Ax Sepehr Harfi, Ahmad Salimi, Dongming Shen, Alex Smola 5/12/2026

ProactBench: Beyond What The User Asked For

ProactBench evaluates LLM conversational proactivity: ability to infer and act on implied user needs beyond explicit requests.

Ax Juanwu Lu, Ziran Wang 5/12/2026

On Variance Reduction in Learning Mean Flows

Theoretical analysis of variance reduction in MeanFlow one-step generative modeling, addressing training instability and gradient variance.

Ax Aditya Sinha, Harald Steck, Vito Ostuni, Matteo Rinaldi 5/12/2026

Beyond Continuity: Challenges of Context Switching in Multi-Turn Dialogue with LLMs

Study of LLM multi-turn dialogue understanding, analyzing failures in context switching and topic detection across conversation turns.

Ax Ruiying Peng, Mengyu Yang, Jing Lei, Xiaohui Li, Xueyu Wu, Xinlei Chen 5/12/2026

Memorize Theorems, Not Instances: Probing SFT Generalization through Mathematical Reasoning

Theorem-SFT approach for supervised fine-tuning that improves LLM generalization on mathematical reasoning by targeting principles over instances.

Ax Muhammed Ustaomeroglu, Guannan Qu 5/12/2026

Towards Effective Theory of LLMs: A Representation Learning Approach

Representational Effective Theory framework for understanding LLM computation through learned macrostates from hidden-state trajectories.

Ax Samuel J. K. Chin, Maximilian Schiffer 5/12/2026

Neural Cluster First, Route Second: One-Shot Capacitated Vehicle Routing via Differentiable Optimal Transport

Neural optimization method using differentiable optimal transport for vehicle routing problems, replacing autoregressive decoding.

Ax Seungwoo Kum 5/12/2026

Hierarchical Attention-based Graph Neural Network with Relevance-driven Pruning

Graph Neural Network framework with attention mechanism for interpretability and computational efficiency on heterogeneous graphs.

Ax Jeongho Bang, Marcin Paw{\l}owski 5/12/2026

Neural Information Causality

Theoretical framework embedding information causality into representation learning via query-separated computation, tangential to core interests.

Ax Luoxi Tang, Rupali Rajendra Vaje, Yuqiao Meng, Sakshi Sunil Narkar, Weicheng Ma, Zeyu Ding, Dazheng Zhang, Zhaohan Xi 5/12/2026

The Trap of Trajectory: Towards Understanding and Mitigating Spurious Correlations in Agentic Memory

arXiv paper investigating spurious correlations in agentic memory systems where retrieved context propagates erroneous reasoning in LLM decision-making.

Ax Xiaocheng Luo, Kang Wang, Zaifu Zhan, Yuechi Zhou, Xiangyu Duan 5/12/2026

RuPLaR : Efficient Latent Compression of LLM Reasoning Chains with Rule-Based Priors From Multi-Step to One-Step

arXiv paper proposing latent chain-of-thought reasoning compression using rule-based priors to replace multi-step reasoning with efficient one-step inference.

Ax Shusaku Egami, Aoi Ohta, Tomoki Tsujimura, Masaki Asada, Tatsuya Ishigaki, Ken Fukuda, Masahiro Hamasaki, Hiroya Takamura 5/12/2026

HOME-KGQA: A Benchmark Dataset for Multimodal Knowledge Graph Question Answering on Household Daily Activities

Benchmark dataset for multimodal knowledge graph question answering integrating LLMs with knowledge graphs to reduce hallucinations and leverage external knowledge.

Ax Yash Vishe, Rohan Surana, Xunyi Jiang, Zihan Huang, Xintong Li, Nikki Lijing Kuang, Tong Yu, Ryan A. Rossi, Jingbo Shang, Julian McAuley, Junda Wu 5/12/2026

Skill-R1: Agent Skill Evolution via Reinforcement Learning

arXiv paper on optimizing reusable skills for agentic LLMs through reinforcement learning with two-level credit assignment, moving beyond prompt engineering.

Ax Zhenghan Song, Yulong Liu, Cheng Wan, Chenjun Li, Lingfu Liu, Yunyi Li, Congcong Yuan 5/12/2026

Your Simulation Runs but Solves the Wrong Physics: PDE-Grounded Intent Verification for LLM-Generated Multiphysics Simulation Code

Framework for verifying LLM-generated PDE simulation code by detecting comprehension-generation gaps where code executes but encodes wrong physics.

