Ax Xuanyu Zhu, Yan Bai, Yang Shi, Yihang Lou, Yuanxing Zhang, Jing Jin, Yuan Zhou 5/12/2026

Beyond the Last Layer: Multi-Layer Representation Fusion for Visual Tokenizatio

Visual tokenization method using multi-layer representation fusion from frozen vision encoders for better reconstruction.

Ax Ari Holtzman, Peter West 5/12/2026

Can You Keep a Secret? Involuntary Information Leakage in Language Model Writing

Study of involuntary information leakage from prompted secrets in language model outputs using multi-model detection.

Ax Gabriel Garcia 5/12/2026

The Last Word Often Wins: A Format Confound in Chain-of-Thought Corruption Studies

Identifies format confound in chain-of-thought corruption studies that detects answer text location rather than actual computation.

Ax Juyi Lin, Arash Akbari, Yumei He, Lin Zhao, Haichao Zhang, Arman Akbari, Xingchen Xu, Zoe Y. Lu, Enfu Nan, Hokin Deng, Edmund Yeh, Sarah Ostadabbas, Yun Fu, Jennifer Dy, Pu Zhao, Yanzhi Wang 5/12/2026

PhyGround: Benchmarking Physical Reasoning in Generative World Models

Benchmark suite for evaluating physical reasoning and dynamics accuracy in generative video world models.

Ax Saba Pourhanifeh, AbdulAziz AbdulGhaffar, Ashraf Matrawy 5/12/2026

Threat Modelling using Domain-Adapted Language Models: Empirical Evaluation and Insights

Empirical evaluation of domain-adapted language models versus general-purpose LLMs for cybersecurity threat modeling tasks.

Ax Avik Kar, Siddharth Chandak, Rahul Singh, Soumitra Sinhahajari, Eric Moulines, Shalabh Bhatnagar, Nicholas Bambos 5/12/2026

Policy Gradient Methods for Non-Markovian Reinforcement Learning

Policy gradient method analysis for reinforcement learning in non-Markovian environments with internal state management.

Ax Zuojin Tang, Haoyun Liu, Xinyuan Chang, Changjie Wu, Dongjie Huo, Yandan Yang, Bin Liu, Zhejia Cai, Feng Xiong, Mu Xu, jiachen Luo, De Ma, Zhiheng Ma, Gang Pan 5/12/2026

ALAM: Algebraically Consistent Latent Transitions for Vision-Language-Action Models

Algebraic framework for latent action models in vision-language-action robotics using action-free video data.

Ax Mingxu Zhang, Yuhan Li, Lujundong Li, Dazhong Shen, Hui Xiong, Ying Sun 5/12/2026

SLIM: Sparse Latent Steering for Interpretable and Property-Directed LLM-Based Molecular Editing

Sparse latent steering technique for interpretable control of molecular editing properties in LLMs with explicit property handles.

Ax Yixuan Yang, Mehak Arora, Ryan Zhang, Baraa Abed, Junseob Kim, Tilendra Choudhary, Md Hassanuzzaman, Kevin Zhu, Ayman Ali, Chengkun Yang, Alasdair Edward Gent, Victor Moas, Rishikesan Kamaleswaran 5/12/2026

Clin-JEPA: A Multi-Phase Co-Training Framework for Joint-Embedding Predictive Pretraining on EHR Patient Trajectories

Multi-phase pretraining framework for learning representations from electronic health records using joint-embedding predictive architecture.

Ax Huynh Trung Kiet, Dao Sy Duy Minh, Tuan Nguyen, Chi-Nguyen Tran, Phu-Hoa Pham, Nguyen Lam Phu Quy, The Anh Han, Long Tran-Thanh 5/12/2026

Training-Free Cultural Alignment of Large Language Models via Persona Disagreement

Training-free method for aligning LLM behavior across cultures without fine-tuning or model access, using black-box prompting techniques.

Ax Tz-Huan Hsu, Jheng-Hong Yang, Jimmy Lin 5/12/2026

Rethinking Agentic Search with Pi-Serini: Is Lexical Retrieval Sufficient?

