Ax Ruta Binkyte, Ivaxi Sheth, Zhijing Jin, Mohammad Havaei, Bernhard Sch\"olkopf, Mario Fritz 5/6/2026

Trustworthy AI Suffers from Invariance Conflicts and Causality is The Solution

Position paper arguing causality is necessary to resolve conflicts between trustworthy AI objectives like fairness and robustness.

Ax Xiayang Li, Kuo Gai, Shihua Zhang 5/6/2026

Deciphering Shortcut Learning from an Evolutionary Game Theory Perspective

Theoretical analysis of shortcut learning in deep neural networks using evolutionary game theory framework.

Ax Junjie Yu, Pengrui Lu, Weiye Si, Hongliang Lu, Jiabao Wu, Kaiwen Tao, Kun Wang, Lingyu Yang, Qiran Zhang, Xiuting Guo, Xuanyu Wang, Yang Wang, Yanjie Wang, Yi Yang, Zijian Hu, Ziyi Yang, Zonghan Zhou, Binghao Qiang, Borui Zhang, Chenning Li, Enchang Zhang, Feifan Chen, Feng Jian, Fengyin Sun, Hao Qiu, Hao Zheng, Haoran Zhu, Hongyu Liu, Jianbin Deng, Jiaxin Song, Jiaying Chi, Jiayou Shi, Jie Fang, Jinghui Zhong, Jingyu Zhou, Jinze Li, Junfeng Yi, Junyan Yu, Junzhi Xue, Ni Song, Pengyi Chen, Qi Chen, Quansheng Li, Rui Tao, Shenghai Gong, Shenhang Lu, Tianqi Shen, Tianxiang Zhu, Tiehan Kang, Tingyu Li, Wendi Wu, Xiao Shen, Xiao Zhou, Xiaotao Zhang, Xinrong Li, Xuankun Yang, Xun Zhang, Yan Li, Ye Lu, Yi Wang, Yibo Zhou, Yichi Zhang, Yihao Sun, Yijun Huang, Yixin Zhu, Yixuan Wu, Yuchen Sun, Yue Wu, Yuheng Sun, Yukun Li, Yutian Tu, Yuxuan Qin, Yuzhuo Wu, Zeyu Li, Zhengyu Lou, Zhenning Ran, Zizhu He, Pengfei Liu 5/6/2026

AcademiClaw: When Students Set Challenges for AI Agents

AcademiClaw: bilingual benchmark of 80 complex long-horizon academic tasks from real student workflows for evaluating AI agents.

Ax Majed El Helou, Benjamin Ryder, Chiara Troiani, Jean Diaconu, Herv\'e Muyal, Marcelo Yannuzzi 5/6/2026

Hybrid Inspection and Task-Based Access Control in Zero-Trust Agentic AI

Zero-trust security framework for LLM-driven agents using hybrid inspection and task-based access control for tool invocation.

Ax Gelei Xu, Ningzhi Tang, Xueyang Li, Toby Jia-Jun Li, Zhi Zheng, Wei Jin, Yiyu Shi 5/6/2026

An Empirical Study of Agent Skills for Healthcare: Practice, Gaps, and Governance

Empirical study of 557 healthcare agent skills, analyzing procedural adaptations and governance for healthcare AI agents.

Ax Guangrui Xie 5/6/2026

ORPilot: A Production-Oriented Agentic LLM-for-OR Tool for Optimization Modeling

ORPilot: open-source agentic LLM system that translates business problems into optimization models for production use.

Ax Joshua Strong, Pramit Saha, Emma Sun, Helen Higham, Alison Noble 5/6/2026

Coherent Hierarchical Multi-Label Learning to Defer for Medical Imaging

Learning to defer approach for hierarchical multi-label medical imaging decisions where models can defer to experts.

