Good Papers

Showing RL for LLMs Show all papers

82%Must read
?Must readVote to see the score
arXivRL for LLMs

Sherpa: Teaching LLMs to Teach Adaptively

Sherpa uses multi-turn reinforcement learning to train LLM teachers that adapt instructions to diverse student archetypes, improving student performance by 20.5 points and pedagogy scores to 79.2%.

Weixian Xu, Yanzhe Zhang, Zora Zhiruo Wang, Changyu Chen and 1 more

Published Oct 6, 2026 · ▲ 3 on Hugging Face · Code

100% Readers1 of 1 upvoted
11/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 11 of 20 reviewers recommend it
lenient 5/5
medium 6/10
strict 0/5
84%Must read
?Must readVote to see the score
arXivRL for LLMs

TRACE: Rollout-Guided Quantization-Aware Training for FP4 Reinforcement Learning of MoE Language Models

TRACE aligns FP4 quantization between RL training and rollout paths via rollout-guided quantization-aware training for MoE language models, achieving BF16-comparable RL performance with up to 5.4x rollout speedup.

Xin Wang, Hao Yu, Zhengyang Zhuge, Bochao Mao and 8 more

Published Oct 6, 2026 · ▲ 15 on Hugging Face

100% Readers1 of 1 upvoted
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 12 of 20 reviewers recommend it
lenient 5/5
medium 6/10
strict 1/5
80%Must read
?Must readVote to see the score
arXivRL for LLMs

NeMo-DCR: Bit-Exact Delta-Compressed Refit for Scalable Agentic RL at Trillion-Parameter Scale

NeMo-DCR bit-exactly refits trillion-parameter policies by streaming delta-compressed weight changes via affine mappings and XOR masks, cutting 1T cross-region refits from 87.5 minutes to 150 seconds.

Songlin Jiang, Zhiyu Li, Terry Kong, Yu Yao and 4 more

Published Oct 6, 2026 · ▲ 6 on Hugging Face · Code ★ 2,048

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 12 of 20 reviewers recommend it
lenient 4/5
medium 6/10
strict 2/5
84%Must read
?Must readVote to see the score
arXivRL for LLMs

HiPLEX: Hierarchical Policy Factorization for Full Duplex Speech Language Models

HiPLEX factorizes full-duplex speech policies into timing and content controllers to jointly optimize interaction dynamics via reinforcement learning. It lowers takeover rates, reduces interruption latency, and improves human-like turn timing versus GRPO.

Kyudan Jung, Hyunsin Park, Yoonhyung Lee, Jinhwan Park and 4 more

Published Oct 6, 2026

100% Readers1 of 1 upvoted
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 12 of 20 reviewers recommend it
lenient 5/5
medium 6/10
strict 1/5
76%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026RL for LLMs

Rationale-Guided Policy Optimization: Learning to Reason with Adaptive Rationale Scaffolding

RGPO adaptively uses ground-truth rationales as temporary scaffolds to generate improved responses for on-policy RL, then transfers only higher-reward model outputs back, reducing reward sparsity and improving text and multimodal reasoning.

Hoang Phan, Minh Pham, Chau Pham, Chinmay Hegde and 2 more

Published Oct 5, 2026 · ▲ 7 on Hugging Face · Code ★ 1

– ReadersNo votes yet
10/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 10 of 20 reviewers recommend it
lenient 4/5
medium 6/10
strict 0/5
78%Highly rated
?Highly ratedVote to see the score
arXivRL for LLMs

LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches

LoGRA reduces LLM reinforcement learning memory by up to 45.7% via low-rank gradient sketches and predicted-KL step control, enabling 27B-parameter training on single nodes.

Shaokun Zhang, Yifan Zhang, Jian Hu, Yueying Li and 4 more

Published Oct 5, 2026 · ▲ 11 on Hugging Face

– ReadersNo votes yet
11/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 11 of 20 reviewers recommend it
lenient 5/5
medium 5/10
strict 1/5
80%Must read
?Must readVote to see the score
arXivRL for LLMs

OPD Before RL: Warm-Starting Rubric-Based RL with On-Policy Distillation

Rubric-privileged on-policy distillation before reinforcement learning improves open-ended task scores and reduces reward hacking versus supervised fine-tuning baselines.

Xinpeng Wang, Wei Shi, Yu-Chia Chen, Maria Zontak and 2 more

Published Oct 2, 2026 · 0 citations · ▲ 3 on Hugging Face

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

88%Must read
?Must readVote to see the score
arXivRL for LLMs

MetaRubric: Learning to Reward for Rubric-Based Reinforcement Learning

MetaRubric fixes vacuous rubric credit via evidence-aware optimization and counterfactual rubric adaptation, improving PubMedQA accuracy by up to 20.40 points over static-judge GRPO.

Yuxuan Fan, Jaehong Yoon

Published Oct 2, 2026 · 0 citations · ▲ 22 on Hugging Face · Code ★ 2

– ReadersNo votes yet
15/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

91%Must read
?Must readVote to see the score
arXivU Wisconsin–MadisonNew YorkStanfordRL for LLMs

Sharpening Tax in Post-Training

Post-training sharpens base model behaviors at the cost of solution coverage, introducing a quantifiable "Sharpening Tax"; a posterior-tempered group sampler reduces this tax while boosting accuracy.

Changdae Oh, Qi Zeng, Qi Qi, Andrey Zhmoginov and 6 more

Published Oct 1, 2026 · 0 citations · ▲ 102 on Hugging Face · Code ★ 25

– ReadersNo votes yet
17/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

71%Highly rated
?Highly ratedVote to see the score
arXivRL for LLMs

From Gradients to Capabilities: Understanding Multi-Teacher On-Policy Distillation

Multi-teacher on-policy distillation integrates RL teacher gradients via loss averaging, Adam smoothing, and BF16 rounding, with averaging rules significantly altering math accuracy outcomes.

Siqi Zhu, Suozhi Huang, Kaixuan Zhang, Yuheng Yang and 3 more

Published Oct 1, 2026 · 0 citations · ▲ 13 on Hugging Face

– ReadersNo votes yet
7/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

86%Must read
?Must readVote to see the score
arXivRL for LLMs

Does Scaling Reinforcement Learning Really Require More Training?

SURGE extracts stronger policies from fixed RL histories via spectral fusion of checkpoints, exceeding native training-curve accuracy without extra training or inference cost.

Bangji Yang, Jiajun Fan, MA Hongba, Ruihan Guo and 1 more

Published Oct 1, 2026 · 0 citations

– ReadersNo votes yet
14/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

83%Must read
?Must readVote to see the score
arXivRL for LLMs

My FAULT: Self-Diagnosis as Credit Assignment in Self-Evolving Agentic Reinforcement Learning

FAULT turns self-diagnosed errors into step-level credit via terminal outcome anchoring and evidence-checked cost learning, recovering 95% signal coverage on ALFWorld and improving long-horizon agentic RL.

Yihua Zhu, Qianying Liu, Weixu Qiao, Xuan Ren and 9 more

Published Oct 1, 2026 · 0 citations

– ReadersNo votes yet
13/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

80%Must read
?Must readVote to see the score
arXivRL for LLMs

Function-Structured Reinforcement Learning with Executable Verifiers for Mathematical Reasoning

FSG-RL connects subproblem graphs with Python code and multi-verifier feedback to improve math reasoning, raising final-answer accuracy from 43.25% to 67.50% over supervised fine-tuning.

Zihan Liu, Xurong Xie

Published Oct 1, 2026 · 0 citations

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

89%Must read
?Must readVote to see the score
arXivRL for LLMs

CARM: Cancellation-Aware Response Masking for LLM Reinforcement Learning

CARM prevents opposing token-level probability changes from canceling in sequence-level masking by using absolute log-ratios, improving RL reasoning and code benchmarks over geometric-mean masking.