Pi-Serini search agent system pairing lexical retrieval with frontier LLMs for agentic research tasks, evaluating if BM25 suffices for modern reasoning-capable agents.

Ax Anna C. Doris, Jacob Thomas Sony, Ghadi Nehme, Era Syla, Amin Heyrani Nobari, Faez Ahmed 5/12/2026

CADBench: A Multimodal Benchmark for AI-Assisted CAD Program Generation

Multimodal benchmark dataset with 18,000 samples for evaluating AI-assisted CAD program generation from images and 3D observations.

Ax Edward De Brouwer, Carl Edwards, Alexander Wu, Jenna Collier, Graham Heimberg, Xiner Li, Meena Subramaniam, Ehsan Hajiramezanali, David Richmond, Jan-Christian H\"utter, Sara Mostafavi, Gabriele Scalia 5/12/2026

AssayBench: An Assay-Level Virtual Cell Benchmark for LLMs and Agents

Benchmark for evaluating LLMs and agents on virtual cell modeling tasks, testing in silico phenotypic screening and biological discovery prediction.

Ax Linus Heck, Filip Mac\'ak, Roman Andriushchenko, Milan \v{C}e\v{s}ka, Sebastian Junges 5/12/2026

Shields to Guarantee Probabilistic Safety in MDPs

Formal framework for shielding techniques in probabilistic Markov decision processes, extending safety guarantees for autonomous agents with acceptable failure probabilities.

Ax Mohammadreza Armandpour, Fatih Ilhan, David Harrison, Ajay Jaiswal, Duc N. M Hoang, Fartash Faghri, Yizhe Zhang, Minsik Cho, Mehrdad Farajtabar 5/12/2026

Unmasking On-Policy Distillation: Where It Helps, Where It Hurts, and Why

Analysis of on-policy distillation for training reasoning models, investigating when teacher-student supervision helps or hurts performance on token-level tasks.

Ax Yaxin Du, Xiyuan Yang, Zhifan Zhou, Wanxu Liu, Zixing Lei, Zimeng Chen, Fenyi Liu, Haotian Wu, Yuzhu Cai, Zexi Liu, Xinyu Zhu, WenHao Wang, Linfeng Zhang, Chen Qian, Siheng Chen 5/12/2026

DataMaster: Towards Autonomous Data Engineering for Machine Learning

Study of autonomous data engineering for ML systems, automating dataset discovery, adaptation, and validation to reduce manual data engineering workflows.

Ax Roxana Geambasu (Google,Columbia University), Mariana Raykova (Google), Pierre Tholoniat (Google), Trishita Tiwari (Google), Lillian Tsai (Google), Wen Zhang (Google) 5/12/2026

Engineering Robustness into Personal Agents with the AI Workflow Store

Research on engineering robustness into AI agents by applying traditional software engineering processes like testing, adversarial evaluation, and staged deployment instead of on-the-fly synthesis.

Ax Keya Hu, Linlu Qiu, Yiyang Lu, Hanhong Zhao, Tianhong Li, Yoon Kim, Jacob Andreas, Kaiming He 5/12/2026

ELF: Embedded Language Flows

Continuous diffusion language models using minimal adaptation to match effectiveness of leading discrete-token language model approaches.

Ax Himanshu Gupta, Shreyas Verma, Ujjwala Anantheswaran, Kevin Scaria, Mihir Parmar, Swaroop Mishra, Chitta Baral 5/12/2026

Polymath: A Challenging Multi-modal Mathematical Reasoning Benchmark

PolyMATH benchmark with 5,000 images evaluating multimodal LLM visual comprehension and abstract reasoning across 10 cognitive challenge categories.

Ax Wenjie Tang, Yuan Zhou, Erqiang Xu, Keyan Cheng, Minne Li, Liquan Xiao 5/12/2026

DSGBench: A Diverse Strategic Game Benchmark for Evaluating LLM-based Agents in Complex Decision-Making Environments

DSGBench evaluation platform for LLM-based agents in strategic games assessing long-horizon reasoning, multi-agent interaction, and decision-making.