Ax Fan Ma, Yuntian Liu, Xiang Lan, Weipeng Zhou, Jun Ni, Mauro Giuffr\`e, Lingfei Qian, Xueqing Peng, Yujia Zhou, Ruey-Ling Weng, Huan He, Lu Li, Qingyu Chen, Andrew Loza, Laila Rasmy, Degui Zhi, Yuan Lu, Chenjie Zeng, Joshua C Denny, Lee Schwamm, Daniella Meeker, Lucila Ohno-Machado, Yong Chen, Hua Xu 5/6/2026

Foundation Models to Unlock Real-World Evidence from Nationwide Medical Claims

ReClaim: generative transformer foundation model for extracting insights from large-scale medical claims data.

Ax Maria Chang, Ronny Luss, Miao Lui, Keerthiram Murugesan, Karthikeyan Ramamurthy, Djallel Bouneffouf 5/6/2026

Mitigating Misalignment Contagion by Steering with Implicit Traits

Research on misalignment contagion between multiple LMs in multi-turn interactions and steering techniques using implicit traits.

Ax Christine P Lee, Xinyu Jessica Wang, Aws Albarghouthi, David Porfirio, Bilge Mutlu 5/6/2026

U-Define: Designing User Workflows for Hard and Soft Constraints in LLM-Based Planning

System for designing user workflows with hard and soft constraints in LLM-based planning, addressing reliability and control challenges.

Ax David Mumford 5/6/2026

AIs and Humans with Agency

Philosophical comparison of human agency development with potential LLM agency. Argues for joint action/planning architecture with humans.

Ax Jiujiu Chen, Yazheng Liu, Sihong Xie, Hui Xiong 5/6/2026

SCPRM: A Schema-aware Cumulative Process Reward Model for Knowledge Graph Question Answering

SCPRM: Schema-aware reward model for knowledge graph question answering. Addresses risk compensation in process rewards for LLM reasoning paths.

Ax Ziqi Liu, Kiljae Lee, Yuan Zhang, Weijing Tang 5/6/2026

First-Order Efficiency for Probabilistic Value Estimation via A Statistical Viewpoint

First-order efficiency methods for Shapley/Semivalues computation via statistical viewpoint. Model-agnostic feature attribution for explainability.

Ax Jingze Ge, Yun Liu, Xue Geng, Wanqi Dong, Wang Zhe Mark, Min Wu, Xulei Yang 5/6/2026

Compress Then Adapt? No, Do It Together via Task-aware Union of Subspaces

JACTUS: Joint adaptation and compression for large models across diverse tasks. Simultaneous parameter-efficient fine-tuning and low-rank compression.

Ax Vicente Pelechanoa, Antoni Mestre, Manoli Albert, Miriam Gil 5/6/2026

HAAS: A Policy-Aware Framework for Adaptive Task Allocation Between Humans and Artificial Intelligence Systems

HAAS framework for policy-aware adaptive task allocation between humans and AI. Addresses complementary roles and contextual task distribution.

Ax Mohamad Khajezade, Fatemeh H. Fard, Mohamed Sami Shehata 5/6/2026

Standing on the Shoulders of Giants: Stabilized Knowledge Distillation for Cross--Language Code Clone Detection

Knowledge distillation approach for cross-language code clone detection using compact open-source models. Improves semantic detection reliability.

Ax Abhinav Singh Parmar 5/6/2026

Separating Intelligence from Execution: A Workflow Engine for the Model Context Protocol

MCP Workflow Engine: Orchestration layer decoupling LLM reasoning from execution via Model Context Protocol. Reduces token consumption for repeated tasks.

Ax Shakya Jayakody, Youpeng Zhao, Chinmay Dhanraj Nehate, Jun Wang 5/6/2026

GhostServe: A Lightweight Checkpointing System in the Shadow for Fault-Tolerant LLM Serving

GhostServe: Fault-tolerant checkpointing system for million-token LLM agent serving. Addresses KV cache challenges in long-running inference tasks.

