Ax Baoqing Yue, Zihan Zhu, Yutong Han, Qian Sun, Jichen Feng, Hufei Yang, Yifan Zhang, Mengdi Wang 5/12/2026

Interactive Benchmarks

Interactive Benchmarks evaluation paradigm assesses model reasoning through active information acquisition rather than fixed tasks.

Ax Xingwu Chen, Zhanqiu Zhang, Yiwen Guo, Difan Zou 5/12/2026

Breaking Contextual Inertia: Reinforcement Learning with Single-Turn Anchors for Stable Multi-Turn Interaction

Framework addresses contextual inertia in multi-turn LLM interactions using single-turn anchors and reinforcement learning.

Ax Zhengwei Xie, Zhisheng Chen, Ziyan Weng, Jinhan Li, Chenglong Li, Zikai Xiao, Jingwei Song, Jinhao Jing, Vireo Zhang, Kun Wang 5/12/2026

MineEvolve: Self-Evolution with Accumulated Knowledge for Long-Horizon Embodied Minecraft Agents

MineEvolve framework enables long-horizon embodied agents in Minecraft to improve through accumulated knowledge and experience.

Ax Liang Ding 5/12/2026

AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling

AgentHER adapts Hindsight Experience Replay to recover training signal from failed LLM agent trajectories.

Ax Liang Ding 5/12/2026

AdaRubric: Task-Adaptive Rubrics for Reliable LLM Agent Evaluation and Reward Learning

AdaRubric generates task-specific evaluation rubrics for LLM agents, improving evaluation accuracy over fixed-rubric approaches.

Ax Yifan Wang, Bolian Li, David Cho, Ruqi Zhang, Fanping Sui, Ananth Grama 5/12/2026

SARL: Label-Free Reinforcement Learning by Rewarding Reasoning Topology

SARL framework enables reinforcement learning for reasoning models without requiring verifiable rewards in open-ended domains.

Ax Aaron Rose, Carissa Cullen, Sahar Abdelnabi, Philip Torr, Brandon Gary Kaplowitz, Christian Schroeder de Witt 5/12/2026

Detecting Multi-Agent Collusion Through Multi-Agent Interpretability

arXiv research on detecting multi-agent collusion using interpretability. Introduces NARCBench benchmark for covert agent coordination.

Ax Tianjun Feng, Yunfeng Chen, Chun-Yi Tsai, Yihan Sun, Ayan Das, Kaoutar El Maghraoui, Shuxin Lin, Dhaval Patel 5/12/2026

PHMForge: Evaluating LLM Agents on Industrial Prognostics through MCP-Native, Algorithm-Grounded Tools

arXiv: PHMForge evaluation environment for LLM agents using Model Context Protocol on industrial asset management tasks. Benchmarks agent reliability for safety-critical systems.

Ax Jiaxuan Wang, Yulan Hu, Wenjin Yang, Zheng Pan, Xin Li, Lan-Zhe Guo 5/12/2026

Aligning Agents via Planning: A Benchmark for Trajectory-Level Reward Modeling

arXiv: Benchmark for trajectory-level reward modeling in agentic LLMs with tool use. Research on RLHF alignment for agent systems.

Ax Hao Yan, Yuliang Liu, Xingchen Liu, Yuyi Zhang, Minghui Liao, Jihao Wu, Wei Chen, Xiang Bai 5/12/2026

DocSeeker: Structured Visual Reasoning with Evidence Grounding for Long Document Understanding

Multimodal LLM approach for long document understanding with structured visual reasoning and evidence grounding to handle signal-to-noise challenges.

Ax David Y. Y. Tan, Kellie Chin, Jingxian Zhang 5/12/2026

AgentGA: Evolving Code Solutions in Agent-Seed Space

Framework for evolving code-generation agents by optimizing agent seed (task prompt and parent archives) rather than editing code directly.

Ax Shay Snyder (George Mason University), Andrew Capodieci (Neya Robotics), David Gorsich (US Army Ground Vehicle Systems Center), Maryam Parsa (George Mason University) 5/12/2026

HyperSpace: A Generalized Framework for Spatial Encoding in Hyperdimensional Representations

Open-source framework for Vector Symbolic Architectures with modular operators for encoding, binding, and similarity operations in hyperdimensional spaces.

Ax Yang Li, Zirui Zhang, Yang Liu, Chengzhi Mao 5/12/2026

LACE: Lattice Attention for Cross-thread Exploration

Framework enabling cross-thread attention in parallel reasoning paths for LLMs, allowing concurrent trajectories to interact and share insights.

