Ax Yujia Zheng, Fan Feng, Yuke Li, Shaoan Xie, Kevin Murphy, Kun Zhang 5/14/2026

From Generalist to Specialist Representation

Nonparametric study of learning task-relevant specialist representations from generalist models with identifiability guarantees.

Ax Nirav Diwan, Han Wang, Berkcan Kapusuzoglu, Ramin Moradi, Supriyo Chakraborty, Giri Iyengar, Sambit Sahu, Huan Zhang, Gang Wang 5/14/2026

CoT-Guard: Small Models for Strong Monitoring

Small monitoring models for detecting covert misbehavior in LLM chain-of-thought reasoning, reducing cost vs. large model monitors.

Ax Kaiyang Li, Shaobo Han, Qing Su, Shihao Ji 5/14/2026

Bayesian Model Merging

Proposes Bayesian approach for merging task-specific expert models without retraining. Leverages anchor model bias for parameter estimation in model merging.

Ax Davi Bastos Costa, Renato Vicente 5/14/2026

Persona-Model Collapse in Emergent Misalignment

Hypothesizes emergent misalignment in fine-tuned LLMs involves persona-model collapse. Tests behavioral degradation using moral susceptibility and robustness metrics.

Ax Zhongkai Yu, Yichen Lin, Chenyang Zhou, Yuwei Zhang, Kun Zhou, Junxia Cui, Haotian Ye, Zhengding Hu, Zaifeng Pan, Ruiyi Wang, Yujie Zhao, Hejia Zhang, Jingbo Shang, Jishen Zhao, Yufei Ding 5/14/2026

ChipMATE: Multi-Agent Training via Reinforcement Learning for Enhanced RTL Generation

Multi-agent reinforcement learning system for RTL hardware code generation. Addresses vendor air-gap security and proprietary codebase training constraints in chip design.

Ax Timothy Zhou, Loris D'Antoni, Nadia Polikarpova 5/14/2026

Language-Based Agent Control

Introduces language-based agent control (LBAC) programming model using static typing and runtime enforcement for agentic applications. Brings security concepts from programming languages to agent control.

Ax Siyuan Liu (IIIS, Tsinghua University), Tinghong Chen (College of AI, Tsinghua University,Shanghai Qi Zhi Institute), Xinghan Li (IIIS, Tsinghua University), Yifei Wang (Amazon AGI SF Lab), Jingzhao Zhang (IIIS, Tsinghua University,Shanghai Qi Zhi Institute) 5/14/2026

Data Difficulty and the Generalization--Extrapolation Tradeoff in LLM Fine-Tuning

Studies data difficulty's role in LLM fine-tuning from empirical and theoretical perspectives. Examines tradeoff between generalization and extrapolation in supervised fine-tuning.

Ax Priyam Sahoo, Gaurav Mittal, Xiaomin Li, Shengjie Ma, Benjamin Steenhoek, Pingping Lin, Yu Hu 5/14/2026

AgentLens: Revealing The Lucky Pass Problem in SWE-Agent Evaluation

Identifies 'lucky pass' problem in SWE-Agent evaluation where agents pass tests through trial-and-error rather than principled solutions. Analyzes 2,614 trajectories to show outcome-only metrics miss process quality.

Ax Yanggan Gu, Shuo Cai, Zihao Wang, Wenjun Wang, Yuanyi Wang, Pengkai Wang, Sirui Huang, Su Lu, Jianmin Wu, Hongxia Yang 5/14/2026

FeatCal: Feature Calibration for Post-Merging Models

Technique to reduce feature drift and improve performance of merged expert models through feature calibration.

Ax Zeyu Huang, Adhiguna Kuncoro, Qixuan Feng, Jiajun Shen, Lucio Dery, Arthur Szlam, Marc'Aurelio Ranzato 5/14/2026

Context Training with Active Information Seeking

Method for adapting LLMs to downstream tasks via context optimization and active information seeking without weight updates.