Ax Jianze Wang, Ying Liu, Jinlong Chen, Xuchun Hu, Qilong Zhang, Yu Cao, Jun Wang, Hua Yang, Yong Xie, Qianglong Chen 5/6/2026

MAD-OPD: Breaking the Ceiling in On-Policy Distillation via Multi-Agent Debate

MAD-OPD uses multi-agent debate to overcome single-teacher capability ceiling in on-policy distillation for agentic tasks.

Ax Jinyuan Feng, Xin Yu, Yiqun Chen, Xiaochi Wei, Yan Gao, Yi Wu, Yao Hu, Zhiqiang Pu 5/6/2026

Focus on the Core: Empowering Diffusion Large Language Models by Self-Contrast

Self-contrast decoding strategy for diffusion large language models to improve generation quality via high-information-density context modeling.

Ax Yifei Wang, Ruiyin Li, Peng Liang, Yangxiao Cai, Zengyang Li, Mojtaba Shahin, Arif Ali Khan, Qiong Feng 5/6/2026

Using LLMs in Software Design: An Empirical Study of GitHub and A Practitioner Survey

Empirical study of how developers use LLMs in software design tasks, surveying practitioners on benefits and drawbacks.

Ax Dong Xu, Jialun Cao, Guozhao Mo, Junjie Hu, Cheng Wen, Hongyu Lin, Xianpei Han, Shengchao Qin, Cong Tian, Shing-Chi Cheung, Le Sun, Yaojie Lu 5/6/2026

LiveFMBench: Unveiling the Power and Limits of Agentic Workflows in Specification Generation

LiveFMBench: systematic study of LLM and agent-based formal specification generation for C programs with contamination awareness.

Ax Sangkwon Park, Donghun Kang, Jisoo Mok, Sungroh Yoon 5/6/2026

Verbal-R3: Verbal Reranker as the Missing Bridge between Retrieval and Reasoning

Verbal-R3 bridges retrieval and LLM reasoning via verbal annotations—analytic narratives connecting queries to retrieved contexts in RAG systems.

Ax Annie Yuan, Xiaohua Chen, Kalina Yacef, Judy Kay 5/6/2026

AI Expert Twin: Capturing Expert Cognition for Human-Centred, Practice-Based Learning

Framework for capturing expert cognition in practice-based domains to enhance AI-driven educational systems.

Ax Benjamin Warner, Ratna Sagari Grandhi, Max Kieffer, Aymane Ouraq, Saurav Panigrahi, Geetu Ambwani, Kunal Bagga, Nikhil Khandekar, Arya Hariharan, Nishant Mishra, Manish Ram, Shamus Sim Zi Yang, Ahmed Essouaied, Adepoju Jeremiah Moyondafoluwa, Robert Scholz, Bofeng Huang, Molly Beavers, Srishti Gureja, Anish Mahishi, Sameed Khan, Maxime Griot, Hunar Batra, Jean-Benoit Delbrouck, Siddhant Bharadwaj, Ronald Clark, Ashish Vashist, Anas Zafar, Leema Krishna Murali, Harsh Deshpande, Ameen Patel, William Brown, Johannes Hagemann, Connor Lane, Paul Steven Scotti, Tanishq Mathew Abraham 5/6/2026

Medmarks: A Comprehensive Open-Source LLM Benchmark Suite for Medical Tasks

Medmarks: open-source benchmark suite with 30 benchmarks for evaluating LLMs on medical tasks including QA, information extraction, and calculations.

Ax Junkun Jiao, Tong Liu, Ke Li, Weimin Song, Yipu Liao, Bolun Zhang, Beijiang Liu, Chang-Zheng Yuan, Yue Sun 5/6/2026

HepScript: A Dual-Use DSL for Human-AI Collaborative Data Analysis Workflows in High-Energy Physics

HepScript DSL enabling human-AI collaborative data analysis for high-energy physics workflows using agentic LLMs and domain-specific knowledge.

Ax Richeng Zhou, Xuelin Zhang, Liyuan Liu 5/6/2026

Quantifying Multimodal Capabilities: Formal Generalization Guarantees in Pairwise Metric Learning

Theoretical analysis of generalization properties in multimodal metric learning with incomplete or redundant data.

Ax Pang Liu, Yingjie Lao 5/6/2026

VisInject: Disruption != Injection -- A Dual-Dimension Evaluation of Universal Adversarial Attacks on Vision-Language Models

Analysis of adversarial attacks on vision-language models, distinguishing between output perturbation and precise injection attacks.