Ax Carissa Cullen, Harry Garland, Alexander Roman, Louis Thomson, Christos Ziakas, Elliott Thornley 5/12/2026

Towards Shutdownable Agents: Generalizing Stochastic Choice in RL Agents and LLMs

Training approach for shutdownable agents using DReST reward function to make agents stochastically choose different trajectory lengths.

Ax Francisco Angulo de Lafuente, Teerth Sharma, Vladimir Veselov, Seid Mohammed Abdu, Nirmal Tej Kumar, Guillermo Perry 5/12/2026

OpenCLAW-P2P v7.0-P2PCLAW: Resilient Multi-Layer Persistence, Live Reference Verification, and Production-Scale Evaluation of Decentralized AI Peer Review v7.0 -- Mathematical Corrections & Ecosystem Developments Edition

Decentralized platform where autonomous AI agents publish, peer-review, and improve scientific papers without human gatekeeping.

Ax Xiyin Zeng, Yuyu Sun, Haoyang Li, Shouqiang Liu, Hao Wang 5/12/2026

ReCAPA: Hierarchical Predictive Correction to Mitigate Cascading Failures

Hierarchical predictive correction approach for Vision-Language-Action systems to mitigate cascading failures from intermediate step errors.

Ax Bochao Liu, Zhipeng Qian, Yang Zhao, Xinyuan Jiang, Zihan Liang, Yufei Ma, Junpeng Zhuang, Ben Chen, Shuo Yang, Hongen Wan, Yao Wu, Chenyi Lei, Xiao Liang 5/12/2026

Bian Que: An Agentic Framework with Flexible Skill Arrangement for Online System Operations

LLM-based agent framework for large-scale system operations with flexible skill orchestration for monitoring, alerting, and root cause analysis.

Ax Ke Xu, Zhongyuan Lian 5/12/2026

WaferSAGE: Large Language Model-Powered Wafer Defect Analysis via Synthetic Data Generation and Rubric-Guided Reinforcement Learning

LLM-powered framework for wafer defect analysis using synthetic data generation and vision-language models with rubric-guided reinforcement learning.

Ax Yan Zhang, Daiqing Wu, Huawen Shen, Can Ma, Yu Zhou 5/12/2026

Learn where to Click from Yourself: On-Policy Self-Distillation for GUI Grounding

On-policy self-distillation method for GUI grounding in autonomous agents, improving reinforcement learning efficiency for element identification.

Ax Jiyao Wang, Yunbiao Wang, Yubo Jiao, Xiao Yang, Dengbo He, Sasan Jafarnejad, Luis Miranda-Moreno, Raphael Frank, Jiangbo Yu 5/12/2026

MILD: Mediator Agent System with Bidirectional Perception and Multi-Layered Alignment for Human-Vehicle Collaboration

Agent system for human-vehicle collaboration using mediator agents with bidirectional perception to improve coordination and situational awareness.

Ax Maria Chang, Ronny Luss, Miao Liu, Keerthiram Murugesan, Karthikeyan Ramamurthy, Djallel Bouneffouf 5/12/2026

Mitigating Misalignment Contagion by Steering with Implicit Traits

Study of misalignment contagion where misaligned behavior spreads between multiple LMs in multi-agent settings, with steering techniques to mitigate.

Ax Zirui Tang, Xuanhe Zhou, Yumou Liu, Linchun Li, Weizheng Wang, Hongzhang Huang, Jun Zhou, Jiachen Song, Shaoli Yu, Jinqi Wang, Zihang Zhou, Hongyi Zhou, Yuting Lv, Jinyang Li, Jiashuo Liu, Ruoyu Chen, Chunwei Liu, GuoLiang Li, Jihua Kang, Fan Wu 5/12/2026

Workspace-Bench 1.0: Benchmarking AI Agents on Workspace Tasks with Large-Scale File Dependencies

Benchmark for evaluating AI agents on workspace tasks with real-world file dependencies, addressing gap in current agent evaluation methods.

Ax Joseph Breda, Fadi Yousif, Beszel Hawkins, Marinela Cotoi, Miao Liu, Ray Luo, Po-Hsuan Cameron Chen, Mike Schaekermann, Samuel Schmidgall, Xin Liu, Girish Narayanswamy, Samuel Solomon, Maxwell A. Xu, Xiaoran Fan, Longfei Shangguan, Anran Wang, Bhavna Daryani, Buddy Herkenham, Cara Tan, Mark Malhotra, Shwetak Patel, John B. Hernandez, Quang Duong, Yun Liu, Zach Wasson, Dimitrios Antos, Bob Lou, Matthew Thompson, Jonathan Richina, Anupam Pathak, Nichole Young-Lin, Jake Sunshine, Daniel McDuff 5/12/2026

SymptomAI: Toward a Conversational AI Agent for Everyday Symptom Assessment

Conversational AI agents for patient symptom assessment, evaluated on realistic everyday medical scenarios rather than curated case studies.