Ax Matthew T. Dearing, Yiheng Tao, Xingfu Wu, Zhiling Lan, Valerie Taylor 5/12/2026

Leveraging LLMs to Automate Energy-Aware Refactoring of Parallel Scientific Codes

Framework using LLMs to automate energy-aware refactoring of parallel scientific code focusing on energy efficiency beyond execution time.

Ax Haorui Wang, Jeff Guo, Lingkai Kong, Rampi Ramprasad, Philippe Schwaller, Yuanqi Du, Chao Zhang 5/12/2026

LLM-Augmented Chemical Synthesis and Design Decision Programs

LLM-augmented retrosynthesis system for chemical synthesis and drug development combining ML and LLMs to navigate combinatorial pathway space.

Ax Chunhui Zhang, Zhongyu Ouyang, Kwonjoon Lee, Nakul Agarwal, Sean Dae Houlihan, Soroush Vosoughi, Shao-Yuan Lo 5/12/2026

Overcoming Multi-step Complexity in Multimodal Theory-of-Mind Reasoning: A Scalable Bayesian Planner

Scalable Bayesian planner for multimodal theory-of-mind reasoning inferring beliefs and intentions without task-specific priors.

Ax Byeongchan Lee, Jonghoon Lee, Dongyoung Kim, Jaehyung Kim, Kyungjoon Park, Dongjun Lee, Jinwoo Shin 5/12/2026

Efficient LLM Collaboration via Planning

Planning-based framework for efficient LLM collaboration combining large and small models to reduce inference costs while maintaining performance.

Ax Shufan Jiang, Sizhou Chen, Chios Chen, Chi Zhang, Xiao-Lei Zhang, Xuelong Li 5/12/2026

HAMLET: A Hierarchical and Adaptive Multi-Agent Framework for Live Embodied Theatrics

HAMLET framework combines hierarchical multi-agent LLMs for interactive theatrical experiences with embodied interaction and initiative.

Ax Zahra Khotanlou, Kate Larson, Amir-Hossein Karimi 5/12/2026

Your Recourse, My Loss? Algorithmic Recourse under Shared Constraints

Study of algorithmic recourse for ML-driven decisions addressing multi-stakeholder scenarios with shared constraints.

Ax Pratik Jayarao, Himanshu Gupta, Neeraj Varshney, Chaitanya Dwivedi 5/12/2026

Explicit Reasoning Makes Better Judges: A Systematic Study on Accuracy, Efficiency, and Robustness

Systematic comparison of reasoning vs non-reasoning LLMs in judge role evaluating accuracy, efficiency, and robustness on small models.

Ax Yuanyi Wang, Yanggan Gu, Yiming Zhang, Qi Zhou, Zhaoyi Yan, Congkai Xie, Xinyao Wang, Jianbo Yuan, Hongxia Yang 5/12/2026

Model Merging Scaling Laws in Large Language Models

Empirical scaling laws for language model merging showing power law relationship between model size, expert number, and merging performance.

Ax Minhui Zhu, Minyang Tian, Xiaocheng Yang, Tianci Zhou, Lifan Yuan, Penghao Zhu, Eli Chertkov, Shengyan Liu, Yufeng Du, Ziming Ji, Indranil Das, Qingzhi Chen, Junyi Cao, Yufeng Du, Jiabin Yu, Peixue Wu, Jinchen He, Yifan Su, Yikun Jiang, Yujie Zhang, Chang Liu, Ze-Min Huang, Weizhen Jia, Yunkai Wang, Farshid Jafarpour, Yong Zhao, Xinan Chen, Jessie Shelton, Aaron W. Young, John Bartolotta, Wenchao Xu, Yue Sun, Anjun Chu, Victor Colussi, Chris Akers, Nathan Brooks, Wenbo Fu, Jinchao Zhao, Marvin Qi, Anqi Mu, Yubo Yang, Allen Zang, Yang Lyu, Peizhi Mai, Christopher Wilson, Xuefei Guo, Juntai Zhou, Daniel Inafuku, Chi Xue, Luyu Gao, Ze Yang, Ya\"ir Hein, Yonatan Kahn, Kevin Zhou, Di Luo, John Drew Wilson, Jarrod T. Reilly, Dmytro Bandak, Ofir Press, Liang Yang, Xueying Wang, Hao Tong, Nicolas Chia, Eliu Huerta, Hao Peng 5/12/2026

Probing the Critical Point (CritPt) of AI Reasoning: a Frontier Physics Research Benchmark

CritPt benchmark evaluates LLM reasoning on complex open-ended frontier physics research challenges beyond high-school math and coding.