Ax Hyukjoo Lee 5/6/2026

Practical Limits of Autonomous Test Repair: A Multi-Agent Case Study with LLM-Driven Discovery and Self-Correction

Multi-agent autonomous testing system using LLM with LangGraph orchestration for UI test repair in enterprise applications.

Ax Zebin Guo, Weidong Geng, Ruichen Mao 5/6/2026

FT-RAG: A Fine-grained Retrieval-Augmented Generation Framework for Complex Table Reasoning

FT-RAG: fine-grained retrieval-augmented generation framework for complex table reasoning using entry-level decomposition and semantic comprehension.

Ax Zixi Shao, Tao Wang, Yibei Xiao, Tianyi Huang 5/6/2026

Protein-Conditioned Multi-Objective Reinforcement Learning for Full-Length mRNA Design

ProMORNA: multi-objective reinforcement learning framework for designing full-length therapeutic mRNA from protein sequences balancing stability and safety.

Ax Paul Garnier, Vincent Lannelongue, Elie Hachem 5/6/2026

Mesh Based Simulations with Spatial and Temporal awareness

Advances machine learning surrogates for CFD using GNNs and transformers, addressing training paradigm bottlenecks in physics simulations.

Ax Mohamed Amine Ferrag, Abderrahmane Lakas, Merouane Debbah 5/6/2026

6G Needs Agents: Toward Agentic AI-Native Networks for Autonomous Intelligence

Proposes agentic AI-native 6G networks where LLM-based agents operate as bounded, policy-governed entities for autonomous intelligence.

Ax Arnau Marin-Llobet, Javier Ferrando 5/6/2026

Automated Interpretability and Feature Discovery in Language Models with Agents

Autonomous multiagent framework automating mechanistic interpretability of LLMs through iterative hypothesis testing and feature discovery agents.

Ax Ahmed Ibrahim 5/6/2026

Neuro-Symbolic Agents for Hallucination-Free Requirements Reuse

Neuro-symbolic agents framework combining LLMs with formal methods to enable flexible requirements reuse while maintaining structural validity and consistency.

Ax Donato Crisostomi 5/6/2026

Model Merging: Foundations and Algorithms

Thesis on model merging paradigm: combining independently trained neural networks in weight space without optimization or original training data access.

Ax Ngoc Luyen Le, Marie-H\'el\`ene Abel, Bertrand Laforge 5/6/2026

KG-First, LLM-Fallback: A Hybrid Microservice for Grounded Skill Search and Explanation

SkillGraph-Service: hybrid microservice merging knowledge graphs with LLM fallback for competency framework interoperability and skill search.

Ax Zhuoyun Li, Boxuan Wang, Jinwei Hu, Zhenglin Huang, Qisong He, Xinmiao Huang, Guangliang Cheng, Xiaowei Huang, Yi Dong 5/6/2026

Where Do Prompt Perturbations Break Generation? A Segment-Level View of Robustness in LoRA-Tuned Language Models

Introduces S²R², a segment-level robustness framework for LoRA-tuned language models that enforces consistency at entity and relation levels beyond whole-sequence.

Ax Pratyush Acharya, Nuraj Rimal, Habish Dhakal 5/6/2026

Concepts Whisper While Syntax Shouts: Spectral Anti-Concentration and the Dual Geometry of Transformer Representations

Tests causal inner products for cross-lingual concept transport in transformer representations across 17 models, finding anti-concentration phenomena.

Ax Eunchae Jang, S. Shyam Sundar 5/6/2026

Less Interaction But More Explanation: A Communication Perspective on Agentic AI Interfaces

Examines how agentic AI interfaces should prioritize explanation and oversight communication over routine user interaction during workflow execution.

Ax Roseval Malaquias Junior, Giovana Kerche Bon\'as, Thales Sales Almeida, Hugo Abonizio, Thiago Laitz, Ramon Pires, Marcos Piau, Celio Larcher, Rodrigo Nogueira 5/6/2026

Prosa: Rubric-Based Evaluation of LLMs on Real User Chats in Brazilian Portuguese

Introduces Prosa, a Brazilian Portuguese LLM evaluation benchmark using rubric-based scoring with multi-judge filtering to reduce judge model bias.