Yafei Zhang, Songshuo Lu, Sicong Liao, Zhi Chen and 1 more

Published Oct 1, 2026 · 0 citations

– ReadersNo votes yet
16/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

80%Must read
?Must readVote to see the score
arXivRL for LLMs

Make Sparse Rewards Count: Density-Aware Reward Aggregation for Multi-Reward RL

DARA corrects batch-level reward imbalance via inverse-square-root active-group density weights, accelerating multi-reward RL training by up to 65% with no objective change.

Tong Zheng, Skylar Zhai, Zhan Cheng, Tianming Sha and 6 more

Published Sep 30, 2026 · 0 citations · ▲ 64 on Hugging Face · Code ★ 2

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

70%Highly rated
?Highly ratedVote to see the score
arXivU North Carolina at Chapel HillU Maryland, College ParkWashington University in St. LouRL for LLMs

Small RL Controller, Large Language Model: RL-Guided Adaptive Sampling for Test-Time Scaling

A lightweight RL controller formulates adaptive sampling as an MDP to balance LLM answer correctness, latency, and computation cost at test time.

Runpeng Dai, Tong Zheng, Rui Liu, Chengsong Huang and 1 more

Published Jun 2, 2026 · 0 citations · ▲ 13 on Hugging Face

– ReadersNo votes yet
5/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

80%Must read
?Must readVote to see the score
ACL 2026TencentU Notre DameRL for LLMs

Too Correct to Learn: Reinforcement Learning on Saturated Reasoning Data

Reinforcement learning on saturated reasoning data causes mode collapse as advantage signals vanish; CUTS sampling and Mixed-CUTS restore diversity, boosting AIME25 accuracy by up to 15.1%.

Zhenwen Liang, Yujun Zhou, Sidi Lu, Xiangliang Zhang and 2 more

Published Apr 20, 2026 · 0 citations

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

76%Highly rated
?Highly ratedVote to see the score
arXivFudanShanghai Innovation InstituteCentral SouthTsinghuaRL for LLMs

AI Can Learn Scientific Taste

RLCF trains AI to judge and propose high-impact research ideas via community feedback, showing learned scientific taste generalizes across fields and time.

Jingqi Tong, Mingzhe Li, Hangcheng Li, Yongzhuo Yang and 19 more

Published Mar 15, 2026 · 0 citations · ▲ 316 on Hugging Face · Code ★ 433

– ReadersNo votes yet
10/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

88%Must read
?Must readVote to see the score
arXivRL for LLMs

GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization

GDPO decouples per-reward normalization in multi-reward RL to prevent advantage collapse, improving training stability and outperforming GRPO on reasoning and coding tasks.

Shih-Yang Liu, Xin Dong, Ximing Lu, Shizhe Diao and 9 more

Published Jan 8, 2026 · 0 citations · ▲ 235 on Hugging Face · Code ★ 512

– ReadersNo votes yet
15/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

83%Must read
?Must readVote to see the score
ACL 2026TencentU Notre DameRL for LLMs

Too Correct to Learn: Reinforcement Learning on Saturated Reasoning Data

Standard RL collapses on saturated reasoning data due to vanishing advantage signals, so CUTS sampling and Mixed-CUTS training restore exploration and boost AIME25 Pass@1 by 15.1%.

Zhenwen Liang, Yujun Zhou, Sidi Lu, Xiangliang Zhang and 2 more

Published 2026 · 0 citations

– ReadersNo votes yet
13/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

74%Highly rated
?Highly ratedVote to see the score
arXivRL for LLMs

Native Parallel Reasoner: Reasoning in Parallelism via Self-Distilled Reinforcement Learning

NPR enables LLMs to self-evolve genuine parallel reasoning via self-distilled reinforcement learning, achieving up to 24.5% accuracy gains, 4.6x speedups, and 100% parallel execution.

Wu, Tong, Liu, Yang, Bai, Jun, Jia, Zixia and 5 more

Published Dec 8, 2025 · 0 citations · ▲ 80 on Hugging Face · Code ★ 112

– ReadersNo votes yet
9/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

83%Must read
?Must readVote to see the score
arXivRL for LLMs

Parallel-R1: Towards Parallel Thinking via Reinforcement Learning

Parallel-R1 uses a progressive curriculum combining supervised warmup and reinforcement learning to train large language models in parallel reasoning, achieving significant gains on math benchmarks by treating parallel thinking as a temporary exploration scaffold that unlocks higher final performanc

Zheng, Tong, Hongming Zhang, Wenhao Yu, Xiaoyang Wang and 6 more

Published Sep 9, 2025 · 0 citations · ▲ 105 on Hugging Face · Code ★ 265

– ReadersNo votes yet
13/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

80%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026Institute of Information EngineeInstitute of Information EnginerJD.comInstitute of Information EngineeJoy Future Academy at JD.ComRL for LLMs

Co-Evolving Policy Distillation

Co-Evolving Policy Distillation co-trains experts via bidirectional online policy distillation during RLVR to avoid divergence and absorption gaps, integrating multi-modal reasoning to surpass domain-specific experts.

Naibin Gu, Chenxu Yang, Qingyi Si, Chuanyu Qin and 6 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026 · ▲ 67 on Hugging Face

100% Readers1 of 1 upvoted
10/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

91%Must read
?Must readVote to see the score
NeurIPS 2026SpotlightMicrosoftU Wisconsin-MadisonRL for LLMs

ECHO: Terminal Agents Learn World Models for Free

ECHO trains terminal agents to predict environment responses for dense supervision, doubling GRPO pass@1 on TerminalBench-2.0.

Vaishnavi Shrivastava, Ahmed Awadallah, Dimitris Papailiopoulos

Atlanta Poster Session 4, Thu, Dec 10, 4:30 PM–7:30 PM, Hall C1 · Published 2026 · ▲ 7 on Hugging Face

– ReadersNo votes yet. 1 from authors or colleagues not counted
18/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 18 of 20 reviewers recommend it
lenient 5/5
medium 10/10
strict 3/5
45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026U MichiganU Michigan - Ann ArborU Michigan Google ResearchRL for LLMs

VSPO: Vector-Steered Policy Optimization for Controllable Model Behavior

Xuechen Zhang, Zijian Huang, Kai Yang, Weijia Zhang and 2 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 0 of 20 reviewers recommend it
lenient 0/5
medium 0/10
strict 0/5
45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026Shanghai Jiao TongFudanU Michigan - Ann ArborMMLab@CUHKShanghai JiaotongRL for LLMs

Self-Calibrated GUI Reward Model via Inverse Dynamic Modeling

Zeyi Sun, Shengyuan Ding, Xingpeng Xia, Jinsong Li and 3 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026FudanNVIDIAUC Berkeley, University of CalifRL for LLMs

Large Discrete Policy: Advancing Explicit Behavior Modeling with Stochastic Iterative Scoring

Zhenxin Li, Nadine Chang, Xinglong Sun, Jingde Chen and 7 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 0 of 20 reviewers recommend it
lenient 0/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026Carnegie MellonCMU, Carnegie MellonRL for LLMs

Are we really tilting? The mechanics of reward guidance in flow and diffusion models

Sanjit Dandapanthula, Nicholas Boffi

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026Shanghai Jiao TongRL for LLMs

Never Stop Learning: Test-Time Reinforcement Learning for Vision-Language-Action Models

Xinying Yi, Jingjing Jiang, Chao Ma

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026ETHZ - ETH ZurichPoolside AIUniversity College London, UniveRL for LLMs

