Ax Zheng Gao, Xiaoyu Li, Xiaoyan Feng, Jiaojiao Jiang, Yang Song, Yulei Sui, Zhenchang Xing, Liming Zhu 19d ago

TRACE: A Two-Channel Robust Attribution Watermark via Complementary Embeddings for LLM-Agent Trajectories

Watermarking scheme for LLM-agent trajectory logs enabling attribution verification against resellers with full data access.

Ax Ali Motahharynia, Mohammadreza Ghaffarzadeh-Esfahani, Mahsa Sheikholeslami, Navid Mazrouei, Matin Irajpour, Yousof Gheisari, Hajar Sirous 19d ago

DrugGen 2: A disease-aware language model for enhancing drug discovery

Disease-aware generative language model for drug discovery that conditions molecular generation on disease ontology and protein sequences.

Ax Shashi Kumar, Yanis Labrak, Hasindri Watawana, Sergio Burdisso, Esa\'u Villatoro-Tello, Kadri Hacio\u{g}lu, Petr Motlicek, Andreas Stolcke 19d ago

When Synthetic Speech Is All You Have: Better Call GRPO

Uses Group Relative Policy Optimization to adapt LLM-based ASR systems trained on synthetic speech for real-world banking domain.

Ax Baoyu Li, Xinchen Yin, Mengying Lin, Yixin Zhang, Danfei Xu 19d ago

EgoWAM: World Action Models Beyond Pixels with In-the-Wild Egocentric Human Data

World Action Models trained on egocentric human video for robot manipulation, separating transferable task semantics from human-specific factors.

Ax Ashit Kumar Subudhi, Bhargav Chirumamilla, Shubham Vaishnav, Mduduzi C. Hlophe, Praveen Kumar Donta, Andrea Fumagalli, Venkateswarlu Gudepu, Koteswararao Kondepu 19d ago

ADORN: Adaptive Drift handling for Open RAN using Reinforcement Learning

Q-learning approach for adaptive model retraining in Open RAN networks to handle traffic-induced performance drift.

Ax Benedikt J. Wagner 19d ago

Two Axes of LLM Abstention: Answer Correctness and Question Answerability

Research on LLM abstention mechanisms distinguishing between incorrect answers and unanswerable questions using separate confidence axes.

Ax Masahiro Fujita 19d ago

The Context Access Divide: Interaction-Level Architecture as a Complementary Dimension of Agentic Inequality

Research paper analyzing interaction-level disparities in AI agent access beyond availability/quality/quantity dimensions.

Ax Feng Wang, Canmiao Fu, Zhipeng Huang, Chen Li, Jing Lyu, Ge Li 19d ago

Cognitive-structured Multimodal Agent for Multimodal Understanding, Generation, and Editing

Multimodal agent with episodic memory for understanding, generation, and editing without context window explosion in long-horizon dialogue.

Ax Zongyou Yang, Yinghan Hou, Xiaokun Yang 19d ago

When the Judge Changes, So Does the Measurement: Auditing LLM-as-Judge Reliability

Audits reliability variation in LLM-as-judge evaluation across model upgrades, showing judge replacements are not interchangeable measurement tools.

Ax Ziqi Chen, Yingli Zhou, Fangyuan Zhang, Quanqing Xu, Chuanhui Yang, Yixiang Fang 19d ago

DocMaster: A Hierarchical Structure-Aware System for Document Analysis

DocMaster hierarchical document analysis system preserving structural relationships for LLM-based analysis of academic papers and technical documents.

Ax ZhiXin Sun 19d ago

VocaDet: Sample-Driven Open-Vocabulary Object Detection and Segmentation via Visual Tokenization and Vector Database Retrieval

VocaDet open-vocabulary object detection and segmentation via visual tokenization and vector database retrieval for scalable category expansion.

Ax Jiahao Wang, Kaizhan Lin, Kaixi Zhang, Jinbo Han, Xingda Wei, Sijie Shen, Chenguang Fang, Wenyuan Yu, Rong Chen, Haibo Chen 19d ago

SMetric: Rethink LLM Scheduling for Serving Agents with Balanced Session-centric Scheduling

SMetric session-centric LLM scheduling system optimized for agentic serving with high KV-cache reuse, balancing throughput and latency.