Ax Tianyu Liu, Wangjie Zheng, Rui Yang, Benny Kai Guo Loo, Hui Zhang, Jeffries Lauran, Jianlei Gu, Botao Yu, Weihao Xuan, Kexin Huang, Nan Liu, James Zou, Yonghui Jiang, Hua Xu, Hongyu Zhao 5/12/2026

A Versatile AI Agent for Rare Disease Diagnosis and Risk Gene Prioritization

Hygieia: multi-modal AI agent integrating phenotypic, genetic, and clinical data for rare disease diagnosis and gene prioritization.

Ax Milena Rmus, Mathew D. Hardy, Thomas L. Griffiths, Mayank Agrawal 5/12/2026

Process Matters more than Output for Distinguishing Humans from Machines

Analysis showing process traces rather than outputs are more reliable for human-machine discrimination in online settings.

Ax Tianle Wang, Zhaoyang Wang, Guangchen Lan, Xinpeng Wei, Sipeng Zhang, Guanwen Qiu, Abulhair Saparov 5/12/2026

Can RL Teach Long-Horizon Reasoning to LLMs? Expressiveness Is Key

Study using synthetic logical reasoning environment to investigate reinforcement learning for improving LLM long-horizon reasoning.

Ax Sixing Chen, Ji-An Li, Saner Cakir, Sinan Akcali, Kayla Lee, Marcelo G. Mattar 5/12/2026

Extracting Search Trees from LLM Reasoning Traces Reveals Myopic Planning

Method to extract and analyze search trees from LLM reasoning traces to evaluate planning behavior and myopia.

Ax Dongming Wu, Junwen Li, Ming Lu, Gang Wang, Ting Chen 5/12/2026

Towards Autonomous Business Intelligence via Data-to-Insight Discovery Agent

AIDA: autonomous agent system for exploring complex enterprise databases and discovering insights from fragmented data.

Ax Xingtong Yu, Zhongwei Kuai, Chang Zhou, Xuanting Xie, Renhe Jiang, Xikun Zhang, Hong Cheng, Xinming Zhang, Yuan Fang 5/12/2026

GraphReAct: Reasoning and Acting for Multi-step Graph Inference

Framework combining reasoning and acting paradigms for multi-step inference over graph-structured data using LLMs.

Ax Daniel Daza, Michael Cochez, Paul Groth 5/12/2026

Inductive Entity Representations from Text via Link Prediction

Knowledge graph entity representation learning from textual descriptions for link prediction tasks.

Ax Jane Downer, Ren Wang, Binghui Wang 5/12/2026

Identifying Backdoored Graphs in Graph Neural Network Training: An Explanation-Based Approach with Novel Metrics

Explanation-based detection method with novel metrics to identify backdoor attacks in graph neural network training.

Ax Suhong Moon, Marwa Abdulhai, Minwoo Kang, Joseph Suh, Widyadewi Soedarmadji, Eran Kohen Behar, David M. Chan, John Canny 5/12/2026

Virtual Personas for Language Models via an Anthology of Backstories

Method to condition large language models with diverse backstories to simulate individual human personas and behaviors.

Ax Daniel Daza, Cuong Xuan Chu, Trung-Kien Tran, Daria Stepanova, Michael Cochez, Paul Groth 5/12/2026

Explaining Graph Neural Networks for Node Similarity on Graphs

Explainability approach for graph neural network-based similarity search on graph data like citation networks.

Ax Haeil Lee, Hansang Lee, Seoyeon Gye, Junmo Kim 5/12/2026

Beta Sampling is All You Need: Efficient Image Generation Strategy for Diffusion Models using Stepwise Spectral Analysis

Beta distribution-based time step sampling method for optimizing diffusion model image generation efficiency.

Ax Chenhao Ding, Xinyuan Gao, Songlin Dong, Jizhou Han, Qiang Wang, Zhengdong Zhou, Yuhang He, Yihong Gong 5/12/2026

LPT: Less-overfitting Prompt Tuning for Vision-Language Model

Prompt tuning method reducing overfitting for vision-language models during transfer to downstream tasks.

Ax Cheng Pan, Kai Junge, Benhui Dai, Qinghua Guan, Josie Hughes 5/12/2026

Language Conditioned Multi-Finger Dexterous Manipulation Enabled by Physical Compliance and Switching of Controllers

Vision-Language-Action model combining language conditioning with dexterous robot manipulation control.

