Ax Vaibhav Devraj, Dhruv Kumar, Jagat Sesh Challa, Parth Agarwal, Navya Kommuri, Trizal Garg, Prisha Singhal, Dhruv Shah 2/24/2026

CricBench: A Multilingual Benchmark for Evaluating LLMs in Cricket Analytics

CricBench multilingual benchmark for evaluating LLMs on cricket analytics tasks including text-to-SQL and complex statistical reasoning.

Ax Rui Li, Zhaoning Zhang, Libo Zhang, Huaimin Wang, Xiang Fu, Zhiquan Lai 2/24/2026

Nightjar: Dynamic Adaptive Speculative Decoding for Large Language Models Serving

Nightjar dynamic speculative decoding adapts draft token verification to varying request loads for improved LLM serving efficiency.

Ax Chenji Lu, Zhuo Chen, Hui Zhao, Zhenyi Wang, Pengjie Wang, Chuan Yu, Jian Xu 2/24/2026

RAIR: A Rule-Aware Benchmark Uniting Challenging Long-Tail and Visual Salience Subset for E-commerce Relevance Assessment

RAIR Chinese benchmark dataset for e-commerce search relevance assessment addressing long-tail and visual salience in LLM evaluation.

Ax Tianyu Zhao, Llion Jones 2/24/2026

Fast-weight Product Key Memory

FwPKM sparse memory layer balances storage capacity and efficiency for sequence models, addressing softmax attention trade-offs.

Ax Xinhao Sun, Huaijin Zhao, Maoliang Li, Zihao Zheng, Jiayu Chen, Yun Liang, Xiang Chen 2/24/2026

STaRR: Spatial-Temporal Token-Dynamics-Aware Responsive Remasking for Diffusion Language Models

STaRR training-free remasking strategy for diffusion language models adapts to spatial-temporal token dynamics for faster inference.

Ax Hector Zenil 2/24/2026

On the Limits of Self-Improving in Large Language Models: The Singularity Is Not Near Without Symbolic Model Synthesis

Formal analysis proving LLM recursive self-training undergoes degenerative dynamics without external grounding, limiting self-improvement scaling.

Ax Bertie Vidgen, Austin Mann, Abby Fennelly, John Wright Stanly, Lucas Rothman, Marco Burstein, Julien Benchek, David Ostrofsky, Anirudh Ravichandran, Debnil Sur, Neel Venugopal, Alannah Hsia, Isaac Robinson, Calix Huang, Olivia Varones, Daniyal Khan, Michael Haines, Austin Bridges, Jesse Boyle, Koby Twist, Zach Richards, Chirag Mahapatra, Brendan Foody, Osvald Nitski 2/24/2026

APEX-Agents

APEX-Agents benchmark evaluates AI agents on long-horizon cross-application tasks in realistic work environments with files and tools.

Ax Saptarshi Roy, Alessandro Rinaldo, Purnamrita Sarkar 2/24/2026

Low-Dimensional Adaptation of Rectified Flow: A Diffusion and Stochastic Localization Perspective

Theoretical analysis of how rectified flow automatically adapts to intrinsic dimensionality to accelerate sampling in diffusion models.

Ax Onkar Susladkar, Tushar Prakash, Adheesh Juvekar, Kiet A. Nguyen, Dong-Hwan Jang, Inderjit S Dhillon, Ismini Lourentzou 2/24/2026

PyraTok: Language-Aligned Pyramidal Tokenizer for Video Understanding and Generation

PyraTok pyramidal tokenizer learns language-aligned discrete video representations across multiple scales for improved video understanding and generation.

Ax Xiaojiang Peng, Jingyi Chen, Zebang Cheng, Bao Peng, Fengyi Wu, Yifei Dong, Shuyuan Tu, Qiyu Hu, Huiting Huang, Yuxiang Lin, Jun-Yan He, Kai Wang, Zheng Lian, Zhi-Qi Cheng 2/24/2026

Emotion-LLaMAv2 and MMEVerse: A New Framework and Benchmark for Multimodal Emotion Understanding

Emotion-LLaMAv2 and MMEVerse benchmark for evaluating multimodal LLMs on emotion understanding tasks with high-quality annotations.

Ax Zihou Zhang, Zheyong Xie, Li Zhong, Haifeng Liu, Yao Hu, Shaosheng Cao 2/24/2026

One Token Is Enough: Improving Diffusion Language Models with a Sink Token

Sink token mechanism stabilizes diffusion language models, improving parallel text generation quality and addressing moving sink phenomenon.