Ax Daemyung Kang, Eunjin Hwang, Hanjeong Lee, HyeokJin Kim, Hyunhoi Koo, Jeongkyu Shin, Jeongseok Kang, Jihyun Kang, Joongi Kim, Junbum Lee, Jungseung Yang, Kyujin Cho, Youngsook Song 5/12/2026

From Detection to Recovery: Operational Analysis on LLM Pre-training with 504 GPUs

Operational analysis of 504-GPU production cluster documenting hardware failures, recovery patterns, and distributed training reliability over 55 days.

Ax Xinyi Wu, Jayant Teotia, Shuai Zhao, Erik Cambria 5/12/2026

EduStory: A Unified Framework for Pedagogically-Consistent Multi-Shot STEM Instructional Video Generation

EduStory framework for pedagogically-consistent multi-shot STEM video generation using pedagogical state tracking and script guidance.

Ax Runze Ma, Shunbo Jia, Haonan Lyu, Guo Liu, Caizhi Liao 5/12/2026

LiteMedCoT-VL: Parameter-Efficient Adaptation for Medical Visual Question Answering

LiteMedCoT-VL enables parameter-efficient medical VQA reasoning on compact models through knowledge distillation of reasoning steps.

Ax Dong Yang, Yiyi Cai, Haoyu Zhang, Yuki Saito, Hiroshi Saruwatari 5/12/2026

Kinetic-Optimal Scheduling with Moment Correction for Metric-Induced Discrete Flow Matching in Zero-Shot Text-to-Speech

Kinetic-optimal scheduler and moment correction method for discrete flow matching in zero-shot text-to-speech generation.

Ax Gabriel Smithline, Chris Mascioli 5/12/2026

Sparsity Moves Computation: How FFN Architecture Reshapes Attention in Small Transformers

Study of how FFN sparsity patterns reshape attention computations in small Transformers across arithmetic and counting tasks.

Ax Weijia Liufu, Xiaoyu Guo, Ruiyi Chen, Jingzhi Liu, Kaidong Zhang, Xiwen Liang, Jianqi Lin, Dawei Sun, Yuze Wang, Rongtao Xu, Bingqian Lin, Bowen Yang, Tongtong Cao, Bowen Peng, Dongyu Zhang, Guangrun Wang, Min Wang, Liang Lin, Xiaodan Liang 5/12/2026

RePO-VLA: Recovery-Driven Policy Optimization for Vision-Language-Action Models

RePO-VLA framework improves vision-language-action models for robotic manipulation by learning from recovery trajectories.

Ax Shogo Noguchi 5/12/2026

AtteConDA: Attention-Based Conflict Suppression in Multi-Condition Diffusion Models and Synthetic Data Augmentation

AtteConDA improves conditional diffusion models through attention-based conflict suppression for data augmentation applications.

Ax Jie Ma, Yihang Liu, Zhike Qiu, Jiayi Ji, Xiaoshuai Sun 5/12/2026

Evading Visual Aphasia: Contrastive Adaptive Semantic Token Pruning for Vision-Language Models

Contrastive pruning technique for vision-language models preserves essential low-attention tokens for compositional reasoning tasks.

Ax Shanwen Tan, Hao Li, Jingtao Zhang, Xiaosong Jia, Xue Yang, Shaofeng Zhang, Yanyong Zhang 5/12/2026

SWIFT: Prompt-Adaptive Memory for Efficient Interactive Long Video Generation

SWIFT enables efficient long video generation with prompt-adaptive memory management for continuous semantic transitions.

Ax Dongqi Zheng 5/12/2026

When Few Steps Are Enough: Training-Free Acceleration of Identity-Preserved Generation

Training-free acceleration of identity-preserved image generation by transferring frozen adapters to distilled diffusion backbones.

Ax Tianyu Zheng, Hong Wu, Jiaji Zhong 5/12/2026

APCD: Adaptive Path-Contrastive Decoding for Reliable Large Language Model Generation

APCD method improves LLM generation reliability by adaptively branching decoding paths to mitigate hallucinations from error accumulation.

Ax Hong Chen, Zixiang Tang, Yuanbao Chen, Yu Liu 5/12/2026

LASSA Architecture-Based Autonomous Fault-Tolerant Control of Unmanned Underwater Vehicles

LASSA architecture applies LLMs to autonomous fault-tolerant control of underwater vehicles in communication-constrained environments.