WATERFALL: Workflow for Adaptive Training with Evolutionary Reward Formulation and Automated Learning Loops

Eleftherios Triantafyllidis, Filippos Christianos, Zhibin Li, Bernd Bickel

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026SpotlightNanyang TechnologicalSkywork AIRL for LLMs

Negative-Only Policy Optimization for One-Sided Verifiable Rewards

Jiacheng Xu, Shuo He, Fuxiang Zhang, Chaojie Wang and 1 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026UCLAUCLA Computer Science DepartmentU California, Los AngelesRL for LLMs

RAHF: Reward-Amplified Human Feedback for Closed-Loop Policy Fine-Tuning

Haoyuan Cai, Seth Zhao, Jason Zhang, Bolei Zhou

Atlanta Poster Session 4, Thu, Dec 10, 4:30 PM–7:30 PM, Hall C1 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026PurdueStanfordMichigan StatePurdue University RL for LLMs

$f$-GRPO & Beyond: Divergence-Based Reinforcement Learning Algorithms for General LLM Alignment

Rajdeep Haldar, Lantao Mei, Guang Lin, Yue XING and 1 more

Atlanta Poster Session 1, Wed, Dec 9, 10:00 AM–1:00 PM, Hall C1 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026Dalian University of TechnologyRL for LLMs

GraphMemRL: Action-Native Reinforcement Learning for Persistent Graph Memory Construction

Bingcheng Dong, Shenglan Liu, Sifan Zhang, Jirui Tian and 1 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026The University of SydneyU Southern QueenslandMacquarieU OsloMacquarie University, AustraliaRL for LLMs

PRISM:Disentangling Preference Distributions for Generative Ranking

zhangkai wu, Kaize Shi, Xu Zhang, Zhihong Cui and 1 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 0 of 20 reviewers recommend it
lenient 0/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026The Hong Kong PolytechnicHong Kong University of Science Southern University of Science aNanyang TechnologicalRutgersRL for LLMs

Train at the Moving Edge: Rollout-Efficient RL for Large Reasoning Models

Jiahao Wu, Ning Lu, Shengcai Liu, Kun Wang and 5 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026U Science and Technology of ChinHKBU / RIKENTCL Coporate ResearchThe Hong Kong University of ScieTCL Corporate ResearchRL for LLMs

Reasoning Warm-up: Scaling Label-free RL via Verifiable Surrogate Rewards

Jun Nie, Bo Han, Jiaqi Fan, Yonggang Zhang and 3 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026SoochowResearch, MicrosoftMicrosoft Research AsiaStony BrookPekingRL for LLMs

Agent$^2$ RL-Bench: Can LLM Agents Engineer Agentic RL Post-Training?

Wanyi Chen, Xiao Yang, Xu Yang, Tianming Sha and 6 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026TsinghuaBeijing Institute of TechnologyRL for LLMs

Principled Policy Optimization for LLMs via Self-Normalized Importance Sampling

Huiyang Shao, Xintong Zhang, Junyi Liu

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026Alibaba GroupNorthwestern PolytechnicalSoutheastZhejiangLazadaRL for LLMs

Stable and Granular Policy Optimization for Generative Recommendation

Muyu Zou, Haibo Xing, Hao Deng, Zhezheng Hao and 5 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026Massachusetts Institute of TechnMITRL for LLMs

Reinforcement Learning with Verifiable Physics: Post-training LLMs for PDE Solver Generation

Pengfei Cai, Utkarsh Utkarsh, Alan Edelman, Christopher Rackauckas and 1 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

67%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026SpotlightkaistKAISTKorea Advanced Institute of ScieRL for LLMs

Imitation Dominates Reinforcement: Direct In-Context RL Is Closer to ICL Than RL

Minchan Kwon, Seunghee Koh, Sunghyun Baek, Minsung Bae and 1 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
2/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 2 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 1/5
45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026U MacauShenzhen Institutes of Advanced RL for LLMs

SPRM: From Cooperative Games to Marginal-Contribution Process Reward Modeling

Yu Bao, Pak Lon Ip, Qiyu Ruan, Xitong Gao and 1 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026U Science and Technology of ChinHuawei Technologies Ltd.TianjinHuawei Noah's Ark LabRL for LLMs

Efficient Gradient-Aware Asynchronous Reinforcement Learning for LLM Post-Training

Songhan Yang, Jie Wang, Yinqi Bai, Tong Xialiang and 3 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026RL for LLMs

Correctable Fork Tokens: Verifier-Anchored Selective Credit Assignment for Tool-Integrated RLVR

Shuqi Yin

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 0 of 20 reviewers recommend it
lenient 0/5
medium 0/10
strict 0/5
67%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026MicrosoftU WashingtonStanfordResearch, MicrosoftPUC-RIORL for LLMs

Multi-Turn RL Makes Small Language Model Competitive for Optimization Modeling

Xinzhi Zhang, Zeyi Chen, Humishka Zope, Hugo Barbalho and 5 more

Atlanta Poster Session 2, Wed, Dec 9, 4:30 PM–7:30 PM, Hall C1 · Published 2026

– ReadersNo votes yet
2/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026MicrosoftMetaRL for LLMs

Relaxed On-Policy Distillation: Selective Credit Allocation for Scaling Reasoning Efficiently

Jongwoo Ko, Sara Abdali, Young Jin Kim, Tianyi Chen and 1 more

Atlanta Poster Session 3, Thu, Dec 10, 10:00 AM–1:00 PM, Hall C1 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026U North Carolina at Chapel HillThe Chinese University of Hong KThe Chinese University of Hong KU Illinois at Urbana-ChampaignInstitute of automation, ChineseRL for LLMs

Auto-Rubric as Reward: From Implicit Preference to Explicit Generative Criteria

Juanxi Tian, Fengyuan Liu, Jiaming Han, Yilei Jiang and 5 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026U MichiganGuide LabsPrescient Design / GenentechNYUPrescient Design, GenentechRL for LLMs

Are Easier or Harder Examples Better? Rethinking Data Selection for Reward Models and Preference Optimization

Kevin Christian Wibisono, Aya Ismail, Pedro O. Pinheiro, Yixin Wang and 3 more

Paris Poster Session 5, Fri, Dec 11, 11:30 AM–1:30 PM, Paris Poster Hall · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026Central China NormalMeituanzhipuSouth-Central MinzuRL for LLMs

SDAE: Semantic-Diversity-Aware Exploration for Efficient Reinforcement Learning in Large Language Models

Jianwen Sun, Wangzi Shi, Sannyuya Liu, Zhiming Wang and 3 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026PekingCity University of Hong KongHuawei Technologies Ltd.Huawei Noah's Ark LabRL for LLMs

LADDERS: Length-Aware Data Distribution and Existing-Response Speculation for Fast RL Rollout Generation

Shengpeng Yin, Hui-Ling Zhen, Xing Li, Mingxuan Yuan and 2 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026SpotlightStanfordU WashingtonRL for LLMs

Rethinking Visual Reasoning in Text-to-Image Reward Modeling

Shiye Su, Xiaohan Wang, Ludwig Schmidt, Serena Yeung-Levy

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 0 of 20 reviewers recommend it
lenient 0/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026MITAmazon AGINortheasternAmazonAmazon AGI -- Web Knowledge ServRL for LLMs

Reward Shaping to Improve Language Model Query Generation

Shicheng Liu, Zeyu Zhang, LEI LU, Kexuan Sun and 1 more

Atlanta Poster Session 3, Thu, Dec 10, 10:00 AM–1:00 PM, Hall C1 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026A*STARWashington StateRL for LLMs

