Ax Jun Liu, Zhenglun Kong, Peiyan Dong, Changdi Yang, Tianqi Li, Hao Tang, Geng Yuan, Wei Niu, Wenbin Zhang, Pu Zhao, Xue Lin, Dong Huang, Yanzhi Wang 3/13/2026

Structured Agent Distillation for Large Language Model

Structured Agent Distillation compresses LLM-based agents into smaller student models while preserving reasoning and action consistency.

Ax Chengyu Shen, Zhen Hao Wong, Runming He, Hao Liang, Meiyi Qiang, Zimo Meng, Zhengyang Zhao, Bohan Zeng, Zhengzhou Zhu, Bin Cui, Wentao Zhang 3/13/2026

Let's Verify Math Questions Step by Step

Framework for verifying correctness of math questions used in LLM training. Focuses on QA data quality beyond answer correctness.

Ax Kai Li, Can Shen, Yile Liu, Jirui Han, Kelong Zheng, Xuechao Zou, Lionel Z. Wang, Shun Zhang, Xingjian Du, Hanjun Luo, Yingbin Jin, Xinxin Xing, Ziyang Ma, Yue Liu, Yifan Zhang, Junfeng Fang, Kun Wang, Yibo Yan, Gelei Deng, Haoyang Li, Yiming Li, Xiaobin Zhuang, Tianlong Chen, Qingsong Wen, Tianwei Zhang, Yang Liu, Haibo Hu, Zhizheng Wu, Xiaolin Hu, Eng-Siong Chng, Wenyuan Xu, XiaoFeng Wang, Wei Dong, Xinfeng Li 3/13/2026

AudioTrust: Benchmarking the Multifaceted Trustworthiness of Audio Large Language Models

AudioTrust benchmark evaluating trustworthiness of audio LLMs. Reveals vulnerabilities from non-semantic acoustic cues like timbre and accent.

Ax Jianing Geng, Biao Yi, Zekun Fei, Ruiqi He, Lihai Nie, Tong Li, Zheli Liu 3/13/2026

Hiding in Plain Sight: A Steganographic Approach to Stealthy LLM Jailbreaks

Steganographic jailbreak attacks on LLMs balancing semantic and linguistic stealth. Bypasses safety mechanisms through hidden malicious intent.

Ax Sicheng Feng, Song Wang, Shuyi Ouyang, Lingdong Kong, Zikai Song, Jianke Zhu, Huan Wang, Xinchao Wang 3/13/2026

ReasonMap: Towards Fine-Grained Visual Reasoning from Transit Maps

ReasonMap benchmark for evaluating multimodal LLM visual reasoning on transit maps. Tests math and logic capabilities on 1,008 questions.

Ax Aida Kostikova, Zhipin Wang, Deidamea Bajri, Ole P\"utz, Benjamin Paa{\ss}en, Steffen Eger 3/13/2026

LLLMs: A Data-Driven Survey of Evolving Research on Limitations of Large Language Models

Data-driven survey of 14,648 papers on LLM limitations from 2022-2025. Systematically categorizes known weaknesses and failure modes.

Ax Sirui Lu, Zhijing Jin, Terry Jingchen Zhang, Pavel Kos, J. Ignacio Cirac, Bernhard Sch\"olkopf 3/13/2026

Can Theoretical Physics Research Benefit from Language Agents?

Investigates LLM limitations in theoretical physics. Identifies gaps in physical intuition and constraint satisfaction beyond prompting improvements.

Ax Nadav Kunievsky, James A. Evans 3/13/2026

Measuring Intent Comprehension in LLMs

Study measuring how well LLMs comprehend user intent beyond surface-level text matching. Analyzes gap between token prediction and actual user goals.

Ax Peibo Li, Shuang Ao, Hao Xue, Yang Song, Maarten de Rijke, Johan Barth\'elemy, Tomasz Bednarz, Flora D. Salim 3/13/2026

Refine-POI: Reinforcement Fine-Tuned Large Language Models for Next Point-of-Interest Recommendation

Refine-POI applies reinforcement fine-tuning to LLMs for point-of-interest recommendation with improved semantic ID indexing and topology awareness.