Ax Weiduo Liao, Yunqiao Yang, Ying Wei 19d ago

When Structured Sparse Autoencoders Learn Consistent Concepts Across Modalities

Structured sparse autoencoders for learning modality-consistent concepts across vision-language models with improved mechanistic interpretability.

Ax Xinlong Zhao, Dongsheng Liu, Hengyu Zhao, Zixuan Fu, Zheng Wang, Jie Cai, Jie Zhou, Qiang Ma, Xuanhe Zhou, Xu Han, Yudong Wang, Zhiyuan Liu 19d ago

UltraX: Refining Pre-Training Data at Scale with Adaptive Programmatic Editing

UltraX adaptive programmatic editing system for large-scale pre-training data refinement beyond rule-based and rule-learning approaches.

Ax Ali Larian, Qian Lin, Chang Zong Wu, Daniel S. Brown 19d ago

Multi-Modal, Multi-Environment Machine Teaching for Robust Reward Learning

Multi-modal machine teaching approach for robust reward learning in autonomous agents across diverse operational environments using inverse RL.

Ax Xiaoshuai Song, Liancheng Zhang, Kangzhi Zhao, Yutao Zhu, Zhongyuan Wang, Guanting Dong, Jinghan Yang, Han Li, Kun Gai, Ji-Rong Wen, Zhicheng Dou 19d ago

WebSwarm: Recursive Multi-Agent Orchestration for Deep-and-Wide Web Search

WebSwarm multi-agent orchestration system for deep-and-wide web search using recursive agent coordination beyond single trajectory limitations.

Ax Guoxuan Xia, Luka Ribar, Paul Balanca 19d ago

A Practical Investigation of Training-free Relaxed Speculative Decoding

Training-free speculative decoding acceleration for LLM sampling with relaxed distribution guarantees enabling speed-capability trade-offs.

Ax QiHong Chen, Aaron Imani, Iftekhar Ahmed 19d ago

ProjAgent: Procedural Similarity Retrieval for Repository-Level Code Generation

ProjAgent retrieves procedurally similar repository functions for repository-level code generation, accounting for cross-file dependencies and project conventions.

Ax Manuel Pita 19d ago

Validity of LLMs as data annotators: AMALIA on authority

Validity assessment of Portugal's AMALIA 9B LLM for data annotation tasks, comparing agreement and reliability against human coders.

Ax David Gonz\'alez-Mart\'inez, Shiwei Liu 19d ago

SLORR: Simple and Efficient In-Training Low-Rank Regularization

In-training low-rank regularization technique for neural network compression without requiring SVD or architecture modification.

Ax Xinyan Chen, Ziyu Guo, Renrui Zhang, Dongzhi Jiang, Hongsheng Li 19d ago

OpenCoF: Learning to Reason Through Video Generation

Chain-of-Frame reasoning approach using video generation models as alternative to chain-of-thought for logical reasoning in large models.

Ax Jinwei He, Feng Lu 19d ago

IFAR: Multi-Perspective and Multi-Level Causal Discovery with LLMs

Multi-perspective causal discovery framework using LLMs for abductive reasoning, includes DeepAbduction dataset for pollution cause analysis.

Ax Mengyang Li, Haozhan Geng, Zhong Zhang, Shuang Liu 19d ago

Dual-Difficulty Curriculum Learning for Direct Preference Optimization

Curriculum learning method for Direct Preference Optimization using two-dimensional difficulty space (prompt complexity and distinguishability) to improve LLM alignment.

Ax Abhijit Chatterjee, Niraj K. Jha, Jonathan D. Cohen, Thomas L. Griffiths, Hongjing Lu, Diana Marculescu, Ashiqur Rasul, Wenrui Xu, Keshab K. Parhi 19d ago

A Vision Toward Energy-Efficient Domain-Specific Artificial Intelligence Models and Agents

Research on energy-efficient domain-specific AI models and agents, addressing computational costs of large language models in production.

Ax Zhiyong Cao, Dunqiang Liu, Qi Dai, Haojun Xu, Huai Yuen Khor, Hao Wang, Huan He, Yafei Liu, Ke Ma, Ruqian Shi, Sicheng Zhou, Sijia Yao 19d ago

SimRPD: Optimizing Recruitment Proactive Dialogue Agents through Simulator-Based Data Evaluation and Selection

System for training task-oriented dialogue agents for recruitment using simulator-based data evaluation and selection.