Ax Alexandra Yost, Shreyans Jain, Shivam Raval, Grant Corser, Allen Roush, Nina Xu, Jacqueline Hammack, Ravid Shwartz-Ziv, Amirali Abdullah 5/12/2026

Measure what Matters: Psychometric Evaluation of AI with Situational Judgment Tests

Framework using situational judgment tests and multidimensional item response theory to measure stable behavioral tendencies in persona-conditioned LLMs.

Ax Adam Tauman Kalai, Yael Tauman Kalai, Or Zamir 5/12/2026

Consensus Sampling for Safer Generative AI

Consensus sampling algorithm aggregates multiple probability distributions to improve generative AI safety with architecture-agnostic approach.

Ax Yannick Brunink, Daniel Daza, Yunjie He, Michael Cochez 5/12/2026

Counting Still Counts: Understanding Neural Complex Query Answering Through Query Relaxation

Analysis of neural complex query answering over knowledge graphs comparing learned patterns with training-free query relaxation strategies.

Ax Miles Q. Li, Benjamin C. M. Fung, Martin Weiss, Pulei Xiong, Khalil Al-Hussaeni, Claude Fachkha 5/12/2026

A Benchmark for Evaluating Outcome-Driven Constraint Violations in Autonomous AI Agents

Benchmark for evaluating outcome-driven constraint violations in autonomous AI agents, addressing safety and alignment in high-stakes deployment scenarios.

Ax Alex L. Zhang, Tim Kraska, Omar Khattab 5/12/2026

Recursive Language Models

Recursive Language Models enable LLMs to process arbitrarily long prompts through inference-time scaling via recursive self-calling over prompt snippets.

Ax Xuan Yang, Furong Jia, Roy Xie, Xiong Xi, Hengwei Bian, Jian Li, Monica Agrawal 5/12/2026

Batch-of-Thought: Cross-Instance Learning for Enhanced LLM Reasoning

Batch-of-Thought method processes related queries jointly to improve LLM reasoning by identifying high-quality reasoning templates and detecting errors through consistency analysis.

Ax Ian Rios-Sialer 5/12/2026

The Homogenization Problem in LLMs: Towards Meaningful Diversity in AI Safety

Framework for characterizing and measuring homogenization and mode collapse in LLMs as an AI safety concern affecting diversity.

Ax Kiana Jafari, Paul Ulrich Nikolaus Rust, Duncan Eddy, Robbie Fraser, Nina Vasan, Darja Djordjevic, Akanksha Dadlani, Max Lamparth, Eugenia Kim, Mykel Kochenderfer 5/12/2026

Expert Evaluation and the Limits of Human Feedback in Mental Health AI Safety Testing

Study on inter-rater reliability limitations of human feedback for mental health LLM evaluation, questioning assumptions in RLHF approaches.

Ax Jingyun Wang, Dian Li, Xiaohan Wang, Gang Liu, Jiahong Yan, Guoliang Kang 5/12/2026

Concise Geometric Description as a Bridge: Unleashing the Potential of LLM for Plane Geometry Problem Solving

Method using concise geometric descriptions to improve multimodal LLM performance on plane geometry problem solving tasks.

Ax Tristan Shah, Stas Tiomkin 5/12/2026

Emergence of Physical Intelligence via Controllable Information Production

Proposes Controllable Information Production (CIP) as a principled intrinsic motivation framework for training autonomous agents without external rewards.

