RRSI regularizes recursive agent harness self-improvement via annealed edit budgets, trajectory exploration, and critical selection to boost out-of-distribution performance and reduce token use. It improves up to 14.1 points in-distribution and 4.7 points out-of-distribution while cutting policy tok

Peng Xia, Rujun Han, Zifeng Wang, Yanfei Chen and 10 more

Published Sep 21, 2026 · 0 citations · ▲ 222 on Hugging Face · Code ★ 1,293

– 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.

86%Must read

?Must readVote to see the score

arXivMultimodal reasoning

Rethinking Latent Visual Reasoning: Grounding Latent Reasoning in Visual Evidence

ReaLVR fixes latent reasoning's weak visual grounding by supervising latent tokens with visual evidence, improving reasoning across scales up to 235B.

Xi Xiao, Tianchen Zhao, Youngeun Kim, Zhuowei Li and 9 more

Published Sep 28, 2026 · 0 citations · ▲ 217 on Hugging Face · Code ★ 40

– 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.

78%Highly rated

?Highly ratedVote to see the score

arXivUnified multimodal models

LTX-2: Efficient Joint Audio-Visual Foundation Model

LTX-2 is an open-source 14B/5B audiovisual transformer generating synchronized high-quality video and audio with state-of-the-art open-source quality at low computational cost.

Yoav HaCohen, Benny Brazowski, Nisan Chiprut, Yaki Bitterman and 25 more

Published Jan 6, 2026 · 0 citations · ▲ 197 on Hugging Face · Code ★ 9,606

– 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

72%Highly rated

?Highly ratedVote to see the score

arXivLLM agents & planning

Apodex 1.1: Scaling Agentic Intelligence for Complex Work

Apodex 1.1 scales agentic intelligence via environment and coordination scaling to achieve leading complex-work performance with smaller models.

B. An, B. An, B. Wang, B. L. Wang and 36 more

Published Aug 24, 2026 · 0 citations · ▲ 212 on Hugging Face · Code ★ 5,146

– 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.

80%Must read

?Must readVote to see the score

arXivSoftware engineering agents

Frontis-MA1: Training an AI4AI Model towards Recursive Self-Improvement in Machine Learning Engineering

Frontis-MA1 improves machine-learning engineering via recursive self-improvement using OpenMLE, boosting MLE-Bench Lite medal average from 39.39% to 71.21% and surpassing larger closed models.

Junlin Yang, Che Jiang, Yu Fu, Tianwei Luo and 20 more

Published Jul 30, 2026 · 0 citations · ▲ 188 on Hugging Face · Code ★ 782

– 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 8/10

strict 0/5

88%Must read

?Must readVote to see the score

arXivU Cambridge Alignment & preference optimization

On-Policy or Off-Policy Learning? A Systematic Study of Distillation Dynamics

In controlled strong-to-weak distillation, rollout policy is less central than token-level KL direction and learning rate, though on-policy data can improve generalization on harder reasoning tasks.

Julianna Piskorz, Antonin Berthon, Mihaela van der Schaar

Published Sep 28, 2026 · 0 citations · ▲ 194 on Hugging Face

– 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

arXivWashington University in St. Lou Google U North Carolina at Chapel Hill Agent benchmarks & environments

EnvHarness: Awakening Static Worlds for Agent Learning

EnvHarness wraps static environments with programmable components to reshape agent behavior without altering underlying logic, improving benchmarks by up to 9.0 points while enabling continuous policy-environment co-evolution.

Chengsong Huang, Zifeng Wang, Rujun Han, Jun Yan and 13 more

Published Aug 20, 2026 · 0 citations · ▲ 175 on Hugging Face · Code ★ 619

– 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.

74%Highly rated

?Highly ratedVote to see the score

arXivShanghai Jiao Tong Shanghai Innovation Institute Multi-agent LLM systems

ARIS: Autonomous Research via Adversarial Multi-Agent Collaboration

ARIS is an open-source autonomous research harness using cross-model adversarial collaboration to coordinate ML workflows and verify experimental claims.