Ax Brice Valentin Kok-Shun, Johnny Chan, Gabrielle Peko, David Sundaram 5/6/2026

Agentopic: A Generative AI Agent Workflow for Explainable Topic Modeling

Agentopic: Multi-agent workflow for explainable topic modeling using LLMs. Collaborative agents handle identification, validation, and hierarchical grouping.

Ax Diana Maria Popa, Simona-Vasilica Oprea, Adela B\^ara 5/6/2026

Generative-AI and the transformation of workforce. A job postings-driven analysis

Analysis of 150,000+ job postings examining generative AI's impact on workforce skill requirements. Tests augmentation vs. substitution across sectors.

Ax Theodor Spiro 5/6/2026

The Oracle's Fingerprint: Correlated AI Forecasting Errors and the Limits of Bias Transmission

Study of correlated forecasting errors across GPT-4o, Claude, and Gemini on 568 binary predictions. Shows epistemic monoculture with mean error correlation r=0.77.

Ax Mengying Wang, Nicolaas Jedema, Rahul Pandey, RaviKiran Krishnan, Jens Lehmann, Yinghui Wu 5/6/2026

Graph Query Generation with Constraint-guided Large Language Agents

UniQGen: LLM agent framework for knowledge graph question answering across RDF/SPARQL and Cypher. Constraint-guided query generation for property graphs.

Ax Cutter Dawes, Aryan Sharma, Angelos Ioannis Lagos, Shivam Raval 5/6/2026

H-Probes: Extracting Hierarchical Structures From Latent Representations of Language Models

H-probes method to extract hierarchical structures from LLM latent representations. Analyzes geometric representations enabling hierarchical reasoning.

Ax Firat Ozdemir, Yun Cheng, Salman Mohebi, Fanny Lehmann, Simon Adamov, Zhenyi Zhang, Leonardo Trentini, Dana Grund, Oliver Fuhrer, Torsten Hoefler, Siddhartha Mishra, Sebastian Schemm, Benedikt Soja, Mathieu Salzmann 5/6/2026

Earth System Foundation Model (ESFM): A unified framework for heterogeneous data integration and forecasting

Open Earth System Foundation Model extending Aurora for weather/climate forecasting. Unified framework for heterogeneous geophysical data integration.

Ax Huanchen Cai, Sten Ternstr\"om 5/6/2026

Voice Mapping of Text-to-Speech Systems: A Metric-Based Approach for Voice Quality Assessment

Evaluation framework for text-to-speech synthesis quality using crest factor, spectrum balance, and cepstral metrics across six TTS models.

Ax Advait Tilak, Jiwon Choi, Nazifa Mouli, Wei Le 5/6/2026

BRITE: A Benchmark for Reliable and Interpretable T2V Evaluation on Implausible Scenarios

BRITE: benchmark for text-to-video evaluation on implausible scenarios including audio-visual alignment and QA-based interpretable assessment.

Ax Alizishaan Khatri, Chiquita Prabhu 5/6/2026

Latent Space Probing for Adult Content Detection in Video Generative Models

Latent space probing framework for detecting adult content in video generative models by analyzing internal representations during generation.

Ax Yida Xue, Ningyu Zhang, Tingwei Wu, Zhe Ma, Daxiong Ji, Zhao Wang, Guozhou Zheng, Huajun Chen 5/6/2026

OceanPile: A Large-Scale Multimodal Ocean Corpus for Foundation Models

OceanPile: multimodal ocean corpus addressing fragmented ocean data for training foundation models on climate and marine biodiversity tasks.

Ax Hao Wang, Limeng Qiao, Chi Zhang, Lin Ma, Guanglu Wan, Xiangyuan Lan, Xiaodan Liang 5/6/2026

X2SAM: Any Segmentation in Images and Videos

X2SAM integrates MLLMs with foundation segmentation models enabling pixel-level perception from conversational instructions across images and videos.