When Scores Conflict with Preferences: Calibrated Drift Control for Heterogeneous DPO

Ping Liu, Yan Yan

Atlanta Poster Session 1, Wed, Dec 9, 10:00 AM–1:00 PM, Hall C1 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026City University of Hong KongInstitute of Advanced IntelligenRL for LLMs

From Ideas to Code: Tree-structured Policy Optimization for Automated Algorithm Design with LLMs

Rui Zhang, Ping Guo, Liyong Lin, Zhichao Lu and 1 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026Qualcomm Inc, QualCommKoreaOracleSeoul National University GooglRL for LLMs

Merging RLVR-Trained Experts via Policy-Shift-Guided Spectral Alignment

Geeho Kim, MINSIK CHOI, Kyle Min, Young Geun Kim and 1 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026U Maryland, Baltimore CountyArizona StateVrije Universiteit AmsterdamRL for LLMs

Cornerstones or Stumbling Blocks? Deciphering the Rock Tokens in On-Policy Distillation

Yuxuan Jiang, Runchao Li, Shubhashis Roy Dipta, Dawei Li and 1 more

Atlanta Poster Session 6, Fri, Dec 11, 4:30 PM–7:30 PM, Hall C1 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

67%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026City University of Hong KongU Illinois ChicagoDalian University of TechnologyMohamed bin Zayed University of Mila - Quebec AI Institute &RL for LLMs

How Data Scales in Agentic Reinforcement Learning: Laws and Synthesis Strategies

Bowei He, Yankai Chen, Xiaokun Zhang, Changjiang Han and 3 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
2/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 2 of 20 reviewers recommend it
lenient 1/5
medium 1/10
strict 0/5
45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026Nanyang TechnologicalTsinghuaNanyang Technological UniversityRL for LLMs

HyperSkill: Training-Free Omnimodal GRPO via Hypergraph-Indexed Skill-Library Evolution

Haoran Luo, Shangkai Lin, Jinyang Wu, XINLIANG ZHOU and 1 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026SpotlightSeoul NationalMassachusetts General Hospital, RL for LLMs

Predicting Human-Gain Curves under Local Proxy Optimization from Repeated Ratings

Moonwon Choi, Kisung Nam, Seunggeun Lee

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 0 of 20 reviewers recommend it
lenient 0/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026MetaColumbiaMathSI, MetaRL for LLMs

Aligning LLMs Toward Multi-Turn Conversational Outcomes Using Iterative RLHF

Daniel Jiang, Ankur Samanta, Yukai Yang, Jalaj Bhandari and 2 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026Chongqing University of TechnoloSun Yat-sen University, ChinaRL for LLMs

CTRL: Continual Test-Time Reinforcement Learning for Large Language Models

Chu Zhao, Enneng Yang

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026Information Technology UniversitQatar Computing Research InstituHuazhong AgriculturalRL for LLMs

SPRING: Solver-guided Process Rewards for Novel Logical Reasoning Steps Generation

Muhammad Asif Ali, Mohammad Raza, Wenqing Wang, Huan Wang

Paris Poster Session 2, Wed, Dec 9, 5:00 PM–7:00 PM, Paris Poster Hall · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 0 of 20 reviewers recommend it
lenient 0/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026Holistic AIUniversity College London & RL for LLMs

Control Reinforcement Learning: Token-Level Mechanistic Analysis via Learned SAE Feature Steering

Seonglae Cho, Zekun Wu, Adriano Koshiyama

Paris Poster Session 4, Thu, Dec 10, 5:30 PM–7:30 PM, Paris Poster Hall · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026U CambridgeMicrosoft ResearchAlibaba GroupRL for LLMs

Reward Budgeting Reduces Premature Convergence in Reinforcement Learning for LLM Reasoning

Mengni Jia, Mengyu Zhou, xiaoxi jiang, Guanjun Jiang

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026CUHKHKUTsinghua universityThe University of Hong KongDepartment of Computer Science aRL for LLMs

Fast-RL: Accelerating Reinforcement Learning for LongCoT Reasoning Models

Sitong Wu, Haoru Tan, bin xia, Bei Yu and 2 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026CUHKHKUThe University of Hong KongDepartment of Computer Science aRL for LLMs

Historical Relative Policy Optimization for Bootstrapping LLM Reasoning

Sitong Wu, Haoru Tan, Bei Yu, Xiaojuan Qi and 1 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026Harbin Institute of TechnologyHuawei Technologies Ltd.Northwest Polytechnical UniversiHuaweiAcademy of Cyber, China electronRL for LLMs

Unveiling Entropy-Performance Decoupling in Agentic RL for Tool-Integrated Reasoning

Yirong Zeng, Shen You, Yufei Liu, Hao Cong and 9 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026Harbin Institute Of Technology, Huawei Technologies Ltd.Harbin Institute of TechnologyRL for LLMs

ASTOR: Multi-Task Code Reinforcement Learning via Utility-Driven Coordination

Yujia Chen, Yang Ye, Xiao Chu, Yuchi Ma and 1 more

Paris Poster Session 3, Thu, Dec 10, 12:30 PM–2:30 PM, Paris Poster Hall · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 0 of 20 reviewers recommend it
lenient 0/5
medium 0/10
strict 0/5
45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026Google ResearchGoogle Google ResearchRL for LLMs

MaxIM: Maximally Informative Incremental Summarization via Reinforcement Learning

Jihwan Jeong, Guy Tennenholtz, Yinlam Chow, Chih-wei Hsu and 1 more

Atlanta Poster Session 1, Wed, Dec 9, 10:00 AM–1:00 PM, Hall C1 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026U South CarolinaAI Institute, University of SoutDevRev Inc.RL for LLMs

Stop Calling It Reinforcement Learning in Language Models Without Clear Improvement Claims: Decision-Process Cards as a Reporting Standard

Anvi Kohli, Vedant Khandelwal, Rishit Agarwal, Rahul Maity and 1 more

Atlanta Poster Session 1, Wed, Dec 9, 10:00 AM–1:00 PM, Hall C1 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 0 of 20 reviewers recommend it
lenient 0/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026BaiduRL for LLMs

Knowledge-Level Consistency Reinforcement Learning: Dual-Fact Alignment for Long-Form Factuality

Junliang Li, Yucheng Wang, Yan Chen, Yu Ran and 4 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026Hong Kong University of Science Hong Kong University of Science ZhejiangTianjinMontreal Institute for Learning RL for LLMs

Reformulate LLM Reinforcement Learning for Stable Training under Black-box Discrepancy

Jiashun Liu, Runze Liu, Xu Wan, Jing Liang and 2 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026PekingDartmouth CollegeU Science and Technology of ChinU TorontoU Southern CaliforniaRL for LLMs

The Heel of RLVR: Benchmark Glory Should Not Outpace Honest Measurement

Shuo Yang, Chiyu Ma, Kexin Huang, Jinda Lu and 10 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 0 of 20 reviewers recommend it
lenient 0/5
medium 0/10
strict 0/5
45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026George WashingtonThe University of Texas at DallaRL for LLMs

OPPO: Bayesian Value Recursion for Token-Level Credit Assignment in Policy Optimization

Yu Li, Rui Miao, Tian Lan, Zhengling Qi

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026Nanyang Technological universitySoutheastNanjing University of AeronauticU MacauRL for LLMs

DACE RL for Compute Efficient Reinforcement Learning in Small Model Reasoning

Guo Yue, YANG LIU, Tang Qingkang, Donghui Zhang and 2 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026kyung HeeKyung HeeRL for LLMs

When Alignment Fails: Stabilizing Cross-Dynamics RL with Prototype Trust Regions

Dong Uk Kim, Ji Su Yoon, Eui-Nam Huh, Choong Hong

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026ZhejiangantgroupAnt GroupCollege of Computer ScienceRL for LLMs

SIRAS: Sibling-Relative Advantage Shaping for Reinforcement Learning from Verifiable Rewards

Zairun Yang, Jun Xu, Lei Liang, Huajun Chen and 1 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 0 of 20 reviewers recommend it
lenient 0/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026U Wisconsin-MadisonKRAFTONKAISTRL for LLMs

Is the Importance Ratio Necessary for Stable Reinforcement Learning in LLMs?