Ruofeng Yang, Yongcan Li, Shuai Li

Published May 4, 2026 · 1 citation · ▲ 154 on Hugging Face · Code ★ 17,051

– 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

88%Must read

?Must readVote to see the score

arXivU Science and Technology of Chin Shanghai Innovation Institute Fudan Shanghai Artificial Intelligence LLM agents & planning

GraphForge: Training Working Agents with Graph-Anchored Workspace Synthesis

GraphForge synthesizes workspace tasks and verifiers over real file evidence graphs to train working agents, and fine-tuning Qwen3.6-27B improves GDPVal, Workspace-Bench-Lite, and SpreadsheetBench II results.

Qisheng Su, Hanchen Wang, 朱冠儒, Huicheng Jiang and 8 more

Published Sep 30, 2026 · 0 citations · ▲ 146 on Hugging Face

– 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.

72%Highly rated

?Highly ratedVote to see the score

arXivLLM agents & planning

AREX-2: Advancing Self-Improving Agents through Long-Horizon Reflective Tasks

AREX-2 synthesizes long-horizon reflective trajectories to train a Qwen3.8-27B agent that self-improves at test time, achieving strong results on MLE-bench, Frontier-CS, and deep research benchmarks while scaling with iteration budget.

Hongjin Qian, Chaofan Li, Kun Luo, Wenqing Wei and 10 more

Published Sep 29, 2026 · 0 citations · ▲ 139 on Hugging Face · Code ★ 31

– 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.

71%Highly rated

?Highly ratedVote to see the score

arXivHong Kong University of Science U Hong Kong Tencent Unified multimodal models

Adaptive Reward Routing: Dynamic Multi-Reward Optimization for Joint Audio-Video Diffusion via Forward-Process RL

Adaptive Reward Routing dynamically routes updates and balances rewards during forward-process RL for joint audio-video diffusion, consistently improving quality, alignment, and synchronization over fixed baselines.

Songlin Yang, Xiaotong Zhao, Jiacheng Zhang, Zhe Wang and 4 more

Published Sep 29, 2026 · 0 citations · ▲ 138 on Hugging Face

– ReadersNo votes yet

6/20 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

arXivShanghai Jiao Tong Fudan Shanghai Innovation Institute Northeastern LLM reasoning & chain of thought

Does Learning Protein Folding Generalize to Broader Reasoning?

Post-training on protein-folding data via discrete answers and continuous geometry improves structure prediction and broad reasoning across ten benchmarks.

Yong Liu, Zhanpeng Shi, Yizhou Dang, Zhongyue Zhang and 3 more

Published Sep 30, 2026 · 0 citations · ▲ 120 on Hugging Face · Code ★ 28

100% Readers1 of 1 upvoted

16/20 AI panelreviewers recommend it

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

88%Must read

?Must readVote to see the score

arXivLLM agents & planning

Mid-Harness: Scaling Actions Between Model and Harness for Terminal Agents

Mid-Harness verifies candidate terminal actions at the model-harness boundary, raising TerminalBench-Lite Pass@1 from 50.00% to 68.03% and improving success at lower token cost than trajectory scaling alone.

Minki Kang, Ryo Hachiuma, Shaokun Zhang, Subhashree Radhakrishnan and 7 more

Published Sep 30, 2026 · 0 citations · ▲ 116 on Hugging Face

– 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

arXivDiffusion models

UniMate: One Unified Model to Animate Diverse Skeletons

UniMate is a unified diffusion transformer that synthesizes motion for arbitrary skeletons from text and rigged assets without test-time optimization, using topology-aware attention and a new dataset to outperform specialized animators.

Linzhan Mou, Lei, Jiahui, Zhiyang Dou, Chenyue Cai and 3 more

Published Sep 4, 2026 · 0 citations · ▲ 23 on Hugging Face · Code ★ 1,553

– 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

arXivU Minnesota Korea Advanced Institute of Scie Seoul National Hanyang Tool use & function calling

EvoDuet: Bilevel Co-Evolution of Web Searching and Task Solving for Scientific Discovery

EvoDuet co-evolves solutions and web queries via a retrieval gate to boost LLM discovery gains up to 82.3% across optimization tasks.

Young-Jun Lee, Jinheon Baek, Soyeong Jeong, Minki Kang and 4 more

Published Sep 30, 2026 · 0 citations · ▲ 109 on Hugging Face · Code ★ 4

– 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.