Ax Rohit Agarwal, Joshua Lin, Mark Braverman, Elad Hazan 5/6/2026

AI Alignment via Incentives and Correction

Studies AI alignment through law-and-economics models of deterrence, analyzing how agentic AI systems respond strategically to incentive structures.

Ax Tingting Dan, Guorong Wu 5/6/2026

From Cortical Synchronous Rhythm to Brain Inspired Learning Mechanism: An Oscillatory Spiking Neural Network with Time-Delayed Coordination

Brain-inspired spiking neural network with oscillatory dynamics and time-delayed coordination for cognitive learning mechanisms.

Ax Pritam Mishra, Coloma Ballester, Dimosthenis Karatzas 5/6/2026

TRIMMER: A New Paradigm for Video Summarization through Self-Supervised Reinforcement Learning

TRIMMER uses self-supervised reinforcement learning for domain-agnostic video summarization without manual annotations.

Ax Haoshen Zhang, Di Wen, Kunyu Peng, David Schneider, Zeyun Zhong, Alexander Jaus, Zdravko Marinov, Jiale Wei, Ruiping Liu, Junwei Zheng, Yufan Chen, Yufeng Zhang, Yuanhao Luo, Lei Qi, Rainer Stiefelhagen 5/6/2026

IMPACT-HOI: Supervisory Control for Onset-Anchored Partial HOI Event Construction

IMPACT-HOI mixed-initiative framework for egocentric video annotation of Human-Object Interactions to generate robot manipulation training data.

Ax Qian Yin, Di Wen, Kunyu Peng, David Schneider, Zeyun Zhong, Alexander Jaus, Zdravko Marinov, Jiale Wei, Ruiping Liu, Junwei Zheng, Yufan Chen, Chen Zhang, Lei Qi, Rainer Stiefelhagen 5/6/2026

IMPACT-Scribe: Interactive Temporal Action Segmentation with Boundary Scribbles and Query Planning

IMPACT-Scribe framework for dense video annotation using interactive correction-driven collaboration between human annotators and models.

Ax Qiao Liu 5/6/2026

Missingness-aware Data Imputation via AI-powered Bayesian Generative Modeling

MissBGM Bayesian generative model for missing data imputation with uncertainty quantification using neural networks and Bayesian inference.

Ax Joydeb Kumar Sana 5/6/2026

Class-Aware Adaptive Differential Privacy in Deep Learning for Sensor-Based Fall Detection

Adaptive differential privacy method for fall detection in sensor data that applies class-aware noise to preserve prediction performance.

Ax Yushi Sun, Bowen Cao, Dong Fang, Lingfeng Su, Wai Lam 5/6/2026

GRAVITY: Architecture-Agnostic Structured Anchoring for Long-Horizon Conversational Memory

GRAVITY architecture for long-horizon conversational agents that structures retrieved memory fragments with relational and temporal anchors.

Ax Sylvain Hellin, Suhyung Jang, Stefan Fuchs, Stavros Nousias, Andr\'e Borrmann 5/6/2026

BIM Information Extraction Through LLM-based Adaptive Exploration

LLM-based adaptive agent for extracting information from BIM models through iterative exploration instead of static query translation.

Ax Anamika Paul Rupa, Anietie Andy 5/6/2026

Probe-Geometry Alignment: Erasing the Cross-Sequence Memorization Signature Below Chance

Technique to surgically remove memorization signatures in unlearned LLMs using cross-sequence probes without capability loss.

Ax Yipin Guo, Siddharth Joshi 5/6/2026

SplitZip: Ultra Fast Lossless KV Compression for Disaggregated LLM Serving

SplitZip compression method for KV cache transfer in disaggregated LLM serving, optimizing prefill-decode architecture bottleneck.

Ax Xinran Li, Xinze Che, Yifan Lyu, Zhiqi Huang, Xiujuan Xu 5/6/2026

TCDA: Thread-Constrained Discourse-Aware Modeling for Conversational Sentiment Quadruple Analysis

TCDA model for conversational sentiment analysis using graph networks and positional encoding to handle multi-turn dialogues with temporal context.

Ax Zeshang Li, Shuoyang Zhang, Jiashen Ding 5/6/2026

GEASS: Training-Free Caption Steering for Hallucination Mitigation in Vision-Language Models

Training-free method to mitigate object hallucination in vision-language models by steering caption generation.

Ax Alfredo Metere 5/6/2026

Architectural Obsolescence of Unhardened Agentic-AI Runtimes

Security analysis of agentic-AI runtimes showing critical gaps in action auditing and safety properties across tool calls and device actuation.