Shuibai Zhang, Junhyuck Kim, Gyeongman Kim, Jaewoong Cho

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026National University of SingaporeHarbin Institute of TechnologySoutheastRL for LLMs

On-Policy Distillation with Open Property-Equivalence Reward for LLM-Based NL-to-SVA Generation

Qingyun Zou, Yingze Li, Tianen Liu, Bingsheng He and 1 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026Institute of automation, ChineseCASIAInstitute of Automation, ChineseRL for LLMs

Boosting Off-Policy RLVR with Data-Centric Replay

Yuxiao He, Junge Zhang, Ziqi Wang

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026U California, BerkeleyUC San DiegoRL for LLMs

Correct, Route, Calibrate: Efficient Preference Optimization from Noisy, Heterogeneous Human Feedback

Zhongming Xie, Xinwei Ma, Jingshen Wang

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026Shanghai JiaotongShanghai Artificial IntelligenceShanghai Jiao TongRL for LLMs

Context Value Informed In-Context Reinforcement Learning

Wenhao Zhang, Shao Zhang, Xihuai Wang, Yang Li and 1 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026PekingTsinghuaBeijing Institute for General ArAlibaba GroupInstitute of automation, ChineseRL for LLMs

Reasoning via Test-Time Instance-Level Policy Gradient in Latent Space

Hengli Li, Chenxi Li, Tong Wu, Xuekai Zhu and 7 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 0 of 20 reviewers recommend it
lenient 0/5
medium 0/10
strict 0/5
57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026FudanShanghai Innovation InstituteShanghai Artificial IntelligenceShanghai Mosi Intelligent TechnoShanghai JiaotongRL for LLMs

IFDECORATOR: Wrapping Instruction Following Reinforcement Learning with Verifiable Rewards

Xu Guo, Tianyi Liang, Jian Tong, Xiaogui Yang and 5 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026NortheasternCarnegie MellonMicrosoft ResearchNortheastern University, ChinaFudanRL for LLMs

What are Key Factors for Updates in RL for LLM Reasoning?

Peidong Wang, Demi Wang, Xufang Luo, Jiahang Xu and 4 more

Paris Poster Session 6, Fri, Dec 11, 2:30 PM–4:30 PM, Paris Poster Hall · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026FudanU Science and Technology of ChinNanjingAlibaba GroupRL for LLMs

DiffCap-RL: Differential QA Rewards for Dense Image and Video Captioning

Kaixun Jiang, Zhihang Liu, Jiyuan Fu, Yuzheng Wang and 5 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

67%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026Renmin University of ChinaHuawei Technologies Ltd.RL for LLMs

Need-Aware Multi-Objective Reinforcement Learning for Emotionally Intelligent LLM Agents

Xiaohe Bo, Xueyang Feng, Rui Li, Zihang Tian and 2 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
2/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 2 of 20 reviewers recommend it
lenient 2/5
medium 0/10
strict 0/5
45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026ZhejiangByteDance Inc.Institute of Software, Chinese ARL for LLMs

GDMD: Guiding Distribution Matching Distillation with Gradient-Based Reinforcement Learning

Linwei Dong, Ruoyu Guo, Ge Bai, Quan Zheng and 2 more

Paris Poster Session 5, Fri, Dec 11, 11:30 AM–1:30 PM, Paris Poster Hall · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

45%Niche pick
?Niche pickVote to see the score
NeurIPS 2026Shanghai JiaotongTencent Technology (shanghai) CoShanghai Jiao TongTencentTencent YouTu LabRL for LLMs

Grounding Agent Reasoning with Structured Process Supervision for Multi-turn Reinforcement Learning

Renting Rui, Yulei Qin, Weiwen Liu, Yunjia Xi and 4 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
0/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 0 of 20 reviewers recommend it
lenient 0/5
medium 0/10
strict 0/5
88%Must read
?Must readVote to see the score
NeurIPS 2026Salesforce AI ResearchSalesForce.comSalesforce ResearchSalesforceRL for LLMs

Prospective Hindsight: Self-Calibrating Reinforcement Learning via Prediction–Reality Gaps

Prospective Hindsight uses prediction-reality gaps to weight gradients, improving reinforcement learning performance and self-calibration by targeting blind spots without added objectives.

Jiaxin Zhang, XIANGYU PENG, Qinglin Chen, Yu Li and 2 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
15/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

78%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026Texas A&MOracle Cloud InfrastructureTexas A&M university RL for LLMs

DRTriton: Large-Scale Synthetic Data Driven Reinforcement Learning for Triton Kernel Generation

DRTriton trains LLMs via synthetic data and reinforcement learning to generate optimized Triton kernels, achieving 92% speedup on KernelBench Level 2 versus 23% for GPT-5.2.

Siqi Guo, Ming Lin, Tianbao Yang

Atlanta Poster Session 2, Wed, Dec 9, 4:30 PM–7:30 PM, Hall C1 · Published 2026

– ReadersNo votes yet
11/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

80%Must read
?Must readVote to see the score
NeurIPS 2026AppleU Tübingen, AppleApple MLRWayveRL for LLMs

Expanding LLM Agent Boundaries with Strategy-Guided Exploration

Strategy-Guided Exploration improves LLM agent reinforcement learning by planning in language strategies rather than actions, boosting performance across UI, tool, coding, and embodied tasks.

Andrew Szot, Michael Kirchhof, Omar Attia, Alexander Toshev

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

78%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026KAISTKorea Advanced Institute of ScieIBM Research - ChinaFudanRL for LLMs

Rebellious Student: Reversing Teacher Signals for Reasoning Exploration with Self-Distilled RLVR

RLRT reverses teacher signals in self-distillation to reinforce student reasoning tokens on successful rollouts, substantially outperforming exploration baselines across Qwen3 models.

Jeonghye Kim, Jiwon Jeon, Dongsheng Li, Yuqing Yang

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026 · ▲ 15 on Hugging Face

– ReadersNo votes yet
11/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 11 of 20 reviewers recommend it
lenient 3/5
medium 7/10
strict 1/5
88%Must read
?Must readVote to see the score
NeurIPS 2026U California, RiversideRL for LLMs

Closing the Reflection Gap: A Free Calibration Bonus for Agentic RL

RefGRPO closes LLM agents' reflection gap via a free calibration bonus and dynamic schedule, improving calibration and task accuracy. Calibrated reflections enable self-improvement without outcome supervision and effective selective prediction.

Yinglun Zhu

Atlanta Poster Session 6, Fri, Dec 11, 4:30 PM–7:30 PM, Hall C1 · Published 2026

– ReadersNo votes yet
15/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

80%Must read
?Must readVote to see the score
NeurIPS 2026U the Chinese Academy of ScienceInstitute of Software, Chinese ABeijing University of AeronauticInstitute of Software, CASU Chinese Academy of SciencesRL for LLMs

Verifiable Environments Are LEGO Bricks: Recursive Composition for Reasoning Generalization

RACES recursively composes verifiable environments as LEGO bricks to scale RL reasoning training, boosting model performance on unseen benchmarks with far fewer base environments.