88%Must read

arXivNational University of Singapore Harbin Institute of Technology Video generation

FrameMorrow: Future-guided Frame Selection with Prospective Tokens for Long-Horizon Video Generation

FrameMorrow guides historical frame selection via prospective tokens representing future needs, improving consistency and quality across diverse long-horizon video generators.

Bo Yin, Xiaobin Hu, Jiaqi Zhao, Shuicheng Yan

Published Sep 30, 2026 · 0 citations · ▲ 105 on Hugging Face · Code ★ 29

100% Readers1 of 1 upvoted

14/20 AI panelreviewers recommend it

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

91%Must read

arXivU Wisconsin–Madison New York Stanford RL 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

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

76%Highly rated

?Highly ratedVote to see the score

arXivAgent benchmarks & environments

WorldAuditBench: Interactive 3D World Auditing with Multimodal Agents

WorldAuditBench benchmarks interactive 3D world auditing with multimodal agents, finding success rates of 6.6% to 42.3% versus 83.4% human performance.

Ziyan Jiang, Jingbo Yang, Jiabao Ji, Yujian Liu and 4 more

Published Sep 30, 2026 · 0 citations · ▲ 102 on Hugging Face · Code ★ 4

– 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.

71%Highly rated

?Highly ratedVote to see the score

arXivLLM evaluation & benchmarks

Predictive Credit: Measuring What Scientific Explanations Add to Experimental Forecasts

Predictive credit for scientific explanations is measured via paired forecasts, but gains over descriptions remain unconfirmed across Tox21, OpenML, and controlled settings.

Jingjie Ning, Xueqi Li, Yibo Kong, Dongting Li

Published Sep 29, 2026 · 0 citations · ▲ 101 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.

80%Must read

?Must readVote to see the score

arXivU Southern California California Southern Southern States Seoul National New Generation University Colleg Efficient inference & serving

Prefill-Free Cross-Family KV Cache Transfer for Heterogeneous Multi-Agent LLMs

HeteroFold enables prefill-free cross-family KV cache transfer between frozen heterogeneous LLM agents, accelerating 32K context transfer up to 10.7x while matching text-based multi-agent performance.

Vincent-Daniel Yun, Woosang Lim, Haneul Yoo, Sungjoo Yoo and 2 more

Published Sep 26, 2026 · 0 citations · ▲ 95 on Hugging Face · Code ★ 1

– 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.

86%Must read

?Must readVote to see the score

arXivAgent memory

Beyond Memory: Harnessing Long-Horizon Agents with Explicit Belief States

PoS maintains explicit belief states for long-horizon LLM agents, detects belief trapping, and recovers to achieve top results across four benchmarks.

Yu Luo, Jiamin Jiang, Yimin Zuo, Xidao Wen and 8 more

Published Oct 1, 2026 · 0 citations · ▲ 93 on Hugging Face · Code ★ 28

– 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

arXivLLM agents & planning

Scaling Trajectories for Complex Tasks through Recursive Self-Rewrite

Recursive Self-Rewrite uses diverse harnesses and recursive revision to rewrite successful terminal trajectories for supervised fine-tuning, boosting pass@3 by up to 7.6x on hard benchmarks.

Zongxia Li, Yucheng Shi, Zhongzhi Li, Junyao Yang and 6 more

Published Oct 2, 2026 · 0 citations · ▲ 91 on Hugging Face

– ReadersNo votes yet. 1 from authors or colleagues not counted

13/20 AI panelreviewers recommend it

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

86%Must read

?Must readVote to see the score

arXivSoftware engineering agents

RSIGame: Autonomous Agentic Game Development with Recursive Self-improvement

RSIGame uses recursive self-improvement with local and global loops to autonomously refine generated games, surpassing one-shot GPT-5.5 scores while cutting generation tokens by 11x.

Wenyi Wu, Minghao Fu, Jieyu You, Kun Zhou and 9 more

Published Sep 30, 2026 · 0 citations · ▲ 91 on Hugging Face · Code ★ 125

– 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.

91%Must read

arXivRobot manipulation

MotorMind: Scaffolding General Vision Language Models for Zero-Shot Robot Manipulation

MotorMind connects general vision-language models to deterministic robot control via mid-level actions and feedback loops, achieving 66.7% zero-shot success on LIBERO-PRO and 95% on real robots without task-specific training or external tools.