Ax Ziyue Wang, Jiangshan Yu, Kaihua Qin, Dawn Song, Arthur Gervais, Liyi Zhou 2/24/2026

TxRay: Agentic Postmortem of Live Blockchain Attacks

TxRay uses agentic approaches for postmortem analysis of blockchain DeFi exploits, analyzing attack patterns and vulnerabilities.

Ax Soyeon Hong, Jinchan Kim, Jaegook You, Seungtaek Choi, Suha Kwak, Hyunsouk Cho 2/24/2026

TextME: Bridging Unseen Modalities Through Text Descriptions

TextME framework projects diverse modalities into LLM embeddings using only text descriptions, enabling multimodal expansion without paired datasets.

Ax Kevin Han, Yuhang Zhou, Mingze Gao, Gedi Zhou, Serena Li, Abhishek Kumar, Xiangjun Fan, Weiwei Li, Lizhu Zhang 2/24/2026

EBPO: Empirical Bayes Shrinkage for Stabilizing Group-Relative Policy Optimization

EBPO technique using empirical Bayes shrinkage to stabilize Group Relative Policy Optimization for LLM reasoning with verifiable rewards.

Ax Chenhang Cui, Binyun Yang, Fei Shen, Yuxin Chen, Jingnan Zheng, Xiang Wang, An Zhang, Tat-Seng Chua 2/24/2026

Transport and Merge: Cross-Architecture Merging for Large Language Models

Method for merging LLMs across different architectures to transfer knowledge from large models to smaller ones with different designs.

Ax Mingfu Liang, Yufei Li, Jay Xu, Kavosh Asadi, Xi Liu, Shuo Gu, Kaushik Rangadurai, Frank Shyu, Shuaiwen Wang, Song Yang, Zhijing Li, Jiang Liu, Mengying Sun, Fei Tian, Xiaohan Wei, Chonglin Sun, Jacob Tao, Shike Mei, Wenlin Chen, Santanu Kolay, Sandeep Pandey, Hamed Firooz, Luke Simon 2/24/2026

Generative Reasoning Re-ranker

LLM-based reranking approach for recommendation systems using generative reasoning to refine candidate rankings with few-shot learning.

Ax Paul Saegert, Ullrich K\"othe 2/24/2026

Breaking the Simplification Bottleneck in Amortized Neural Symbolic Regression

Amortized neural symbolic regression method addressing expression normalization bottleneck to improve scalability for discovering analytical expressions from data.

Ax Yusheng Zheng, Jiakun Fan, Quanzhi Fu, Yiwei Yang, Wei Zhang, Andi Quinn 2/24/2026

AgentCgroup: Understanding and Controlling OS Resources of AI Agents

Systematic characterization of OS-level resource dynamics in sandboxed AI coding agents with resource control mechanisms.