Ax Xianhe Chen, Hao Chen, Yingzhen Li 5/12/2026

Spectral Transformer Neural Processes

Spectral Transformer Neural Processes extend TNPs with frequency-aware Spectral Aggregator for time series and spatial data with strong periodicity patterns.

Ax Michael A. Riegler, Inga Str\"umke 5/12/2026

Position: AI Security Policy Should Target Systems, Not Models

Swarm-attack: open-source framework where multiple LLM agents coordinate to discover safety bypasses and software vulnerabilities at minimal cost, demonstrating systemic risks.

Ax Jakob Sturm, Josef Pichlmeier, Christian Bernhard, Maka Karalashvili, Johannes Klepsch, Georg Groh, Andre Luckow 5/12/2026

Assessment of RAG and Fine-Tuning for Industrial Question-Answering-Applications

Empirical comparison of RAG versus fine-tuning for industrial LLM-based question-answering systems, evaluating cost-accuracy trade-offs in domain-specific enterprise scenarios.

Ax Elyson A. De La Cruz, Rishikesh Sahay, Md Rasel Al Mamun 5/12/2026

Governing AI-Assisted Security Operations: A Design Science Framework for Operational Decision Support

Design science framework for governing AI-assisted security operations, addressing accountability, privacy, cost, and auditability in high-risk operational decision support.

Ax Haoyang Zhou, Li Kong, Shijie Ren, Xiting Wang, Shuang Liang, Guowei Wang, Zhenxuan Pan 5/12/2026

TAD: Temporal-Aware Trajectory Self-Distillation for Fast and Accurate Diffusion LLM

TAD framework addresses accuracy-parallelism trade-off in diffusion LLMs through temporal-aware trajectory self-distillation for faster, accurate parallel text generation.

Ax Guanyi Du, Lintao Wang, Kun Hu, Ziyang Wang 5/12/2026

KAN Text to Vision? The Exploration of Kolmogorov-Arnold Networks for Multi-Scale Sequence-Based Pose Animation from Sign Language Notation

KANMultiSign uses Kolmogorov-Arnold Networks with multi-scale supervision to generate 2D pose sequences from HamNoSys sign language notation.

Ax Aishik Nagar, Arun-Kumar Kaliya-Perumal, Yu-Hsuan Han, Andrew Sheng-Han Huang, Kristen Kee, Yushi Cao, Yiming Chen, Hongchao Jiang 5/12/2026

CLR-voyance: Reinforcing Open-Ended Reasoning for Inpatient Clinical Decision Support with Outcome-Aware Rubrics

CLR-voyance framework evaluates clinical LLM reasoning as sequential decision-making under partial observability using outcome-aware rubrics instead of closed-form retrieval.

Ax Tzeh Yuan Neoh, Nicholas Teh, Je Qin Chooi, Paul W. Goldberg, Milind Tambe 5/12/2026

Efficient Ensemble Selection from Binary and Pairwise Feedback

Method for efficient ensemble selection from multiple AI systems using minimal model calls and human evaluation, formulated as distributional multiwinner voting.

Ax Abhinav Goel, Agostino Capponi, Alfio Gliozzo, Chaitya Shah 5/12/2026

SmartEval: A Benchmark for Evaluating LLM-Generated Smart Contracts from Natural Language Specifications

SmartEval benchmark for evaluating LLM-generated Solidity smart contracts against specifications, with 9,000 contracts and five-dimensional rubric for code quality assessment.

Ax Yuhan Wang, Zihan Li, Han Liu, Simon Arberet, Martin Kraus, Yuyin Zhou, Florin-Cristian Ghesu, Dorin Comaniciu, Ali Kamen, Riqiang Gao 5/12/2026

Any2Any 3D Diffusion Models with Knowledge Transfer: A Radiotherapy Planning Study

DiffKT3D uses diffusion models with knowledge transfer from vision domains for voxel-wise dose prediction in radiotherapy planning, demonstrating cross-domain generalization.

Ax Akuen Akoi Deng, Eimantas Butkus, Alfreds Lapkovskis, Praveen Kumar Donta 5/12/2026

Adaptive DNN Partitioning and Offloading in Heterogeneous Edge-Cloud Continuum

Framework for dynamic DNN partitioning and offloading across edge-cloud devices, tested on real hardware rather than simulation for resource-constrained AI deployment.