Ax Nils Einecke 19d ago

Conversational AI for Rapid Scientific Prototyping: A Case Study on ESA's ELOPE Competition

Case study using ChatGPT for rapid scientific prototyping in lunar trajectory estimation competition, achieving second place.

Ax Linghua Zhang, Jun Wang, Jingtong Wu, Zhisong Zhang 19d ago

RetailBench: Evaluating Long-Horizon Autonomous Decision-Making and Strategy Stability of LLM Agents in Realistic Retail Environments

Benchmark for evaluating long-horizon autonomous decision-making and strategy stability of LLM agents in supermarket simulation.

Ax Aditya Agrawal, Alwarappan Nakkiran, Darshan Fofadiya, Alex Karlsson, Harsha Aduri, Aman Singh Thakur 19d ago

Retrieval-Augmented Generation Must Move Beyond Factual Grounding to Represent Diverse Opinions

Position paper arguing RAG systems need redesign to handle opinion-rich content beyond factual grounding.

Ax Jayanta Dey, Shikhar Srivastava, Itamar Lerner, Christopher Kanan, Dhireesha Kudithipudi 19d ago

SHARP: Sleep-based Hierarchical Accelerated Replay for Long Range Non-Stationary Temporal Pattern Recognition

Neural architecture for learning long-range non-stationary temporal patterns in streaming settings without revisiting past data.

Ax Lei (Rachel), Chen, Guilin Zhang, Kai Zhao, Dalmo Cirne, Andy Olsen, Xu Chu, Zeke Miller, Alet Blanken, Amine Anoun, Jerry Ting 19d ago

Deployment-Time Memorization in Foundation-Model Agents

Study of memory design in long-lived foundation model agents, analyzing personalization, extraction risk, and deletion fidelity tradeoffs.

Ax Minh-Ha Nguyen, Erica Gray, Bryce A. Schuler, Chih-Ting Yang, Rizwan Hamid, Lingyao Li, Siyuan Ma, Thomas A. Cassini, Cathy Shyr 19d ago

LiteOdyssey: A Lightweight Reasoning AI Agent for Interpretable Rare-Disease Diagnosis

Lightweight LLM-based agent framework for rare disease diagnosis built through policy iteration with human feedback.

Ax Abdelrahman Sayed Sayed, Pierre-Jean Meyer, Mohamed Ghazel 19d ago

TNODEV: Toolbox for Neural ODE Verification

Tool for formal verification of neural ordinary differential equations in safety-critical applications.

Ax Mingguang Chen, Bo Qu 19d ago

InvestPhilBench: A Multi-Layer Benchmark for Evaluating Large Language Model Procedural Reasoning in Expert Investment Philosophy

Multi-layer benchmark with 118 problems for evaluating LLMs on reconstructing and applying expert investment decision frameworks.

Ax Patrick Cooper, Alvaro Velasquez 19d ago

Narration-of-Thought: Inference-Time Scaffolding for Defeasible Ethical Reasoning in Large Language Models

System prompt technique (narration-of-thought) for improving LLM ethical reasoning on moral dilemmas by reducing stakeholder collapse.

Ax Chiwang Luk, Matin Mohammad Najafi, Zhifeng Jia, Wei Yang, Xiuchang Li, Jinwei Zhu, Yang Ren, Lei Chen, Gao Cong 19d ago

ContextSniper: AntTrail's Token-Efficient Code Memory for Repository-Level Program Repair

Token-efficient context management module for LLM agents performing repository-level program repair with precision evidence selection.

Ax Josh Hills, Ida Caspary, Asa Cooper Stickland 19d ago

Distributed Attacks in Persistent-State AI Control

Analysis of security vulnerabilities in persistent-state AI coding agents that ship code iteratively across sessions.

Ax Hongliang Li, Yijin Liu, Zhiwei Zhang, Zihe Liu, Xinyue Lou, Jinan Xu, Fandong Meng, Kaiyu Huang 19d ago

PolyWorkBench: Benchmarking Multilingual Long-Horizon LLM Agents

Benchmark for evaluating LLM agents across multilingual long-horizon tasks requiring planning, tool use, and environment interaction.

Ax Heting Mao 19d ago

From Application-Layer Simulation to Native Meta-Architecture: Structural Tension as an Endogenous Driver for Heterogeneous AI Evolution

Theoretical framework for embedding cognitive architectures natively into LLMs rather than simulating via prompting and context management.