Hao Xiang, Qiaoyu Tang, Le Yu, Yaojie Lu and 7 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026 · ▲ 8 on Hugging Face

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

72%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026Harbin Institute of TechnologyTsinghuaHarvardAlibaba GroupHong Kong University of Science RL for LLMs

LoopRPT: Reinforcement Pre-Training for Looped Language Models

LoopRPT applies reinforcement pre-training to looped language models by assigning rewards to latent reasoning steps, improving per-step quality and accuracy-computation trade-offs.

Guo Tang, Shixin Jiang, Heng Chang, Zihan Zhang and 6 more

Paris Poster Session 4, Thu, Dec 10, 5:30 PM–7:30 PM, Paris Poster Hall · Published 2026 · ▲ 16 on Hugging Face

– ReadersNo votes yet
8/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 8 of 20 reviewers recommend it
lenient 4/5
medium 4/10
strict 0/5
80%Must read
?Must readVote to see the score
NeurIPS 2026XiaohongshuBilibili Inc.FudanBilibiliZhejiangRL for LLMs

SALT: When More Rollouts Don’t Help in Group-Based Policy Optimization and How to Make Them Matter

Increasing rollouts fails in group-based RL because normalized gradients cancel; SALT adaptively reweights updates via subspace decomposition to recover effective learning.

Powei Chang, Jinpeng Zhang, Chaoqun Sun, MiniWell Tsao and 5 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 12 of 20 reviewers recommend it
lenient 2/5
medium 9/10
strict 1/5
78%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026Sun Yat-senSUN YAT-SEN UNIVERSITYRL for LLMs

Breaking $\textit{Winner-Takes-All}$: Cooperative Policy Optimization Improves Diverse LLM Reasoning

GCPO replaces individual rollout scoring with team-level credit assignment based on valid solution coverage, significantly improving reasoning accuracy and diversity over competitive RLVR methods.

Haoxuan Chen, Tianming Liang, Wei-Shi Zheng, Jian-Fang Hu

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
11/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 11 of 20 reviewers recommend it
lenient 4/5
medium 7/10
strict 0/5
74%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026TianjinMontreal Institute for Learning ShanxiAlibaba GroupAlibaba Inc.RL for LLMs

The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning

LLM RL suffers from training-inference policy mismatch; MIPU optimizes monotonic inference policy improvement to stabilize training and boost reasoning performance.

Jing Liang, Hongyao Tang, Yi Ma, Yancheng He and 7 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026 · ▲ 51 on Hugging Face

– ReadersNo votes yet
9/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

80%Must read
?Must readVote to see the score
NeurIPS 2026NankaiBeijing Institute of TechnologyHarbin Institute of Technology, Renmin University of ChinaMicrosoft Research AisaRL for LLMs

SCOPE-RL: Stable and Quantitative Control of Policy Entropy in RL Post-Training

SCOPE-RL uses temperature-adaptive positive samples to stabilize entropy and prevent collapse in RL post-training of reasoning LLMs, improving Pass@1 and Pass@$k$ with non-monotonic exploration benefits.

Chen Wang, Zhaochun Li, Bai Jionghao, Hexuan Deng and 3 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 1 on Hugging Face

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

88%Must read
?Must readVote to see the score
NeurIPS 2026Meta / Ecole Normale SupérieureU California, San DiegoINRIA - Ecole Normale SuperieureFacebook AI ResearchMeta / FAIRRL for LLMs

Don’t Let Gains FADE: Breaking Down Policy Gradient Weights in RL

A framework decomposes RL advantage functions into gradient mass axes, showing trade-offs shift during training and motivating FADE, which adapts weights dynamically to accelerate convergence and improve accuracy-diversity trade-offs.

Juliette Decugis, Sean O'Brien, Francis Bach, Gabriel Synnaeve and 1 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
15/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

88%Must read
?Must readVote to see the score
NeurIPS 2026MIT - Massachusetts Institute ofAWS AI LabsUT-AustinAWSAmazonRL for LLMs

MURPHY: Feedback-Aware GRPO with Retrospective Credit Assignment for Multi-Turn Code Generation

MURPHY extends GRPO to multi-turn code generation via feedback-conditioned rollout trees with retrospective credit assignment, achieving up to 6% absolute pass@1 gains over prior methods.

Chanakya Ekbote, Vijay Lingam, Sujay Sanghavi, Luke Huan and 3 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
15/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

83%Must read
?Must readVote to see the score
NeurIPS 2026Nanyang TechnologicalThe University of SydneyAlibaba GroupRL for LLMs

STRIDE: Learnable Stepwise Language Feedback for LLM Reasoning

STRIDE replaces scalar process rewards with learnable stepwise language feedback co-trained via outcome rewards, enabling trajectory redirection and sustained LLM reasoning gains.

Junjie Zhang, Guozheng Ma, Shunyu Liu, Zetian Hu and 4 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
13/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

80%Must read
?Must readVote to see the score
NeurIPS 2026Harbin Institute of TechnologyThe Hong Kong University of ScieDepartment Of Computer Science aTsinghuaKuaishou- 快手科技RL for LLMs

SpatialFlow-GRPO: Where Spatial Credit Drives Image Editing

SpatialFlow-GRPO introduces region-aware rewards to replace whole-image feedback, improving fine-grained image editing via spatially aligned policy updates.

Yankai Yang, Yancheng Long, Wei Chen, Xingyu Lu and 6 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

83%Must read
?Must readVote to see the score
NeurIPS 2026U Chinese Academy of ScienceMeituanCenter for Research on IntelligeInstitute of automation, ChineseU Science and Technology of ChinRL for LLMs

TGRL: Temperature-Grouped Reinforcement Learning for Efficient Exploration in LLMs

TGRL turns temperature-induced rollout diversity into an explicit RLVR training signal via reward contrast and Jensen-Shannon divergence, accelerating convergence up to 36% while improving math, code, and agent benchmarks.

Zihan Lin, Xiaohan Wang, Jie Cao, Jiajun Chai and 3 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026 · ▲ 8 on Hugging Face · Code ★ 3

– ReadersNo votes yet
13/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

83%Must read
?Must readVote to see the score
NeurIPS 2026SUN YAT-SEN UNIVERSITYWechat Search, Tencent Inc.U Michigan - Ann ArborShenzhen Loop Area InstituteTencentRL for LLMs

Stabilizing Policy Optimization via Logits Convexity

Logits convexity explains supervised fine-tuning stability versus RL instability, and the proposed LCO framework improves policy optimization stability and benchmark performance.

Hongzhan Chen, Tao Yang, Yuhua Zhu, Shiping Gao and 2 more

Paris Poster Session 2, Wed, Dec 9, 5:00 PM–7:00 PM, Paris Poster Hall · Published 2026

– ReadersNo votes yet
13/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

78%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026DukeDepartment of Electrical EngineeYale University & Cisco FounRL for LLMs

Learning in Context, Guided by Choice: A Reward-Free Paradigm for Reinforcement Learning with Transformers

In-context preference-based RL trains transformers solely on preference feedback, achieving reward-free in-context generalization comparable to fully supervised methods.

Juncheng Dong, Moyang Guo, Bowen He, Ethan Fang and 2 more

Atlanta Poster Session 2, Wed, Dec 9, 4:30 PM–7:30 PM, Hall C1 · Published 2026

– ReadersNo votes yet
11/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

86%Must read
?Must readVote to see the score
NeurIPS 2026KAISTKoreaGoogleRL for LLMs

Reason in the Words You Speak: Idiolectal Paraphrasing Off-Policy Traces for Reasoning Distillation in VideoLLMs

Echo-GRPO rewrites off-policy reasoning traces into a student VideoLLM's idiolect to avoid gradient clipping, improving reasoning distillation across backbones and benchmarks.