Ax Huy Hoang Ha, Benoit Favre, Francois Portet 5/12/2026

MedMeta: A Benchmark for LLMs in Synthesizing Meta-Analysis Conclusion from Medical Studies

MedMeta benchmark evaluates LLMs on higher-order reasoning by synthesizing medical meta-analysis conclusions from study abstracts, addressing a gap in medical LLM evaluation.

Ax Jingze Song, Zihao Chen, Wenqing Chen, Zibin Zheng 5/12/2026

Learning Multi-Indicator Weights for Data Selection: A Joint Task-Model Adaptation Framework with Efficient Proxies

Data selection framework using multi-indicator weights for efficient instruction tuning of LLMs with task-model adaptation.

Ax Yixiong Chen, Wenjie Xiao, Pedro R. A. S. Bassi, Boyan Wang, Liang He, Xinze Zhou, Sezgin Er, Ibrahim Ethem Hamamci, Zongwei Zhou, Alan Yuille 5/12/2026

DeepTumorVQA: A Hierarchical 3D CT Benchmark for Stage-Wise Evaluation of Medical VLMs and Tool-Augmented Agents

Hierarchical benchmark for medical VLMs and tool-augmented agents on 3D CT tumor analysis with stage-wise evaluation.

Ax Monika Jotautait\.e, Maria Angelica Martinez, Ollie Matthews, Tyler Tracy 5/12/2026

MonitoringBench: Semi-Automated Red-Teaming for Agent Monitoring

Red-teaming methodology for coding-agent monitors exposing attack generation gaps with attack taxonomy for coverage.

Ax Santhosh Kumar Ramakrishnan, Carl Vondrick, Raja Giryes, Philipp Kr\"ahenb\"uhl, Vladlen Koltun 5/12/2026

Do multimodal models imagine electric sheep?

Fine-tuned Qwen VLM demonstrates spatial reasoning and mental imagery on visual puzzles like tangram and 3D rotation.

Ax V\'ictor Gallego 5/12/2026

Metal-Sci: A Scientific Compute Benchmark for Evolutionary LLM Kernel Search on Apple Silicon

Benchmark for evolutionary LLM-based kernel search on Apple Silicon Metal compute across scientific computing tasks.

Ax Alaa Asfour, Christopher Indris, Leihan Chen, Tejas Vyas, Guanghui Wang 5/12/2026

Distilling 3D Spatial Reasoning into a Lightweight Vision-Language Model with CoT

Knowledge distillation framework transferring 3D spatial reasoning from 7B to 2.29B vision-language model via CoT.

Ax Hardik Goel 5/12/2026

Security Risks in Tool-Enabled AI Agents: A Systematic Analysis of Privileged Execution Environments

Systematic security analysis of tool-enabled AI agents in privileged cloud environments identifying attack vectors and risks.

Ax Bowen He, Juncheng Dong, Lin Lin, Xiang Cheng 5/12/2026

One for All: A Non-Linear Transformer can Enable Cross-Domain Generalization for In-Context Reinforcement Learning

Transformer architecture for in-context RL enabling cross-domain generalization without parameter updates via in-context learning.

Ax Vishnu Vardhan Reddy, Sagnik Chatterjee, Soumik Bhatta 5/12/2026

Trajectory Supervision for Continual Tool-Use Learning in LLMs

Studies tool-use learning in LLMs by fine-tuning Llama 3.1 with trajectory supervision across sequential API domains.

Ax Zhiqing Zhong, Zhijing Ye, Jian Zhang, Weijian Zheng, Bolun Sun, Xiaodong Yu 5/12/2026

KV-RM: Regularizing KV-Cache Movement for Static-Graph LLM Serving

KV-cache management optimization for static-graph LLM decoders to reduce memory overallocation and latency.

Ax Sanket Badhe, Priyanka Tiwari, Deep Shah 5/12/2026

The Silent Vote: Improving Zero-Shot LLM Reliability by Aggregating Semantic Neighborhoods

Addresses zero-shot LLM classification bias by aggregating semantic neighborhoods to recover discarded probability mass.

Ax Taylan Soydan, Miguel A. Bessa, Dirk Mohr, Rui Barreira 5/12/2026

TIDES: Implicit Time-Awareness in Selective State Space Models

TIDES: state space model improvements for irregular time series by preserving physical meaning of time discretization steps.

Ax Benjamin Patrick Evans, Sumitra Ganesh, Leo Ardon 5/12/2026

Entropy-informed Decoding: Adaptive Information-Driven Branching

Novel decoding strategy for LLMs that adaptively branches based on entropy to improve output quality while reducing computation.