Ax Ondrej Bohdal, Mete Ozay, Jijoong Moon, Kyeng-Hun Lee, Hyeonmok Ko, Umberto Michieli 3/13/2026

Efficient Compositional Multi-tasking for On-device Large Language Models

Research on adapter parameters and task merging for efficient multi-task learning in on-device LLMs, enabling multiple tasks via parameter merging.

Ax Zhejun Zhao, Yuchen Li, Alley Liu, Yuehu Dong, Xiaolong Wei, Lixue Zheng, Pingsheng Liu, Dongdong Shen, Long Xia, Jiashu Zhao, Dawei Yin 3/13/2026

TURA: Tool-Augmented Unified Retrieval Agent for AI Search

TURA proposes a tool-augmented retrieval agent for conversational AI search that handles real-time data and structured queries beyond traditional RAG limitations.

Ax Maolin Sun, Yibiao Yang, Yuming Zhou 3/13/2026

Once4All: Skeleton-Guided SMT Solver Fuzzing with LLM-Synthesized Generators

Once4All uses LLM-synthesized test generators guided by skeleton templates to fuzz SMT solvers and uncover correctness bugs.

Ax Wei-Teng Chu, Tianyi Zhang, Matthew Johnson-Roberson, Weiming Zhi 3/13/2026

Efficient Construction of Implicit Surface Models From a Single Image for Motion Generation

Fast Image-to-Neural Surface constructs implicit distance representations from single images for robotics obstacle avoidance and path planning.

Ax Haoyang Zheng, Xinyang Liu, Cindy Xiangrui Kong, Nan Jiang, Zheyuan Hu, Weijian Luo, Wei Deng, Guang Lin 3/13/2026

Ultra-Fast Language Generation via Discrete Diffusion Divergence Instruct

DiDi-Instruct distills fast student models from diffusion LLMs for ultra-fast language generation matching teacher performance.

Ax Songmei Yu, Andrew Zagula 3/13/2026

TRACE: AI-Assisted Assessment of Collaborative Projects in Computer Science Education

TRACE uses AI for semi-automated assessment of individual contributions in collaborative computer science group projects.

Ax Yeonseo Lee, Jungwook Mun, Hyosup Shin, Guebin Hwang, Junhee Nam, Taeyeop Lee, Sungho Jo 3/13/2026

XGrasp: Gripper-Aware Grasp Detection with Multi-Gripper Data Generation

XGrasp detects robotic grasps that generalize across multiple gripper types without retraining using gripper-aware architecture.

Ax Jingyu Song, Zhenxin Li, Shiyi Lan, Xinglong Sun, Nadine Chang, Maying Shen, Joshua Chen, Katherine A. Skinner, Jose M. Alvarez 3/13/2026

DriveCritic: Towards Context-Aware, Human-Aligned Evaluation for Autonomous Driving with Vision-Language Models

DriveCritic framework uses vision-language models to provide context-aware evaluation of autonomous driving planners aligned with human judgment.

Ax Zhangquan Chen, Manyuan Zhang, Xinlei Yu, Xufang Luo, Mingze Sun, Zihao Pan, Xiang An, Yan Feng, Peng Pei, Xunliang Cai, Ruqi Huang 3/13/2026

Think with 3D: Geometric Imagination Grounded Spatial Reasoning from Limited Views

Vision-language model approach for 3D spatial reasoning from limited views using geometric imagination grounding.

Ax Eric Bigelow, Daniel Wurgaft, YingQiao Wang, Noah Goodman, Tomer Ullman, Hidenori Tanaka, Ekdeep Singh Lubana 3/13/2026

Belief Dynamics Reveal the Dual Nature of In-Context Learning and Activation Steering

Unifying framework explaining in-context learning and activation steering through belief dynamics, treating both as instances of broader control mechanism.