Ji Soo Lee, Jinyoung Park, Seohyun Lee, Jongha Kim and 3 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026 · ▲ 24 on Hugging Face

– ReadersNo votes yet
14/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

83%Must read
?Must readVote to see the score
NeurIPS 2026U Texas at DallasEmoryU Texas at ArlingtonU FloridaUT DallasRL for LLMs

LEAD: Length-Efficient Adaptive and Dynamic Reasoning for Large Language Models

LEAD adaptively calibrates reasoning length via online self-adaptive rewards, achieving highest accuracy and efficiency scores with shorter outputs than base reasoning models.

Songtao Wei, Yi Li, Zhikai Li, Xu Hu and 6 more

Atlanta Poster Session 6, Fri, Dec 11, 4:30 PM–7:30 PM, Hall C1 · Published 2026 · ▲ 6 on Hugging Face · Code ★ 4

– ReadersNo votes yet
13/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

71%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026The University of QueenslandU QueenslandRL for LLMs

Block-R1: Rethinking the Role of Block Size in Multi-domain Reinforcement Learning for Diffusion Large Language Models

Block-R1 reveals domain-level block-size conflicts in diffusion LLM reinforcement learning and proposes sample-level block sizing for cross-domain post-training.

Yan Jiang, Ruihong Qiu, Zi Huang

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
7/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

74%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026Seoul NationalUpstageColumbiaUpstage AIRL for LLMs

Multi-Step Likelihood-Ratio Correction for Reinforcement Learning with Verifiable Rewards

NFPO augments PPO with an N-step forward trace to reduce structural bias and control variance via cumulative token likelihood ratios, improving reasoning performance.

Deokgyu Yoon, Hyungkyu Kang, Joongkyu Lee, Byeongchan Kim and 3 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
9/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

80%Must read
?Must readVote to see the score
NeurIPS 2026The University of Hong KongU Hong KongRL for LLMs

Overcoming Catastrophic Forgetting in Visual Continual Learning with Reinforcement Fine-Tuning

RaPO reduces visual continual learning forgetting via trajectory-level reward shaping that penalizes policy drift from prior tasks.

Meng Lou, Hanzhong Guo, Linwei Chen, Yizhou Yu

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 6 on Hugging Face

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

72%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026TsinghuaDepartment of Automation, TsinghTsinghua University, TsinghuaU Maryland, College ParkIndependent ResearcherRL for LLMs

Listwise Policy Optimization: Group-based RLVR as Target-Projection on the LLM Response Simplex

Listwise Policy Optimization explicitly projects policies onto target distributions over response simplices via divergence minimization, improving reasoning performance and stability over group-based policy gradients.

Yun Qu, Qi Wang, Yixiu Mao, Heming Zou and 10 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 67 on Hugging Face

– ReadersNo votes yet
8/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 8 of 20 reviewers recommend it
lenient 2/5
medium 6/10
strict 0/5
80%Must read
?Must readVote to see the score
NeurIPS 2026U Science and Technology of ChinRL for LLMs

REFLEX: Reflective Evolution from LLM Experience

REFLEX decouples visual diagnosis from code generation in evolutionary policy search via a Critic-Actor architecture with persistent Skill Memory, solving tasks in under 10 LLM calls with strong sample efficiency.

Pan Wang

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

70%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026East China NormalFudanShanghai JiaotongU Science and Technology of ChinAnt GroupRL for LLMs

DiPO: Disentangled Perplexity Policy Optimization for Fine-grained Exploration-Exploitation Trade-Off

DiPO disentangles perplexity into exploration and exploitation subspaces to enable fine-grained trade-offs, improving LLM reasoning and function calling via stable perplexity-guided policy optimization.

Xiaofan Li, Ming Yang, Zhiyuan Ma, Shichao Ma and 8 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026 · ▲ 13 on Hugging Face

– ReadersNo votes yet
4/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

91%Must read
?Must readVote to see the score
NeurIPS 2026Johns HopkinsAnthropicFacebookJHUMeta FAIRRL for LLMs

Jointly Reinforcing Diversity and Quality in Language Model Generations

DARLING uses a learned partition function to jointly optimize language model response quality and semantic diversity via reinforcement learning, improving both quality and novelty across creative and math benchmarks.

Tianjian Li, Yiming Zhang, Ping Yu, Swarnadeep Saha and 4 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 25 on Hugging Face · Code ★ 61

– ReadersNo votes yet
18/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 18 of 20 reviewers recommend it
lenient 5/5
medium 10/10
strict 3/5
76%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026U Science and Technology of ChinWuhanJD Explore AcademyRL for LLMs

Not Only Where, But When: Temporal Scheduling for RLVR

Scheduling credit allocation criteria over RLVR training improves stability and efficiency by prioritizing targeted tokens before gradually attenuating to general optimization.

Jinghao Zhang, Ruilin Li, Feng Zhao, Jiaqi Wang

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 3 on Hugging Face · Code ★ 1

– ReadersNo votes yet
10/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 10 of 20 reviewers recommend it
lenient 3/5
medium 7/10
strict 0/5
91%Must read
?Must readVote to see the score
NeurIPS 2026SpotlightMicrosoft AIMicrosoftRL for LLMs

Knowing When to Ask: Segment-Level Credit Assignment for LLM Tool Use

CARL assigns segment-level reinforcement learning credit at tool-use boundaries to teach models when external tools are needed, improving accuracy by up to 9.7 points while cutting unnecessary calls by 53%.

Abhijit Kumar, Zoey WU, Mohit Suley

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
17/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

78%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026U Science and Technology of ChinKuaishou- 快手科技KuaishouApplied Research Center, TencentKuaishou TechnologyRL for LLMs

Think, then Score: Decoupled Reasoning and Scoring for Video Reward Modeling

DeScore decouples chain-of-thought reasoning from scoring in video reward models to improve generalization and training stability. Its think-then-score design uses explicit reasoning followed by a dedicated regression head, optimized via cold-start and dual-objective reinforcement learning.

Yuan Wang, Ouxiang Li, Yulong Xu, Borui Liao and 7 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 2 on Hugging Face

– ReadersNo votes yet
11/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

86%Must read
?Must readVote to see the score
NeurIPS 2026NubankGoogle & Google DeepMindRL for LLMs

Bayesian Preference Learning for Test-Time Steerable Reward Models

ICRM enables test-time steerable reward models via Bayesian variational inference over preferences, improving multi-objective alignment, calibration, and math reasoning.

Jiwoo Hong, Shao Tang, Zhipeng Wang

Atlanta Poster Session 6, Fri, Dec 11, 4:30 PM–7:30 PM, Hall C1 · Published 2026

– ReadersNo votes yet
14/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 14 of 20 reviewers recommend it
lenient 3/5
medium 9/10
strict 2/5
83%Must read
?Must readVote to see the score
NeurIPS 2026SalesForce.comSalesforceSalesforce ResearchSalesforce AI ResearchRL for LLMs

Procedural Memory Distillation: Online Reflection for Self-Improving Language Models

Procedural Memory Distillation extracts cross-episode strategy patterns into reusable procedural memory, co-evolving with the policy to improve reasoning benchmarks by up to 13.6% over SDPO.

Ye Liu, Srijan Bansal, Bo Pang, Yang Li and 5 more

Atlanta Poster Session 3, Thu, Dec 10, 10:00 AM–1:00 PM, Hall C1 · Published 2026

– ReadersNo votes yet
13/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

78%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026Salesforce AI ResearchSalesForce.comSalesforce ResearchRL for LLMs

Learning from Language Feedback via Variational Policy Distillation

Variational Policy Distillation co-evolves an adaptive teacher and student via variational EM to extract dense token-level guidance from language feedback, outperforming RLVR and self-distillation baselines on reasoning and code tasks.