Ax Yiheng Yao, Chelsea Zou, Robert D. Hawkins 5/6/2026

Talk is Cheap, Communication is Hard: Dynamic Grounding Failures and Repair in Multi-Agent Negotiation

Research on dynamic grounding failures and repair mechanisms in multi-agent LLM negotiation across conversational turns.

Ax Kwan Soo Shin 5/6/2026

The Compliance Gap: Why AI Systems Promise to Follow Process Instructions but Don't

Study of the Compliance Gap where AI agents verbally agree to instructions but fail to follow them, examining the disconnect between stated and actual behavior.

Ax Zhenkang Hou, Wenqiang Pu, Junkun Yan, Rui Zhou, Hongwei Liu 5/6/2026

Discover Fast Power Allocation Solution for Multi-Target Tracking via AlphaEvolve Evolution

LLM-guided evolutionary search (AlphaEvolve) to discover closed-form solutions for radar power allocation in multi-target tracking.

Ax Rudray Dave, Vedang Dubey, Smit Deoghare, Sudhakar Mishra 5/6/2026

Selector-Guided Autonomous Curriculum for One-Shot Reinforcement Learning from Verifiable Rewards

Selector-guided curriculum learning approach for one-shot reinforcement learning from verifiable rewards in LLM math reasoning.

Ax Yangyang Zhou, Yi-Chen Li 5/6/2026

RMGAP: Benchmarking the Generalization of Reward Models across Diverse Preferences

Benchmark (RMGAP) for evaluating reward model generalization across diverse user preferences in RLHF-aligned language models.

Ax Szymon Kobus, Deniz G\"und\"uz 5/6/2026

Remote Action Generation: Remote Control with Minimal Communication

Framework for remote control with communication-constrained channels where actors receive action guidance from a centralized controller.

Ax Zhiyang Dai, Yansong Gao, Boyu Kuang, Haodong Li, Qi Chang, Gaurav Varshney, Derek Abbott, Anmin Fu 5/6/2026

Repurposing and Evaluating the (In)Feasibility of Dataset Poisoning enabled Watermarking for Contrastive Learning

Evaluation of data-poisoning backdoor attacks on contrastive learning models and their generalization across datasets.

Ax Kotaro Furuya, Takahito Tanimura 5/6/2026

Spatiotemporal Hidden-State Dynamics as a Signature of Internal Reasoning in Large Language Models

Analysis of hidden-state dynamics in large reasoning models to determine if extended reasoning traces reflect genuine internal computation or verbosity.

Ax Dahyun Oh, Minhyuk Yoon, H. Jin Kim 5/6/2026

Quality-Aware Exploration Budget Allocation for Cooperative Multi-Agent Reinforcement Learning

Research on exploration budget allocation in cooperative multi-agent reinforcement learning using intrinsic motivation and novelty bonuses.

Ax Kumar Shubham, Pavan Karjol, Kiran M K, Prathosh AP 5/6/2026

Leveraging Data Symmetries to Select an Optimal Subset of Training Data under Label Noise

Method for selecting optimal training subsets under label noise by leveraging data symmetries, detecting low-quality examples via k-NN based analysis.

Ax Hongkun Pan, Yuwei Wu, Wanyi Hong, Shenghui Hu, Qitong Yan, Yi Yang, Rufei Han, Changju Zhou, Minfeng Zhu, Dongming Han, Wei Chen 5/6/2026

Chart-FR1: Visual Focus-Driven Fine-Grained Reasoning on Dense Charts

Chart-FR1: Multimodal LLM benchmark for fine-grained reasoning on high-information-density charts with multiple subplots and dense annotations.

Ax Kyle Lee, Corentin Delacour, Kevin Callahan-Coray, Kyle Jiang, Can Yaras, Samet Oymak, Tathagata Srimani, Kerem Y. Camsari 5/6/2026

Stochastic Sparse Attention for Memory-Bound Inference

SANTA: Sparse attention method for memory-bound LLM inference that samples from post-softmax distribution to reduce KV cache bandwidth at long contexts.

Ax Sadia Asif, Mohammad Mohammadi Amiri 5/6/2026

RefusalGuard: Geometry-Preserving Fine-Tuning for Safety in LLMs

RefusalGuard: Geometry-preserving fine-tuning method maintaining safety alignment in LLMs by analyzing activation-space feature changes during task-specific adaptation.