Ax Xin Sun, Daniel St{\aa}hl, Kristian Sandahl, Christoph Kessler 3/13/2026

Quality Assurance of LLM-generated Code: Addressing Non-Functional Quality Characteristics

Study evaluates non-functional quality characteristics of LLM-generated code using ISO/IEC 25010 model across functional correctness, maintainability, and security.

Ax Junbo Zou, Haotian Xia, Zhen Ye, Shengjie Zhang, Christopher Lai, Vicente Ordonez, Weining Shen, Hanjie Chen 3/13/2026

DeepSport: A Multimodal Large Language Model for Comprehensive Sports Video Reasoning via Agentic Reinforcement Learning

DeepSport is an end-to-end trained multimodal LLM for multi-sport video understanding using agentic reinforcement learning for iterative reasoning.

Ax Seyed Mohssen Ghafari, Ronny Kol, Juan C. Quiroz, Nella Luan, Monika Patial, Chanaka Rupasinghe, Herman Wandabwa, Luiz Pizzato 3/13/2026

ConCISE: A Reference-Free Conciseness Evaluation Metric for LLM-Generated Answers

ConCISE is a reference-free evaluation metric for measuring conciseness of LLM-generated responses to reduce verbosity and token costs.

Ax Chunzheng Zhu, Yangfang Lin, Shen Chen, Yijun Wang, Jianxin Lin 3/13/2026

MedEyes: Learning Dynamic Visual Focus for Medical Progressive Diagnosis

MedEyes applies vision-language models with reinforcement learning for medical diagnosis via dynamic visual focusing and iterative clinical reasoning.

Ax Khushboo Thaker, Yony Bresler 3/13/2026

Knowledge Distillation with Structured Chain-of-Thought for Text-to-SQL

Knowledge distillation method using structured chain-of-thought to improve text-to-SQL performance in small language models while reducing cost and security risks.

Ax Zhongyu Xia, Wenhao Chen, Yongtao Wang, Ming-Hsuan Yang 3/13/2026

KnowVal: A Knowledge-Augmented and Value-Guided Autonomous Driving System

KnowVal combines visual-language reasoning, driving knowledge, and value alignment for autonomous driving using knowledge-augmented approaches beyond data-driven learning.

Ax Alistair Plum, Laura Bernardy, Tharindu Ranasinghe 3/13/2026

Do LLMs Judge Distantly Supervised Named Entity Labels Well? Constructing the JudgeWEL Dataset

Dataset (judgeWEL) for named entity recognition in Luxembourgish using LLMs to verify automatically labeled data for underrepresented languages.

Ax Pan Liao, Feng Yang, Di Wu, Jinwen Yu, Yuhua Zhu, Wenhui Zhao, Dingwen Zhang 3/13/2026

LLMTrack: Semantic Multi-Object Tracking with Multi-modal Large Language Models

Benchmark and framework (Grand-SMOT) for semantic multi-object tracking using multimodal LLMs to handle complex relational queries.

Ax Yicheng Di 3/13/2026

LLM-driven Multimodal Recommendation

arXiv: LMMRec framework using LLMs to model user motivations in multimodal recommendation systems from heterogeneous information sources.

Ax Xinwu Ye, Yicheng Mao, Jia Zhang, Yimeng Liu, Li Hao, Fang Wu, Zhiwei Li, Yuxuan Liao, Zehong Wang, Zhiyuan Liu, Zhenfei Yin, Li Yuan, Philip Torr, Huan Sun, Xiangxiang Zeng, Mengdi Wang, Le Cong, Shenghua Gao, Xiangru Tang 3/13/2026

LatentChem: From Textual CoT to Latent Thinking in Chemical Reasoning

arXiv: LatentChem decouples chemical reasoning from natural language by using latent representations instead of chain-of-thought prompting.

Ax Zhuoxu Huang, Mengxi Jia, Hao Sun, Xuelong Li, Jungong Han 3/13/2026

Controllable Exploration in Hybrid-Policy RLVR for Multi-Modal Reasoning

arXiv: Hybrid-policy reinforcement learning approach for enhancing multi-modal LLM reasoning with controlled exploration strategies.