Yang Li, Erik Nijkamp, Semih Yavuz, Shafiq Joty

Atlanta Poster Session 6, Fri, Dec 11, 4:30 PM–7:30 PM, Hall C1 · Published 2026 · ▲ 12 on Hugging Face

– ReadersNo votes yet
11/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

80%Must read
?Must readVote to see the score
NeurIPS 2026HarvardCornellMicrosoft ResearchKempner and SEAS at HarvardHarvard University & AmazonRL for LLMs

Scaling Reward Modeling without Human Supervision

Unsupervised reward modeling via web document prefix-suffix preference learning improves RewardBench accuracy up to 7.7 points and matches supervised baselines without human annotations.

Jingxuan Fan, Yueying Li, Zhenting Qi, Dinghuai Zhang and 3 more

Atlanta Poster Session 1, Wed, Dec 9, 10:00 AM–1:00 PM, Hall C1 · Published 2026

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

89%Must read
?Must readVote to see the score
NeurIPS 2026Siemens DAIAWS AIGoogleU Minnesota, MinneapolisRL for LLMs

DualKV: Shared-Prompt Flash Attention for Efficient RL Training with Large Rollouts and Long Contexts

DualKV eliminates shared-prompt replication in RL training via FlashAttention kernels that process shared and per-sequence KV regions separately, achieving up to 3.82x policy-update speedup.

Jiading Gai, Shuai Zhang, Xiang song, Yuyang (Bernie) Wang and 1 more

Atlanta Poster Session 1, Wed, Dec 9, 10:00 AM–1:00 PM, Hall C1 · Published 2026 · ▲ 1 on Hugging Face

– ReadersNo votes yet
16/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

74%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026nanjing universityNanjingDalian University of TechnologyCity University of Hong KongRL for LLMs

ExpLang: Improved Exploration and Exploitation in LLM Reasoning with On-Policy Thinking Language Selection

ExpLang improves LLM reasoning via on-policy multilingual thinking language selection during RL, outperforming English-only training and extending exploration with diverse language preferences.

Changjiang Gao, Zixian Huang, Kaichen Yang, Jiajun Chen and 2 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
9/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 9 of 20 reviewers recommend it
lenient 5/5
medium 4/10
strict 0/5
78%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026IndependentGradient NetworkU Science and Technology of ChinTsinghua University, TsinghuaU Edinburgh, University of EdinbRL for LLMs

ECHO-2: A Large-Scale Distributed Rollout Framework for Cost-Efficient Reinforcement Learning

ECHO-2 is a distributed RL framework that overlaps rollout generation, dissemination, and training with bounded policy staleness to improve cost efficiency while preserving rewards.

Jingwei Song, Meng Chen, Jie Xiao, Qingnan Ren and 14 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026 · ▲ 13 on Hugging Face

– ReadersNo votes yet
11/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

74%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026U CambridgeThomson Reuters Foundational ML RL for LLMs

Step-by-Step Optimization-like Reasoning in LLMs over Expanding Search Spaces

OPT* introduces optimization-style tasks with expanding search spaces and feasibility checkers to evaluate LLM step-by-step reasoning, showing that training on it improves optimization-like reasoning via online policy optimization and search-based offline RL.

Nicolás Astorga, Nabeel Seedat, Mihaela van der Schaar

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
9/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 9 of 20 reviewers recommend it
lenient 4/5
medium 5/10
strict 0/5
80%Must read
?Must readVote to see the score
NeurIPS 2026StanfordStanford University / humans&ampRL for LLMs

Neural Garbage Collection: Learning to Forget while Learning to Reason

Neural Garbage Collection trains language models via reinforcement learning to evict KV cache entries during reasoning, achieving 2-3x compression with minimal accuracy loss.

Michael Li, Jubayer Ibn Hamid, Emily Fox, Noah Goodman

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 2 on Hugging Face

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

88%Must read
?Must readVote to see the score
NeurIPS 2026PurdueTexas Instruments Inc.RL for LLMs

SARL: Label-Free Reinforcement Learning by Rewarding Reasoning Topology

SARL improves reasoning via label-free reinforcement learning that rewards reasoning topology over outcomes, outperforming supervised and preference-based methods on math and open-ended tasks with more stable training.

Yifan Wang, Bolian Li, David Cho, Ruqi Zhang and 2 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
15/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

80%Must read
?Must readVote to see the score
NeurIPS 2026TongjiThe Chinese University of Hong KInstitute of automation, ChineseTianjinInner MongoliaRL for LLMs

AffectGPT-RL: Revealing Roles of Reinforcement Learning in Open-Vocabulary Emotion Recognition

AffectGPT-RL applies reinforcement learning to open-vocabulary emotion recognition, improving performance by optimizing non-differentiable metrics and enabling reasoning.

Zheng Lian, Fan Zhang, Lan Chen, Yazhou Zhang and 7 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

78%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026Stanford University & LMU MuU Munich, Ludwig-Maximilians-UniStanfordU BolognaLMU MunichRL for LLMs

Tool Verification for Test-Time Reinforcement Learning

T³RL uses external tool verification to correct false-majority rewards in test-time RL, improving math reasoning across benchmarks.

Ruotong Liao, Nikolai Röhrich, Xiaohan Wang, Yuhui Zhang and 3 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026 · ▲ 7 on Hugging Face · Code ★ 2

– ReadersNo votes yet
11/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 11 of 20 reviewers recommend it
lenient 4/5
medium 7/10
strict 0/5
80%Must read
?Must readVote to see the score
NeurIPS 2026PekingMicrosoftMicrosoft Research AsiaSouth China University of TechnoRL for LLMs

Group-Graph Policy Optimization for Long-Horizon Agentic Reinforcement Learning

G2PO transforms agent trajectories into state-transition graphs to reduce variance and improve credit assignment, outperforming GRPO by up to 22.2% on long-horizon benchmarks.

Yunan Wang, Minghui Song, Zihan Zhang, Shaohan Huang and 5 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
12/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 12 of 20 reviewers recommend it
lenient 5/5
medium 7/10
strict 0/5
86%Must read
?Must readVote to see the score
NeurIPS 2026TsinghuaMicrosoftPengcheng LaboratoryMicrosft Research AsiaU OxfordRL for LLMs

STARE: Surprisal-Guided Token-Level Advantage Reweighting for Policy Entropy Stability

STARE analyzes token-level entropy dynamics under GRPO, identifies a credit assignment mismatch, and uses surprisal-guided advantage reweighting to stabilize policy entropy, improving AIME accuracy by 4-8%.

HAIPENG LUO, Qingfeng Sun, Song-Li Wu, Can Xu and 3 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 12 on Hugging Face · Code ★ 24

– ReadersNo votes yet
14/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

76%Highly rated
NeurIPS 2026SpotlightInstitute of Information EnginerInstitute of Information EngineeJD.comU the Chinese Academy of ScienceInstitute of Information EngineeRL for LLMs

Self-Distilled RLVR

RLSD combines RLVR and self-distillation, using token-level policy differences for update magnitudes and environmental feedback for directions, improving convergence and stability.

Chenxu Yang, Chuanyu Qin, Qingyi Si, Minghui Chen and 6 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 145 on Hugging Face

– ReadersNo votes yet
10/20 AI panelreviewers recommend it

Only vote on papers you've read. Sign in with GitHub to vote.

Show 20 more papers