Ax Yueyi Yang, Haotian Liu, Fang Kang, Mengqi Zhang, Zheng Lian, Hao Tang, Haoyu Chen 5/12/2026

SayNext-Bench: Why Do LLMs Struggle with Next-Utterance Anticipation?

Research benchmark (SayNext-Bench) examining why LLMs struggle with next-utterance anticipation in dialogue compared to human multimodal understanding.

Ax Xingyuan Hua, Sheng Yue, Xinyi Li, Yizhe Zhao, Jinrui Zhang, Ju Ren 5/12/2026

Context Learning for Multi-Agent Discussion

M2CL framework improves multi-agent discussion by learning context representations that align individual LLM instances toward coherent solutions.

Ax Wenzhe Fan, Tommaso Tognoli, Henry Peng Zou, Chunyu Miao, Yibo Wang, Xinhua Zhang 5/12/2026

TodyComm: Task-Oriented Dynamic Communication for Multi-Round LLM-based Multi-Agent System

TodyComm enables dynamic communication topology for multi-agent LLM systems that adapts across conversation rounds based on task progression.

Ax Yansong Ning, Jun Fang, Naiqiang Tan, Hao Liu 5/12/2026

Agent-Omit: Adaptive Context Omission for Efficient LLM Agents

Agent-Omit adaptively omits unnecessary context during multi-turn agent interactions to improve efficiency without sacrificing performance.

Ax Khurram Yamin, Jingjing Tang, Santiago Cortes-Gomez, Amit Sharma, Eric Horvitz, Bryan Wilder 5/12/2026

When Agents Say One Thing and Do Another: Validating Elicited Beliefs from LLMs

Decision-theoretic framework validates whether LLMs hold coherent beliefs by comparing probability judgments to decision behavior.

Ax Shengxuan Qiu, Haochen Huang, Shuzhang Zhong, Pengfei Zuo, Meng Li 5/12/2026

HyPER: Bridging Exploration and Exploitation for Scalable LLM Reasoning with Hypothesis Path Expansion and Reduction

HyPER framework optimizes test-time compute for LLM reasoning by balancing exploration-exploitation through hypothesis path expansion and reduction.

Ax Jialin Wu, Wei Shi, Han Shen, Peigui Qi, Kunsheng Tang, Zhicong Huang, Binghao Wang, Zhou Yang 5/12/2026

Revis: Sparse Latent Steering to Mitigate Object Hallucination in Large Vision-Language Models

REVIS training-free framework reduces object hallucination in vision-language models through sparse latent steering.

Ax Varun Ursekar, Apaar Shanker, Veronica Chatrath, Yuan Xue, Sam Denton 5/12/2026

VeRO: An Evaluation Harness for Agents to Optimize Agents

VeRO evaluation harness systematically measures coding agent performance on agent optimization through iterative edit-execute-evaluate cycles.

Ax Ik-hwan Kim, Hyeongrok Han, Mingi Jung, Sangwon Yu, Jinseok Hong, Sang Hun Kim, Yoonyoung Choi, Sungroh Yoon 5/12/2026

Metacognitive Behavioral Tuning of Large Language Models for Multi-Hop Question Answering

Metacognitive behavioral tuning improves LLM multi-hop reasoning by strengthening self-regulation of intermediate conclusions.

Ax Elron Bandel, Asaf Yehudai, Lilach Eden, Yehoshua Sagron, Yotam Perlitz, Elad Venezian, Natalia Razinkov, Natan Ergas, Shlomit Shachor Ifergan, Segev Shlomov, Michal Jacovi, Leshem Choshen, Liat Ein-Dor, Yoav Katz, Michal Shmueli-Scheuer 5/12/2026

General Agent Evaluation

First systematic comparison of agent architectures (tool-calling, MCP, code-generation, CLI) across heterogeneous environments and benchmarks.

Ax Junkai Lu, Peng Chen, Xingjian Wu, Yang Shu, Chenjuan Guo, Christian S. Jensen, Bin Yang 5/12/2026

PATRA: Pattern-Aware Alignment and Balanced Reasoning for Time Series Question Answering

PATRA framework improves LLM performance on time series question answering by capturing temporal patterns and balancing task complexity.