Ax Rafael R. Baptista, Andr\'e de Lima Salgado, Ricardo V. Godoy, Marcelo Becker, Thiago Boaventura, Gustavo J. G. Lahr 3/13/2026

Evaluating Zero-Shot and One-Shot Adaptation of Small Language Models in Leader-Follower Interaction

arXiv: Evaluating small language models for zero-shot and one-shot role classification in robot leader-follower interaction tasks.

Ax Jose Javier Gonzalez Ortiz, Abhay Gupta, Christopher Rinard, Davis Blalock 3/13/2026

FlashOptim: Optimizers for Memory-Efficient Training

arXiv: FlashOptim memory-efficient optimizers for mixed-precision neural network training reducing parameter storage overhead.

Ax Wei Luo, Yangfan Ou, Jin Deng, Zeshuai Deng, Xiquan Yan, Zhiquan Wen, Mingkui Tan 3/13/2026

ProtoDCS: Towards Robust and Efficient Open-Set Test-Time Adaptation for Vision-Language Models

arXiv: ProtoDCS framework for robust test-time adaptation of Vision-Language Models under distribution shift and open-set scenarios.

Ax Weizhi Feng, Shidong Shen, Jiaxiang Liu, Taolue Chen, Fu Song, Zhilin Wu 3/13/2026

Can LLM Aid in Solving Constraints with Inductive Definitions?

arXiv: Using LLMs with structured prompts to generate auxiliary lemmas for constraint solving with inductive definitions.

Ax John Driscoll, Yulin Chen, Viki Shi, Izak Vucharatavintara, Yaxing Yao, Haojian Jin 3/13/2026

Understanding Parents' Desires in Moderating Children's Interactions with GenAI Chatbots through LLM-Generated Probes

Study of parental moderation preferences for children's GenAI chatbot interactions using LLM-generated probe scenarios.

Ax Hengquan Guo 3/13/2026

BLOCK: An Open-Source Bi-Stage MLLM Character-to-Skin Pipeline for Minecraft

BLOCK: open-source two-stage MLLM pipeline generating Minecraft character skins from text descriptions via 3D preview synthesis.

Ax Junjie Chu, Xinyue Shen, Ye Leng, Michael Backes, Yun Shen, Yang Zhang 3/13/2026

Benchmark of Benchmarks: Unpacking Influence and Code Repository Quality in LLM Safety Benchmarks

Multi-dimensional evaluation of LLM safety benchmarks assessing their academic influence and code repository quality.

Ax Siddharth Boppana, Annabel Ma, Max Loeffler, Raphael Sarfati, Eric Bigelow, Atticus Geiger, Owen Lewis, Jack Merullo 3/13/2026

Reasoning Theater: Disentangling Model Beliefs from Chain-of-Thought

Analysis of performative chain-of-thought in reasoning LLMs showing discrepancy between generated explanations and internal model beliefs.

Ax Runlong Ye, Naaz Sibia, Angela Zavaleta Bernuy, Tingting Zhu, Carolina Nobre, Viktoria Pammer-Schindler, Michael Liut 3/13/2026

From Toil to Thought: Designing for Strategic Exploration and Responsible AI in Systematic Literature Reviews

Design study on LLM-assisted tools for systematic literature reviews to reduce cognitive load and enable iterative exploration.

Ax Hadi Sotoudeh, Payel Mukhopadhyay, Ruben Ohana, Michael McCabe, Neil D. Lawrence, Shirley Ho, Miles Cranmer 3/13/2026

On the Value of Tokeniser Pretraining in Physics Foundation Models

Investigation of tokenizer pretraining impact on physics foundation models for emulating complex multiphysics simulations.