Ax Ailin Huang, Ang Li, Aobo Kong, Bin Wang, Binxing Jiao, Bo Dong, Bojun Wang, Boyu Chen, Brian Li, Buyun Ma, Chang Su, Changxin Miao, Changyi Wan, Chao Lou, Chen Hu, Chen Xu, Chenfeng Yu, Chengting Feng, Chengyuan Yao, Chunrui Han, Dan Ma, Dapeng Shi, Daxin Jiang, Dehua Ma, Deshan Sun, Di Qi, Enle Liu, Fajie Zhang, Fanqi Wan, Guanzhe Huang, Gulin Yan, Guoliang Cao, Guopeng Li, Han Cheng, Hangyu Guo, Hanshan Zhang, Hao Nie, Haonan Jia, Haoran Lv, Hebin Zhou, Hekun Lv, Heng Wang, Heung-Yeung Shum, Hongbo Huang, Hongbo Peng, Hongyu Zhou, Hongyuan Wang, Houyong Chen, Huangxi Zhu, Huimin Wu, Huiyong Guo, Jia Wang, Jian Zhou, Jianjian Sun, Jiaoren Wu, Jiaran Zhang, Jiashu Lv, Jiashuo Liu, Jiayi Fu, Jiayu Liu, Jie Cheng, Jie Luo, Jie Yang, Jie Zhou, Jieyi Hou, Jing Bai, Jingcheng Hu, Jingjing Xie, Jingwei Wu, Jingyang Zhang, Jishi Zhou, Junfeng Liu, Junzhe Lin, Ka Man Lo, Kai Liang, Kaibo Liu, Kaijun Tan, Kaiwen Yan, Kaixiang Li, Kang An, Kangheng Lin, Lei Yang, Liang Lv, Liang Zhao, Liangyu Chen, Lieyu Shi, Liguo Tan, Lin Lin, Lina Chen, Luck Ma, Mengqiang Ren, Michael Li, Ming Li, Mingliang Li, Mingming Zhang, Mingrui Chen, Mitt Huang, Na Wang, Peng Liu, Qi Han, Qian Zhao, Qinglin He, Qinxin Du, Qiuping Wu, Quan Sun, Rongqiu Yang, Ruihang Miao, Ruixin Han, Ruosi Wan, Ruyan Guo, Shan Wang, Shaoliang Pang, Shaowen Yang, Shengjie Fan, Shijie Shang, Shiliang Yang, Shiwei Li, Shuangshuang Tian, Siqi Liu, Siye Wu, Siyu Chen, Song Yuan, Tiancheng Cao, Tianchi Yue, Tianhao Cheng, Tianning Li, Tingdan Luo, Wang You, Wei Ji, Wei Yuan, Wei Zhang, Weibo Wu, Weihao Xie, Wen Sun, Wenjin Deng, Wenzhen Zheng, Wuxun Xie, Xiangfeng Wang, Xiangwen Kong, Xiangyu Liu, Xiangyu Zhang, Xiaobo Yang, Xiaojia Liu, Xiaolan Yuan, Xiaoran Jiao, Xiaoxiao Ren, Xiaoyun Zhang, Xin Li, Xin Liu, Xin Wu, Xing Chen, Xingping Yang, Xinran Wang, Xu Zhao, Xuan He, Xuanti Feng, Xuedan Cai, Xuqiang Zhou, Yanbo Yu, Yang Li, Yang Xu, Yanlin Lai, Yanming Xu, Yaoyu Wang, Yeqing Shen, Yibo Zhu, Yichen Lv, Yicheng Cao, Yifeng Gong, Yijing Yang, Yikun Yang, Yin Zhao, Yingxiu Zhao, Yinmin Zhang, Yitong Zhang, Yixuan Zhang, Yiyang Chen, Yongchi Zhao, Yongshen Long, Yongyao Wang, Yousong Guan, Yu Zhou, Yuang Peng, Yuanhao Ding, Yuantao Fan, Yuanwei Lu, Yuanzhen Yang, Yuchu Luo, Yudi Zhao, Yue Peng, Yueqiang Lin, Yufan Lu, Yuling Zhao, Yunzhou Ju, Yurong Zhang, Yusheng Li, Yuxiang Yang, Yuyang Chen, Yuzhu Cai, Zejia Weng, Zetao Hong, Zexi Li, Zhe Xie, Zheng Ge, Zheng Gong, Zheng Zeng, Zhenyi Lu, Zhewei Huang, Zhichao Chang, Zhiguo Huang, Zhiheng Hu, Zidong Yang, Zili Wang, Ziqi Ren, Zixin Zhang, Zixuan Wang 2/24/2026

Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters

Step 3.5 Flash: sparse MoE model with 11B active parameters for frontier-level agentic reasoning and tool execution.

Ax Rushuai Yang, Hecheng Wang, Chiming Liu, Xiaohan Yan, Yunlong Wang, Xuan Du, Shuoyu Yue, Yongcheng Liu, Chuheng Zhang, Lizhe Qi, Yi Chen, Wei Shan, Maoqing Yao 2/24/2026

ALOE: Action-Level Off-Policy Evaluation for Vision-Language-Action Model Post-Training

Action-level off-policy evaluation method for improving vision-language-action models through online RL in robotics.

Ax Stephan Vonschallen, Rahel H\"ausler, Theresa Schmiedel, Friederike Eyssel 2/24/2026

Never say never: Exploring the effects of available knowledge on agent persuasiveness in controlled physiotherapy motivation dialogues

Study of generative social agents in healthcare motivation dialogues examining effects of knowledge availability on persuasiveness.

Ax Jiahao Lu, Ziwei Xu, William Tjhi, Junnan Li, Antoine Bosselut, Pang Wei Koh, Mohan Kankanhalli 2/24/2026

Buy versus Build an LLM: A Decision Framework for Governments

Strategic decision framework for governments choosing between buying, building, or hybrid approaches to LLM deployment.

Ax Md. Najib Hasan, Mst. Jannatun Ferdous Rain, Fyad Mohammed, Nazmul Siddique 2/24/2026

BETA-Labeling for Multilingual Dataset Construction in Low-Resource IR

BETA-labeling framework using multiple diverse LLM annotators to construct multilingual low-resource IR datasets.