Ax Md. Enamul Hoq, Wataru Uegami, Saghir Alfasly, Ghazal Alabtah, Sahar Rahimi Malakshan, Armita Kazemi, Alex T. Schmitgen, Fred Prior, H. R. Tizhoosh 5/6/2026

Retrieval-Guided Generation for Safer Histopathology Image Captioning

Retrieval-guided generation approach for medical image captioning reducing hallucinations and factual inconsistency in histopathology image descriptions.

Ax Anirudh Iyengar Kaniyar Narayana Iyengar, Tampu Ravi Kumar, Manan Suri, Raviteja Bommireddy, Dinesh Manocha, Puneet Mathur, Vivek Gupta 5/6/2026

DIAGRAMS: A Review Framework for Reasoning-Level Attribution in Diagram QA

DIAGRAMS: review framework for creating reasoning-level attribution in diagram QA, providing structured evidence annotation across visual diagrams and infographics.

Ax Han Gong, Zhen Zhou, Yunyang Shi, Yan Tan, Jinbiao Huo, Qi Hong, Zhiyuan Liu 5/6/2026

TRIP-Evaluate: An Open Multimodal Benchmark for Evaluating Large Models in Transportation

TRIP-Evaluate: open multimodal benchmark for assessing LLMs and MLLMs on transportation tasks including regulation QA, traffic management, and autonomous driving reasoning.

Ax Bla\v{z} Bertalani\v{c}, Carolina Fortuna 5/6/2026

The Cost of Consensus: Isolated Self-Correction Prevails Over Unguided Homogeneous Multi-Agent Debate

Empirical study of multi-agent LLM debate showing isolated self-correction outperforms unguided homogeneous debate across teams of 10 models on high-difficulty benchmarks.

Ax Katja Karhu, Kari Smolander, Jussi Kasurinen 5/6/2026

To Vibe Research or Not to Vibe Research? Generative AI in Qualitative Research

Survey of generative AI use in qualitative research, discussing suitability for different research approaches (small-q positivist vs Big Q non-positivist) in software engineering.

Ax Guantian Zheng 5/6/2026

StyleShield: Exposing the Fragility of AIGC Detectors through Continuous Controllable Style Transfer

StyleShield: demonstrates fragility of AIGC detectors through continuous controllable style transfer, exposing reliability paradox as language models improve.

Ax Daniel Song, Peter Ney, Cristina Menghini, Faizan Ahmad, Aidan Boyd, Nathaniel Li, Ziwen Han, Jean-Christophe Testud, Saisuke Okabayashi, Maeve Ryan, Jinpeng Miao, Hamza Kwisaba, Felix Binder, Spencer Whitman, Jim Gust, Esteban Arcaute, Dhaval Kapil, Jacob Kahn, Ayaz Minhas, Tristan Goodman, Lauren Deason, Alexander Vaughan, Shengjia Zhao, Summer Yue 5/6/2026

Code World Model Preparedness Report

Code World Model preparedness report: Meta's code generation and reasoning model assessed for frontier AI risks; found no additional catastrophic risks beyond current ecosystem.

Ax Anton Kolonin 5/6/2026

Interpretable experiential learning based on state history and global feedback

Interpretable experiential learning model based on state history and global feedback for reinforcement learning in resource-constrained environments, evaluated on Atari.

Ax Zelin Guan, Shengda Zhuo, Zeyan Li, Jinchun He, Wangjie Qiu, Zhiming Zheng, Shuqiang Huang 5/6/2026

E-MIA: Exam-Style Black-Box Membership Inference Attacks against RAG Systems

E-MIA: black-box membership inference attack against RAG systems that infers whether documents are in retrieval corpus by analyzing LLM response interactions.

Ax Daan Di Scala, Maaike de Boer, P{\i}nar Yolum 5/6/2026

"I Don't Know" -- Towards Appropriate Trust with Certainty-Aware Retrieval Augmented Generation

Certainty-aware RAG system enabling LLMs to express appropriate confidence and say 'I don't know' to improve user trust.

