Science Utopia? Closed-Loop LLM Simulation of Academic Research Ecosystems

SciUtopia is a closed-loop LLM simulation framework modeling entire academic ecosystems; it finds resubmission amplifies reviewer burden, cautious exploration balances impact and diversity, and inequality can emerge without cumulative funding advantage.

Yiqiao Jin, Yiyang Wang, Lucheng Fu, Bing He and 7 more

Published Oct 1, 2026 · 0 citations · ▲ 40 on Hugging Face · Code ★ 29

When do Prophets Profit in Prediction Markets?

Anri Gu, Nicole Kagan, Alec Sun, Jibang Wu and 1 more

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

Towards Financial World Modeling

Humzah Merchant, Alec Guthrie, Simon Mahns, Randall Balestriero and 1 more

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

The $1/\mathcal{W}$ Law: Context Length is the Dominant Energy Lever in LLM Inference Fleets

Huamin Chen, Xunzhuo Liu, Yuhan Liu, Junchen Jiang and 3 more

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

Model Incrimination: Investigating Whether Concerning Behavior Reflects Misalignment

Gerson Kroiz, Aditya Singh, Senthooran Rajamanoharan, Neel Nanda

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

RAG in a Trenchcoat: When Minimal Memory Is Enough for Agentic Systems, and When It Isn’t

Jingyu Liu, Zongze Li, Zach Xu, Zhanhui Zhou and 3 more

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

NitroBox: Lightning-Fast Sandbox for Large-Scale RL Training

Yuzhou Nie, Ruilin Zhou, Zhaorun Chen, Jingyang Zhang and 5 more

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

Quantifying and Optimizing Path Uncertainty in Masked Diffusion Models

Ziyu Chen, Xinbei Jiang, PENG SUN, Tao Lin

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

Beyond Ground Truth: Evaluating Non-Verifiable Reasoning in LLMs through Moral Robustness

Elizaveta Tennant, Benjamin Henke, Anita Keshmirian, Murray Shanahan and 4 more

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

Is Decentralized LLM Agent RL Robust to Heterogeneity? An Asymmetric Tale

Canyu Chen, Kangyu Zhu, Zhaorun Chen, Zhanhui Zhou and 5 more

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

ML-assisted Randomization Tests for A/B Experiments

Wenxuan Guo, JungHo Lee, Panos Toulis

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

SAFE-DRIFT: Data Selection for Supervised Fine-tuning with Controllable Off-Target Drifts

Yeo Jin Jung, Yating Liu, Lalchand Pandia, Claire Donnat

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

Personalized LLM Alignment Should Be Counterfactually Verifiable

Cristina Garbacea

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

Introspection Tools Help LLMs Understand and Control Themselves

Xiao Liu, Junsol Kim, Shiyang Lai, Jacy Anthis and 3 more

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

Bias, Measurement Error, and Double-Dipping: When Can GNN Convolutions Help Brain Connectome Prediction?

Tommaso Castellani, Jiaqi Li, Muriah D Wheelock, Rezwana R Razzaque and 3 more

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

Loyalty Capture: Reporting Relationships and Structural Sycophancy in Frontier AI Models

Eric So, Alex Imas

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

Agent-Native Research Artifacts

Jiachen Liu, Jiaxin Pei, Jintao Huang, Chenglei Si and 33 more

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

HYDRA: Representation Harmonized Tokenization for Multimodal Generation and Understanding

Xuerui Qiu, Yutao Cui, Guozhen Zhang, Junzhe Li and 8 more

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

Statistical Matching via Schr\"odinger Bridge beyond Conditional Independence

Eunho Koo, Jinwon Sohn, Tongseok Lim

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

Precise Debugging Benchmark: Is Your Model Debugging or Regenerating?

Miaosen Chai, Wang Bill Zhu, Shangshang Wang, Yejia Liu and 4 more

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

Data Auctions for Retrieval Augmented Generation

Minbiao Han, Seyed A Esmaeili, Michael Albert, Haifeng Xu

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

Overcoming Attention Distraction: Training-Free Latent Communication for Multi-Agent Systems

zhengjie zhou, Yahao Liu, Ziyue Feng, Tengfei LIU and 1 more

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

Bypassing PC1 Makes SAEs More Reproducible

Nathan Delisle, Chenhao Tan

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

Empirical Bayes Rebiasing

An empirical Bayes rebiasing method learns the bias distribution to recover shorter calibrated intervals from noisy biased estimates, improving precision in LLM evaluations and genetic analysis.

Wanyi Ling, Sida Li, Junming Guan, Nikolaos Ignatiadis

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

INFUSER: Influence-Guided Self-Evolution Improves Reasoning

INFUSER co-evolves a question generator and solver via influence-guided rewards, improving reasoning by over 20% on math benchmarks without curated data.

Siyu Chen, Miao Lu, Beining Wu, Heejune Sheen and 6 more

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

CruxBench: A Benchmark of Information Discovery

CruxBench evaluates LLM information discovery via Value of Information for decomposing forecasting problems into key subquestions, finding frontier models barely exceed random baselines.

Lina Piao, Amelia Hui Dai, Nick Merrill, Nadja Flechner and 2 more

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

Distilling Sequential Computation in Transformer Language Models

A lightweight merge module replaces token spans with surrogate embeddings, cutting Transformer sequence lengths by up to 40% with minimal accuracy loss and no retraining.