Ax Benjamin Fresz, Vincent Philipp G\"obels, Safa Omri, Danilo Brajovic, Andreas Aichele, Janika Kutz, Jens Neuh\"uttler, Marco F. Huber 19d ago

The Contribution of XAI for the Safe Development and Certification of AI: An Expert-Based Analysis

Expert-based analysis of explainable AI methods for safe AI development and certification under regulatory frameworks.

Ax Pengcheng Huang, Zhenghao Liu, Yukun Yan, Haiyan Zhao, Xiaoyuan Yi, Hao Chen, Zhiyuan Liu, Maosong Sun, Tong Xiao, Ge Yu, Chenyan Xiong 19d ago

ParamMute: Suppressing Knowledge-Critical FFNs for Faithful Retrieval-Augmented Generation

ParamMute method suppresses knowledge-critical feed-forward networks to improve faithfulness in retrieval-augmented generation systems.

Ax Ruichuan An, Kai Zeng, Ming Lu, Sihan Yang, Renrui Zhang, Huitong Ji, Hao Liang, Wentao Zhang 19d ago

Concept-as-Tree: A Controllable Synthetic Data Framework Makes Stronger Personalized VLMs

Synthetic data generation framework for personalizing vision-language models using concept hierarchies.

Ax Li Qiao, Mahdi Boloursaz Mashhadi, Zhen Gao, Robert Schober, Deniz G\"und\"uz 19d ago

ToDMA: Large Model-Driven Massive Token Communications for Semantic Multiple Access

Large-model-driven semantic multiple access scheme for token-based communications integrating pretrained models.

Ax Gyeongje Cho, Yeonkyoung So, Sangmin Lee, Jaejin Lee 19d ago

Less Is More: Reducing Token Counts Without Compromising Performance

Thunder-Tok tokenizer reduces token fertility for language models while maintaining performance and inference efficiency.

Ax Jiaming Wang, Jizhuo Chen, Diwen Liu, Harold Soh 19d ago

TOPO-Bench: An Open-Source Topological Mapping Evaluation Framework with Quantifiable Perceptual Aliasing

Open-source evaluation framework for topological mapping systems with standardized metrics and perceptual aliasing quantification.

Ax Robin Staab, Jasper Dekoninck, Maximilian Baader, Martin Vechev 19d ago

Adaptive Generation of Bias-Eliciting Questions for LLMs

Adaptive framework for generating bias-eliciting questions to evaluate and expose biases in large language models.

Ax Zhiheng Guo, Zhaoyang Liu, Zihan Cen, Chenyuan Feng, Xinghua Sun, Xiang Chen, Tony Q. S. Quek, Xijun Wang 19d ago

CoCo-Fed: A Unified Framework for Memory- and Communication-Efficient Federated Learning at the Wireless Edge

Federated learning framework reducing memory and communication overhead for edge AI deployment in wireless networks.

Ax Ke Ren, Ali Salamatian, Kieran Pattison, Cyrus Neary 19d ago

V-VLAPS: Value-Guided Planning for Vision-Language-Action Models

Value-guided planning framework for vision-language-action robotic models that improves performance under distribution shift and long-horizon tasks.

Ax Zhuochun Li, Yong Zhang, Ming Li, Yuelyu Ji, Yiming Zeng, Ning Cheng, Yun Zhu, Yanmeng Wang, Shaojun Wang, Jing Xiao, Daqing He 19d ago

Rethinking LLM-as-a-Judge: Representation-as-a-Judge with Small Language Models via Semantic Capacity Asymmetry

Small language models can serve as efficient reference-free evaluators by using internal representations instead of generation.

Ax Moritz Miller, Florent Draye, Bernhard Sch\"olkopf 19d ago

Towards Isolated Interventions via Almost Orthogonal Features in Language Models

Study of feature entanglement in language models and methods for isolated interventions via almost orthogonal feature representations.

Ax Sein Kim, Sangwu Park, Hongseok Kang, Wonjoong Kim, Jimin Seo, Yeonjun In, Kanghoon Yoon, Hyunsik Jeon, Chanyoung Park 19d ago

Self-EvolveRec: Self-Evolving Recommender Systems with LLM-based Directional Feedback

Self-EvolveRec uses LLM-based directional feedback for automated recommender system design in open-ended program spaces.