Bingxuan Li, Siqi Song, Yizhuo Wu, Jiarui Yao and 2 more

Published Sep 29, 2026 · 0 citations · ▲ 90 on Hugging Face

– ReadersNo votes yet

18/20 AI panelreviewers recommend it

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

78%Highly rated

?Highly ratedVote to see the score

arXivLLM reasoning & chain of thought

Hierarchical Continuous Diffusion Language Models

HC-DLM couples discrete token generation with a continuous latent trajectory via a unified variational denoising objective, outperforming diffusion baselines on Sudoku, Countdown, and language modeling.

Hui Ren, Zihan Li, Chang Liu, Huidong Liu and 1 more

Published Oct 1, 2026 · 0 citations · ▲ 89 on Hugging Face · Code ★ 57

– 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.

72%Highly rated

?Highly ratedVote to see the score

arXivVideo generation

World Observer: Joint Actor-Observer Generation for Persistent World Modeling

World Observer jointly generates actor and panoramic observer views to continuously model out-of-view dynamics via shared geometric warping and observer sinks.

Hyunwook Choi, Dahyun Chung, Hyunsung Kim, Siyoon Jin and 3 more

Published Oct 1, 2026 · 0 citations · ▲ 85 on Hugging Face · Code ★ 30

– 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.

88%Must read

?Must readVote to see the score

arXivNational University of Singapore Robot manipulation

Agent Priors-guided Policy Learning

Agent Priors-guided Policy Learning embeds structural priors in skill interfaces to enable compositional and out-of-distribution skill generalization.

Puming (Oscar) Jiang, Tao Hu, Haozhe Du, Yibo Li and 3 more

Published Sep 28, 2026 · 0 citations · ▲ 84 on Hugging Face · Code ★ 3

– 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.

89%Must read

arXivVision-language-action models

World Action Modeling with Progressive Visual Planning

ProWAM predicts sparse visual sub-goals and actions via progressive planning, achieving state-of-the-art long-horizon robotic control and strong zero-shot real-world generalization.

Fei Zhang, Zhaochong An, Duncan Frost, Yikai Wang and 4 more

Published Oct 1, 2026 · 0 citations · ▲ 83 on Hugging Face

– ReadersNo votes yet

16/20 AI panelreviewers recommend it

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

76%Highly rated

?Highly ratedVote to see the score

arXivLLM agents & planning

Learning Meta-Skills for Agent Harness Design in Test-Time AI4AI

A Builder learns reusable meta-skills from target feedback to construct execution harnesses that boost target performance on unseen tasks. Meta-skills improve macro-average scores by 8.95 points over no-skill construction and 12.02 over direct delivery.

Cheng Qian, Kunlun Zhu, Beibin Li, Zhenhailong Wang and 1 more

Published Sep 29, 2026 · 0 citations · ▲ 83 on Hugging Face · Code ★ 9

– 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.

80%Must read

?Must readVote to see the score

arXivLLM agents & planning

ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization

ActiveSaddler automates curriculum learning for agent harness optimization via non-stationary bandits that adapt training scenarios to evolving failure patterns, boosting Pass@1 by 4.4, 7.5 points.

Sungho Park, Wonjoong Kim, Jue Zhang, Wook-Shin Han and 7 more

Published Oct 1, 2026 · ▲ 82 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.

AI panel: 12 of 20 reviewers recommend it

lenient 4/5

medium 8/10

strict 0/5

76%Highly rated

?Highly ratedVote to see the score

arXivInstitute of Computing Technolog U Chinese Academy of Sciences Alignment & preference optimization

On-Policy Parameter Update Direction Underlies Generalization in LLM Post-Training

On-policy methods continuously adjust parameter update directions, unlike consistent SFT updates; constraining SFT to these directions via OPSFT transfers on-policy generalization advantages to supervised fine-tuning.

Shufan Shen, Zhongni Hou, Junshu Sun, Yufei Zhang and 4 more

Published Sep 29, 2026 · 0 citations · ▲ 81 on Hugging Face

– 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.

80%Must read

arXivVision-language-action models

Native Action-Prior Learning from Videos for World Action Models

NAVA-WAM pretrains robot action policies directly from observation-only videos via flow-matching and joint attention, improving control accuracy and label efficiency.