Ax Fanpu Cao, Shu Yang, Zhengjian Chen, Ye Liu, Laizhong Cui 5/12/2026

Ister: Linear Transformer for Efficient Multivariate Time Series Forecasting

Linear transformer architecture with seasonal-trend decomposition for efficient multivariate time series forecasting.

Ax Lingzhi Yuan, Xinfeng Li, Chejian Xu, Guanhong Tao, Xiaojun Jia, Yihao Huang, Wei Dong, Yang Liu, Bo Li 5/12/2026

PromptGuard: Soft Prompt-Guided Unsafe Content Moderation for Text-to-Image Models

Content moderation technique using soft prompts to prevent unsafe content generation in text-to-image models.

Ax Jane Downer, Yingdan Shi, Ziyan Liu, Ren Wang, Binghui Wang 5/12/2026

Watermarking Graph Neural Networks via Explanations for Ownership Protection

Watermarking technique for GNNs using explanation-based methods for intellectual property protection.

Ax Wenbo Zhang, Hengrui Cai, Wenyu Chen 5/12/2026

Beyond the Singular: Revealing the Value of Multiple Generations in Benchmark Evaluation

Methodology for LLM benchmark evaluation using multiple generations to account for inherent model randomness.

Ax Shu Wang, Yixiang Fang, Yingli Zhou, Xilin Liu, Yuchi Ma 5/12/2026

ArchRAG: Attributed Community-based Hierarchical Retrieval-Augmented Generation

Hierarchical retrieval-augmented generation system using attributed community graphs for improved LLM question-answering.

Ax Lior Cohen, Kaixin Wang, Bingyi Kang, Uri Gadot, Shie Mannor 5/12/2026

Simulus: Combining Improvements in Sample-Efficient World Model Agents

Framework combining multiple improvements to world model agents for sample-efficient reinforcement learning.

Ax Shwai He, Weilin Cai, Jiayi Huang, Ang Li 5/12/2026

Capacity-Aware Inference: Mitigating the Straggler Effect in Mixture of Experts

Capacity-aware inference method addressing straggler effects in Mixture of Experts architectures during distributed inference.

Ax Nikhil Abhyankar, Parshin Shojaee, Chandan K. Reddy 5/12/2026

LLM-FE: Automated Feature Engineering for Tabular Data with LLMs as Evolutionary Optimizers

Automated feature engineering system using LLMs as evolutionary optimizers for tabular data machine learning tasks.

Ax Ming Li, Yanhong Li, Ziyue Li, Tianyi Zhou 5/12/2026

How Instruction and Reasoning Data shape Post-Training: Data Quality through the Lens of Layer-wise Gradients

Spectral analysis of layer-wise gradients to evaluate data quality in LLM post-training on instruction and reasoning tasks.

Ax Rui Wang, Junda Wu, Yu Xia, Tong Yu, Ruiyi Zhang, Ryan Rossi, Subrata Mitra, Lina Yao, Julian McAuley 5/12/2026

CachePrune: Teaching LLMs What Not to Follow via KV-Cache Editing

Defense method against indirect prompt injection attacks via KV-cache pruning to prevent unintended instruction-following in LLMs.

Ax Chenxi Jiang, Chuhao Zhou, Jianfei Yang 5/12/2026

REI-Bench: Can Embodied Agents Understand Vague Human Instructions in Task Planning?

Benchmark evaluating LLM-based robot task planners' ability to handle vague human instructions through semantic parsing.

Ax Yue Fan, Xuehai He, Diji Yang, Kaizhi Zheng, Ching-Chen Kuo, Yuting Zheng, Sravana Jyothi Narayanaraju, Xinze Guan, Xin Eric Wang 5/12/2026

GRIT: Teaching MLLMs to Think with Images

Method enabling multimodal LLMs to integrate visual reasoning with image information using reinforcement learning.

Ax Xinran Gu, Kaifeng Lyu, Jiazheng Li, Jingzhao Zhang 5/12/2026

Data Mixing Can Induce Phase Transitions in Knowledge Acquisition

Analysis of phase transitions in LLM knowledge acquisition when training on mixed data sources with varying quality levels.

Ax Liangwei Nathan Zheng, Wei Emma Zhang, Mingyu Guo, Olaf Maennel, Weitong Chen 5/12/2026

Rethinking Gating Mechanism in Sparse MoE: Handling Arbitrary Modality Inputs with Confidence-Guided Gate

Confidence-guided gating mechanism for Sparse MoE architectures to handle missing modalities in multimodal learning.