Zixuan Lan, Jessica Yang, Yanhong Li, Karen Livescu and 1 more

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

GenEnv: Difficulty-Aligned Co-Evolution Between LLM Agents and Environment Simulators

GenEnv co-evolves LLM agents with generative simulators via difficulty-aligned curricula, improving 7B agents by up to 40.3% with 3.3x less data.

Jiacheng Guo, Ling Yang, Peter Chen, Qixin Xiao and 4 more

Atlanta Poster Session 2, Wed, Dec 9, 4:30 PM–7:30 PM, Hall C1 · Published 2026 · ▲ 19 on Hugging Face · Code ★ 67

Fine-Tuning Improves Information Conveyance in Language Models

Canopy Entropy reveals fine-tuning reorganizes language model uncertainty into longer, more semantically diverse outputs rather than reducing it. Fine-tuned models show stronger positive correlation between output length and per-token information efficiency, tripling entropy-diversity alignment.

Yuwei Cheng, Weiyi Tian, Haifeng Xu

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

Learning to Trigger: Reinforcement Learning at the Large Hadron Collider

Reinforcement learning agents adapt Large Hadron Collider trigger thresholds online to maximize signal efficiency while maintaining background rates within tolerance bands, improving in-tolerance intervals by up to 56% on real CMS collision data without fine-tuning.

Zixin Ding, Shaghayegh Emami, Giovanna Salvi, Cecilia Tosciri and 6 more

Atlanta Poster Session 3, Thu, Dec 10, 10:00 AM–1:00 PM, Hall C1 · Published 2026 · ▲ 6 on Hugging Face · Code ★ 3

CELEUS: Certifiable and Efficient LLM Evaluation via E-Processes

CELEUS uses E-processes with uncertainty-guided sampling and surrogate approximations to provide anytime-valid confidence intervals for LLM evaluation, cutting required samples by 54-62%.

Zhijian Zhou, Zesheng Ye, Zhaorun Chen, Bo Li and 1 more

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

Flow Matching from Viewpoint of Proximal Operators

Optimal transport conditional flow matching equals exact proximal operators via extended Brenier potentials without density assumptions, yields explicit vector fields, converges with batch size, and contracts exponentially normal to manifold-supported targets.

Kenji Fukumizu, Wei Huang, Han Bao, Shuntuo Xu and 1 more

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

Zeroth-Order Sharpness-Aware Learning with Exponential Tilting

An exponential tilting objective unifies zeroth-order smoothing and sharpness-aware minimization, yielding gradient-free algorithms that improve generalization over baselines.

Xuchen Gong, Tian Li

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

Local linear convergence of gradient methods for overparameterized Gaussian mixtures

Overparameterized Gaussian mixtures have a loss manifold of slow growth where Polyak steps achieve geometric loss reduction, and alternating short gradient steps with long Polyak steps yields local linear convergence to near-optimal solutions.

Jingxing Wang, Vasileios Charisopoulos, Maryam Fazel

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

Mecha-nudges for Machines

Mecha-nudging changes online choice environments to systematically influence AI agents without harming human usability, and Etsy listings show a 0.143-bit rise in machine-usable information after ChatGPT's release.

Giulio Frey, Kawin Ethayarajh

Atlanta Poster Session 4, Thu, Dec 10, 4:30 PM–7:30 PM, Hall C1 · Published 2026 · ▲ 1 on Hugging Face

ActWorld: From Explorable to Interactive World Model via Action-Aware Memory

ActWorld extends interactive world models to object interaction via a 100K dataset and hierarchical action-aware memory, improving fidelity over navigation-only baselines.

Zhexiao Xiong, Yizhi Song, Hao Kang, Qing Yan and 9 more

Atlanta Poster Session 2, Wed, Dec 9, 4:30 PM–7:30 PM, Hall C1 · Published 2026 · ▲ 8 on Hugging Face

SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning

SkillRL evolves agents via recursive skill-augmented reinforcement learning with automatic skill discovery and hierarchical library co-evolution, cutting token use while achieving state-of-the-art results across complex tasks.

Peng Xia, Jianwen Chen, Hanyang Wang, Jiaqi Liu and 9 more

Atlanta Poster Session 3, Thu, Dec 10, 10:00 AM–1:00 PM, Hall C1 · Published 2026 · ▲ 75 on Hugging Face · Code ★ 998

Hydra-X: Native Unified Multimodal Models with Holistic Visual Tokenizers

Hydra-X unifies image and video tokenization in one vision transformer via causal temporal attention and hierarchical compression, achieving strong unified understanding and generation performance.

Guozhen Zhang, Xuerui Qiu, Yutao Cui, Tianhui Song and 10 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 31 on Hugging Face

JudgmentBench: Comparing Rubric and Preference Evaluation for Quality Assessment

JudgmentBench compares rubric and comparative evaluation by collecting both expert judgments on 30 legal tasks, finding pairwise preferences recover quality rankings far better with less time.

Russell Yang, Ruishi Chen, Pierce Kelaita, Riya Ranjan and 5 more

Atlanta Poster Session 5, Fri, Dec 11, 10:00 AM–1:00 PM, Hall C1 · Published 2026

High-dimensional Gaussian Graphical Model Testing for Long-Memory Time Series

A direct data-adaptive test for Gaussian graphical models in high-dimensional long-memory time series achieves asymptotic size and power consistency via block bootstrap.

Percy S. Zhai, Ping-Shou Zhong, Wei Biao Wu

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