Zhaochong An, Fei Zhang, Menglin Jia, Duncan Frost and 9 more

Published Oct 2, 2026 · 0 citations · ▲ 81 on Hugging Face

– ReadersNo votes yet

12/20 AI panelreviewers recommend it

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.

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

arXivLLM agents & planning

EVOKE: Eliciting World Knowledge in Agents for Transferable Decision-Making

EVOKE improves LLM agent transfer by ranking actions under diverse goals at fixed states to elicit pretrained world knowledge for robust decision-making.

Yuhan Guo, Jinming Liu, Liang Xu, Ziqiang Li and 7 more

Published Sep 29, 2026 · 0 citations · ▲ 78 on Hugging Face · Code ★ 7

– 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.

86%Must read

?Must readVote to see the score

arXivRepresentation & contrastive learning

JEPA-Anything: Learning Predictive Models across Different Worlds

JEPA-Anything uses orthogonal predictive factorization to learn cross-domain predictive models that outperform baselines in vision, biology, clinical, control, molecular, physical, and weather domains.

Taoyong Cui, Zhongyao Wang, Xinyue Xu, Weiyang Liu and 9 more

Published Sep 17, 2026 · 0 citations · ▲ 77 on Hugging Face · Code ★ 261

– 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.

89%Must read

?Must readVote to see the score

arXivVirginia Tech U Maryland, College Park Mohamed bin Zayed University of LLM evaluation & benchmarks

A Missing Piece for Trustworthy AI Reviewers: From Benchmarking Rhetorical Robustness to SciCore Review

Rhetorical robustness requires stable judgments across content-preserving rewrites and discrimination across papers; SciCore improves both via dual-branch science-core review.

Chenguang Wang, Ming Li, Chengrui Fan, Jianpeng Chen and 3 more

Published Sep 30, 2026 · 0 citations · ▲ 76 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.

86%Must read

?Must readVote to see the score

arXivU Waterloo Duke National University of Singapore LLM agents & planning

X-Tree: Tokenizing Reusable Experience for Efficient Agent Generalization

X-Tree learns reusable hierarchical skills from agent trajectories and improves success rates up to 5.8% across web and science benchmarks.

Sitao Cheng, Xunjian Yin, Zhiyuan Sun, Yuxuan Li and 3 more

Published Sep 26, 2026 · 0 citations · ▲ 76 on Hugging Face · Code ★ 2

– 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.

88%Must read

?Must readVote to see the score

arXivLong-context modeling

Transformers Stop Thinking Too Early, and a Tiny LoRA Fixes It

Pretrained transformers stop following references after 1.4, 3.6 lines, but a rank-8 LoRA at one early layer extends computation to 50, 160 lines without changing frozen weights.

Zehao Jin, Ruixuan Deng, 君然王

Published Sep 29, 2026 · 0 citations · ▲ 75 on Hugging Face · Code ★ 3

– 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.

72%Highly rated

?Highly ratedVote to see the score

arXivThe University of Texas at Austi Zhejiang Hong Kong Polytechnic U Science and Technology of Chin Beijing University of Posts and LLM reasoning & chain of thought

On the Geometry of On-Policy Distillation

On-policy distillation updates occupy a sparse, low-dimensional parameter subspace that is functionally sufficient and geometrically distinct from supervised fine-tuning and reinforcement learning.

Zhennan Shen, Yanshu Li, Qingyu Yin, Chak Tou Leong and 5 more

Published Jun 5, 2026 · 0 citations · ▲ 75 on Hugging Face

– ReadersNo votes yet. 1 from authors or colleagues not counted

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 3/5

medium 4/10

strict 1/5

78%Highly rated

NeurIPS 2026Multimodal reasoning

Imagine3D-LLM: Teaching MLLMs to Imagine 3D Scenes Before Answering

Imagine3D-LLM learns compact scene representations via learnable summary tokens decoded into 3D Gaussian splatting with photometric loss, improving multi-view spatial reasoning benchmarks.

Jaewoo Jung, Hyeonseo Yu, Honggyu An, Jisang Han and 9 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published Sep 29, 2026 · ▲ 72 on Hugging Face · Code ★ 37

– ReadersNo votes yet

11/20 AI panelreviewers recommend it

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 6/10

strict 1/5

Trending