Ax Xingjian Wu, Xvyuan Liu, Junkai Lu, Siyuan Wang, Xiangfei Qiu, Yang Shu, Jilin Hu, Chenjuan Guo, Bin Yang 2/24/2026

ST-EVO: Towards Generative Spatio-Temporal Evolution of Multi-Agent Communication Topologies

Framework for self-evolving multi-agent systems to dynamically construct task-adaptive communication topologies for LLM-powered collaboration.

Ax Shiqi Liu, Zeyu He, Guojian Zhan, Letian Tao, Zhilong Zheng, Jiang Wu, Yinuo Wang, Yang Guan, Kehua Sheng, Bo Zhang, Keqiang Li, Jingliang Duan, Shengbo Eben Li 2/24/2026

STAPO: Stabilizing Reinforcement Learning for LLMs by Silencing Rare Spurious Tokens

Stability improvement for RL fine-tuning of LLMs by filtering rare spurious tokens to prevent training collapse.

Ax Samira Nazari, Mohammad Saeed Almasi, Mahdi Taheri, Ali Azarpeyvand, Ali Mokhtari, Ali Mahani, Christian Herglotz 2/24/2026

HAWX: A Hardware-Aware FrameWork for Fast and Scalable ApproXimation of DNNs

Hardware-aware framework for DNN approximation using multi-level sensitivity scoring and heterogeneous approximate computing blocks.

Ax Zun Li, John Schultz, Daniel Hennes, Marc Lanctot 2/24/2026

Discovering Multiagent Learning Algorithms with Large Language Models

Using large language models to automatically discover multi-agent reinforcement learning algorithms for imperfect-information games.

Ax Ziyuan Liu, Shizhao Sun, Danqing Huang, Yingdong Shi, Meisheng Zhang, Ji Li, Jingsong Yu, Jiang Bian 2/24/2026

DesignAsCode: Bridging Structural Editability and Visual Fidelity in Graphic Design Generation

Graphic design generation system balancing visual fidelity with structural editability through layered approach.

Ax Guoqi Yu, Juncheng Wang, Chen Yang, Jing Qin, Angelica I. Aviles-Rivero, Shujun Wang 2/24/2026

Decentralized Attention Fails Centralized Signals: Rethinking Transformers for Medical Time Series

Transformer architecture analysis for medical time series data addressing temporal and channel dependencies in EEG/ECG.

Ax Kehao Zhang, Shangtong Gui, Sheng Yang, Wei Chen, Yang Feng 2/24/2026

Learning to Remember: End-to-End Training of Memory Agents for Long-Context Reasoning

Unified Memory Agent framework using reinforcement learning to enable LLMs to track state and aggregate evidence in long-context reasoning tasks.

Ax Attila Dobi, Aravindh Manickavasagam, Benjamin Thompson, Xiaohan Yang, Faisal Farooq 2/24/2026

Measuring the Prevalence of Policy Violating Content with ML Assisted Sampling and LLM Labeling

Design-based measurement system using ML-assisted sampling and LLM labeling for cost-effective policy violation prevalence estimation.

Ax Abeer Alsheikhi, Amirfarhad Farhadi, Azadeh Zamanifar 2/24/2026

Deep Reinforcement Learning for Optimizing Energy Consumption in Smart Grid Systems

Deep reinforcement learning approach for optimal power flow problems in smart grids addressing sample inefficiency.

Ax Sheng Ran 2/24/2026

Learning Beyond Optimization: Stress-Gated Dynamical Regime Regulation in Autonomous Systems

Proposes stress-gated dynamical regime regulation as alternative to optimization-based learning for autonomous systems operating without fixed objectives.

Ax Guanghui Min, Tianhao Huang, Ke Wan, Chen Chen 2/24/2026

GIST: Targeted Data Selection for Instruction Tuning via Coupled Optimization Geometry

GIST algorithm for efficient instruction tuning via targeted data selection using coupled optimization geometry and optimizer statistics.

Ax Afiya Ayman, Ayan Mukhopadhyay, Aron Laszka 2/24/2026

Ensemble Prediction of Task Affinity for Efficient Multi-Task Learning

Ensemble method for predicting task affinity in multi-task learning to identify which task groups benefit from joint training.

Ax Ziqiao Shang, Lingyue Ge, Yang Chen, Shi-Yu Tian, Zhenyu Huang, Wenbo Fu, Yu-Feng Li, Lan-Zhe Guo 2/24/2026

MapTab: Can MLLMs Master Constrained Route Planning?

MapTab benchmark evaluating multimodal LLM reasoning capabilities on constrained route planning tasks with visual grounding requirements.