Ax Qian Da, Yijiang Chen, Min Ju, Zheyi Ji, Albert Zhou, Wenwen Wang, Matthew A Abikenari, Philip Chikontwe, Guillaume Larghero, Bowen Chen, Peter Neidlinger, Dingrong Zhong, Shuhao Wang, Wei Xu, Drew Williamson, German Corredor, Sen Yang, Le Lu, Xiao Han, Kun-Hsing Yu, Jun-zhou Huang, Laura Barisoni, Geert Litjens, Anant Madabhushi, Lifeng Zhu, Chaofu Wang, Junhan Zhao, Weiguo Hu 3/13/2026

Computational Pathology in the Era of Emerging Foundation and Agentic AI -- International Expert Perspectives on Clinical Integration and Translational Readiness

Expert perspectives on integrating foundation models and AI agents into clinical computational pathology with translational readiness assessment.

Ax William Thorne, Joseph James, Yang Wang, Chenghua Lin, Diana Maynard 3/13/2026

Evaluating LLM-Based Grant Proposal Review via Structured Perturbations

Framework using structured perturbations to evaluate LLM performance on high-stakes grant proposal reviews across quality dimensions.

Ax Chengjun Yu, Xuhan Zhu, Chaoqun Du, Pengfei Yu, Wei Zhai, Yang Cao, Zheng-Jun Zha 3/13/2026

EXPLORE-Bench: Egocentric Scene Prediction with Long-Horizon Reasoning

EXPLORE-Bench benchmark evaluating multimodal LLMs' ability to reason about long-horizon physical consequences from egocentric viewpoints.

Ax Omar Elshehy, Omer Nacar, Abdelbasset Djamai, Muhammed Ragab, Khloud Al Jallad, Mona Abdelazim 3/13/2026

AraModernBERT: Transtokenized Initialization and Long-Context Encoder Modeling for Arabic

AraModernBERT adapter for Arabic NLP with transtokenized embeddings and 8K token context window support.

Ax Yannis Karmim (ALMAnaCH), Renato Pino (UCHILE), Hernan Contreras (UCHILE), Hernan Lira (CENIA), Sebastian Cifuentes (CENIA), Simon Escoffier (PUC), Luis Mart\'i (ALMAnaCH), Djam\'e Seddah (ALMAnaCH), Valentin Barri\`ere (UCHILE, CENIA) 3/13/2026

Leveraging Wikidata for Geographically Informed Sociocultural Bias Dataset Creation: Application to Latin America

Dataset creation using Wikidata to detect cultural biases in LLMs across Latin American languages and contexts.

Ax Youness Dkhissi (LIUM), Valentin Vielzeuf (LIUM), Elys Allesiardo (LIUM), Anthony Larcher (LIUM) 3/13/2026

SENS-ASR: Semantic Embedding injection in Neural-transducer for Streaming Automatic Speech Recognition

Semantic embedding injection in neural transducers for low-latency streaming automatic speech recognition.

Ax Moriba Kemessia Jah 3/13/2026

The Epistemic Support-Point Filter: Jaynesian Maximum Entropy Meets Popperian Falsification

Epistemic Support-Point Filter for recursive estimation using maximum entropy and falsification principles.

Ax Maxwell Miller-Golub, Collin Coil, Kamil Faber, Marcin Pietron, Panpan Zheng, Pasquale Minervini, Roberto Corizzo 3/13/2026

Rethinking the Harmonic Loss via Non-Euclidean Distance Layers

Proposes harmonic loss as alternative to cross-entropy for training deep neural networks with improved interpretability.

Ax Anna Chistyakova, Mikhail Pautov 3/13/2026

Contract And Conquer: How to Provably Compute Adversarial Examples for a Black-Box Model?

Method to provably compute adversarial examples for black-box neural networks using Contract And Conquer approach.

Ax Artem Dvirniak, Evgeny Kushnir, Dmitrii Tarasov, Artem Iudin, Oleg Kiriukhin, Mikhail Pautov, Dmitrii Korzh, Oleg Y. Rogov 3/13/2026

Towards Robust Speech Deepfake Detection via Human-Inspired Reasoning

Human-inspired reasoning approach for robust speech deepfake detection with improved generalization to unseen audio domains and interpretability.