Ax Zi Tian, Guanting Shen 5/6/2026

Ablation Study of Multimodal Perception, Language Grounding, and Control for Human-Robot Interaction in an Object Detection and Grasping Task

Ablation study isolating contributions of LLM, vision perception, and control components in human-robot interaction for object detection.

Ax Lennard C. Froma, Tom Kouwenhoven, Maaike H. T. de Boer, Catholijn M. Jonker, Max J. van Duijn 5/6/2026

Seeking Information with RAG-Assistants: Does Model Size Matter in Human-AI Collaborations?

Evaluation of retrieval-augmented generation chatbots in realistic multi-turn information-seeking workflows with compliance constraints.

Ax Chenyu Zhang, Xinchen Lyu, Chenshan Ren, Shuhan Liu, Qimei Cui 5/6/2026

Adaptive 3D-RoPE: Physics-Aligned Rotary Positional Encoding for Wireless Foundation Models

Physics-aligned rotary positional encoding method for wireless foundation models in channel state information tasks.

Ax Harshit Rajgarhia, Shuubham Ojha, Asif Shaik, Akhil Pothanapalli, Rachuri Lokesh, Abhishek Mukherji, Prasanna Desikan 5/6/2026

MedMosaic: A Challenging Large Scale Benchmark of Diverse Medical Audio

Medical audio QA dataset with diverse clinical scenarios for benchmarking language and audio reasoning models.

Ax Nihanth W. Cherukuru, Matt Rehme, Kirsten J. Mayer, David John Gagne, John Schreck, John Clyne, Charlie Becker 5/6/2026

Toward a Scientific Discovery Engine for Weather and Climate Data: A Visual Analytics Workbench for Embedding-Based Exploration

Visual analytics workbench for exploring weather and climate data using embedding-based representations and similarity search.

Ax Mohammed Abu Baker, Luca Baroni, Dan Wilhelm 5/6/2026

Model Organisms Are Leaky: Perplexity Differencing Often Reveals Finetuning Objectives

Method using perplexity differencing to reveal finetuning objectives in LLMs, enabling detection of intentionally introduced behaviors.

Ax Kevin H. Guo, Chao Yan, Avinash Baidya, Katherine Brown, Xiang Goa, Juming Xiong, Zhijun Yin, Bradley A. Malin 5/6/2026

CLEAR: Revealing How Noise and Ambiguity Degrade Reliability in LLMs for Medicine

Framework assessing how ambiguity and uncertainty in real-world medical queries degrade LLM reasoning compared to simplified benchmarks.

Ax Yueru Sun, Yimeng Zhang, Haoyu Gu, Nuo Chen, Dong She, Xianrong Yao, Yang Gao, Zhanpeng Jin 5/6/2026

EmoMM: Benchmarking and Steering MLLM for Multimodal Emotion Recognition under Conflict and Missingness

Benchmark and investigation of multimodal LLM behavior for emotion recognition under modality conflict and missingness conditions.

Ax Alan L. McCann 5/6/2026

Certified Purity for Cognitive Workflow Executors: From Static Analysis to Cryptographic Attestation

Certified purity architecture converting governance enforcement into structural boundaries for cognitive workflow execution systems.

Ax Iman Sharifi, Hyeong Tae Kim, Maheed Hatem Ahmed, Mahsa Ghasemi, Peng Wei 5/6/2026

Separation Assurance between Heterogeneous Fleets of Small Unmanned Aerial Systems via Multi-Agent Reinforcement Learning

Multi-agent reinforcement learning approach for tactical deconfliction and separation assurance between heterogeneous unmanned aerial systems.

Ax Joseph Spracklen, Pedram Aghazadeh, Farinaz Koushanfar, Murtuza Jadliwala 5/6/2026

LLM Ghostbusters: Surgical Hallucination Suppression via Adaptive Unlearning

Unlearning method to suppress hallucinations in LLMs, specifically addressing supply-chain attacks from fictional package recommendations.