Ax Baris Arat, Emre Sefer 2/24/2026

Diagnosing LLM Reranker Behavior Under Fixed Evidence Pools

Diagnostic framework isolating LLM reranker behavior using fixed evidence pools to evaluate ranking policy independent of retrieval quality.

Ax Sarim Chaudhry 2/24/2026

Non-Interfering Weight Fields: Treating Model Parameters as a Continuously Extensible Function

Research on non-interfering weight fields for LLMs to address catastrophic forgetting by treating model parameters as continuously extensible functions rather than fixed weights.

Ax Leonardo F. Toso, Davit Shadunts, Yunyang Lu, Nihal Sharma, Donglin Zhan, Nam H. Nguyen, James Anderson 2/24/2026

Learning Invariant Visual Representations for Planning with Joint-Embedding Predictive World Models

World model using joint-embedding predictive architectures with invariant representations for robust planning.

Ax Shvat Messica, Jiawen Zhang, Kevin Li, Theodoros Tsiligkaridis, Marinka Zitnik 2/24/2026

Adaptive Time Series Reasoning via Segment Selection

Time series reasoning system using segment selection and LLMs to adaptively analyze relevant portions of sequences.

Ax Gabriel Raya, Bac Nguyen, Georgios Batzolis, Yuhta Takida, Dejan Stancevic, Naoki Murata, Chieh-Hsin Lai, Yuki Mitsufuji, Luca Ambrogioni 2/24/2026

Information-Guided Noise Allocation for Efficient Diffusion Training

Information-theoretic noise schedule allocation for efficient diffusion model training across datasets and resolutions.

Ax Yongzhong Xu 2/24/2026

Global Low-Rank, Local Full-Rank: The Holographic Encoding of Learned Algorithms

Investigates grokking phenomenon showing how transformers develop low-dimensional algorithms with high-dimensional parameters.

Ax Yinan Zou, Md Kamran Chowdhury Shisher, Christopher G. Brinton, Vishrant Tripathi 2/24/2026

Communication-Efficient Personalized Adaptation via Federated-Local Model Merging

Federated learning method for parameter-efficient model adaptation using LoRA merging with theoretical guarantees.

Ax Nikolaos Kougioulis, Nikolaos Gkorgkolis, MingXue Wang, Bora Caglayan, Dario Simionato, Andrea Tonon, Ioannis Tsamardinos 2/24/2026

Large Causal Models for Temporal Causal Discovery

Large causal models framework for temporal causal discovery enabling multi-dataset pretraining.

Ax Anthony Bao, Jeffrey Lai, William Gilpin 2/24/2026

Transformers for dynamical systems learn transfer operators in-context

Studies how transformers learn transfer operators for dynamical systems, enabling zero-shot cross-scale generalization.

Ax Dhruvesh Patel, Benjamin Rozonoyer, Gaurav Pandey, Tahira Naseem, Ram\'on Fernandez Astudillo, Andrew McCallum 2/24/2026

Insertion Based Sequence Generation with Learnable Order Dynamics

Insertion-based sequence generation with learnable order dynamics using discrete flow matching.

Ax Trishita Tiwari, Ari Trachtenberg, G. Edward Suh 2/24/2026

Prior Aware Memorization: An Efficient Metric for Distinguishing Memorization from Generalization in Large Language Models

Metric distinguishing genuine memorization from generalization in LLMs to assess training data leakage risk.

Ax Zhixiang Guo, Siyuan Liang, Andras Balogh, Noah Lunberry, Rong-Cheng Tu, Mark Jelasity, Dacheng Tao 2/24/2026

When World Models Dream Wrong: Physical-Conditioned Adversarial Attacks against World Models

White-box adversarial attack on world models via physical condition perturbations for autonomous driving systems.

Ax Ziyan Zhang, Changxin Wan, Peng Hao, Kanok Boriboonsomsin, Matthew J. Barth, Yongkang Liu, Seyhan Ucar, Guoyuan Wu 2/24/2026

HONEST-CAV: Hierarchical Optimization of Network Signals and Trajectories for Connected and Automated Vehicles with Multi-Agent Reinforcement Learning

Hierarchical multi-agent reinforcement learning framework for traffic signal control and vehicle eco-driving optimization.

Ax Osman Onur Kuzucu, Tunca Do\u{g}an 2/24/2026

GLaDiGAtor: Language-Model-Augmented Multi-Relation Graph Learning for Predicting Disease-Gene Associations

Graph neural network augmented with language models for predicting disease-gene associations in biomedical research.