Ax Zheyu Zhang, Shuo Yang, Bardh Prenkaj, Gjergji Kasneci 5/12/2026

Active Tabular Augmentation via Policy-Guided Diffusion Inpainting

Active tabular data augmentation using policy-guided diffusion to reduce fidelity-utility gap in generative augmentation for downstream models.

Ax Sean Feeney, Pooja Rao, Andreas Klappenecker, Reuben Tate, Yuri Alexeev, Stefano Mensa, Elica Kyoseva, Stephan Eidenbenz 5/12/2026

SCALAR: A Neurosymbolic Framework for Automated Conjecture and Reasoning in Quantum Circuit Analysis

SCALAR: neurosymbolic framework integrating quantum simulation, symbolic reasoning, and LLMs for automated conjecture generation in quantum circuit analysis.

Ax Yao Lu, Dengdong Fan, Shixun Zhang, Yonghong Tian 5/12/2026

PowerStep: Memory-Efficient Adaptive Optimization via $\ell_p$-Norm Steepest Descent

PowerStep: memory-efficient adaptive optimizer achieving coordinate-wise adaptivity without storing second-moment statistics for transformer training.

Ax Zichen Wen, Boxue Yang, Junlong Ke, Jiajie Huang, Chenfei Liao, Junxi Wang, Xuyang Liu, Linfeng Zhang 5/12/2026

EvoStreaming: Your Offline Video Model Is a Natively Streaming Assistant

EvoStreaming: framework enabling offline video-language models to function as native streaming assistants with real-time decision-making.

Ax Rashi Sharma, Justin Timothy C. Bersamin, Karthikk Subramanian 5/12/2026

Portable Active Learning for Object Detection

Portable active learning method for object detection that minimizes annotation costs without requiring model-specific feature dependencies.

Ax Danni Xu, Shaojing Fan, Harry Cheng, Mohan Kankanhalli 5/12/2026

RW-Post: Auditable Evidence-Grounded Multimodal Fact-Checking in the Wild

RW-Post: multimodal fact-checking benchmark with auditable annotations linking social media posts to human-verified evidence via LLM reasoning traces.

Ax Team Phoenix, :, Arka Ray, Askar Ali Mohamed Jawad, Biondi Lee, Elijah Seah, Eva Lim, Fiona Teo, Grace Toh, Guang Xiang Teo, Jun En Tan, Jia Hui Bong, Jiale Wang, Jonathan Ng, Justin Tan, Kai Zhe Yew, Matthew Ong, Shun Yi Yeo, Wen Jett Lam, Wen Xiu Tan, Ze Yu Zhang, Gee Wah Ng, Chee Wee Ang, Mistral AI, :, Adrien Sad\'e, Guillaume Kunsch, Jia Sin Loh, Nicolas Schuhl, Rupert Menneer, Umar Jamil, Vincent Maladi\`ere, Yimu Pan 5/12/2026

Phoenix-VL 1.5 Medium Technical Report

Phoenix-VL 1.5 Medium: 123B-parameter multimodal multilingual foundation model adapted for regional languages and Singapore context.

Ax Xi Jiang, Yinjie Zhao, Zesheng Yang, Feng Zheng 5/12/2026

AnomalyClaw: A Universal Visual Anomaly Detection Agent via Tool-Grounded Refutation

AnomalyClaw: visual anomaly detection agent using vision-language models with tool-grounding for cross-domain industrial and medical applications.

Ax Andreas Xenofontos, Pavlos Fafalios 5/12/2026

Can Language Models Analyze Data? Evaluating Large Language Models for Question Answering over Datasets

Evaluation of LLMs for answering questions over datasets, including direct analysis and SQL query generation with various prompting strategies.

Ax Jiaqi W. Ma 5/12/2026

Toward an Engineering of Science: Rebalancing Generation and Verification in the Age of AI

Analysis of epistemic pollution risk from AI-generated scientific artifacts and need for rebalancing generation and verification in scientific systems.

Ax Minqing Huang, Yujiao Xiang, Zihan Liang, Jiajie Huang, Jingqi Wang, Zhi Xu, Feiyang Tan, Hangning Zhou, Mu Yang, Gong Che 5/12/2026

CoWorld-VLA: Thinking in a Multi-Expert World Model for Autonomous Driving

CoWorld-VLA: vision-language-action model for autonomous driving using multi-expert world models and spatiotemporal reasoning.

Ax George Panagopoulos 5/12/2026

Real vs. Semi-Simulated: Rethinking Evaluation for Treatment Effect Estimation

Comparative study of evaluation methods for heterogeneous treatment effect estimation, contrasting semi-simulated benchmarks with real-world observable metrics.

Ax Pierre Le Jeune, \'Etienne Duchesne, Weixuan Xiao, Stefano Palminteri, Bazire Houssin, Beno\^it Mal\'ezieux, Matteo Dora 5/12/2026

StereoTales: A Multilingual Framework for Open-Ended Stereotype Discovery in LLMs

StereoTales: multilingual dataset and evaluation framework for detecting social bias in open-ended LLM generation across 10 languages and 79 demographic attributes.

Ax Haoyu Huang, Jiaxin Bai, Shujie Liu, Yang Wei, Hong Ting Tsang, Yisen Gao, Zhongwei Xie, Yufei Li, Yangqiu Song 5/12/2026

DeepRefine: Agent-Compiled Knowledge Refinement via Reinforcement Learning

DeepRefine: reinforcement learning approach for refining agent-compiled knowledge bases by addressing incompleteness, incorrectness, and redundancy.

Ax Marius Miron, David Robinson, Masato Hagiwara, Titouan Parcollet, Jules Cauzinille, Gagan Narula, Milad Alizadeh, Ellen Gilsenan-McMahon, Sara Keen, Emmanuel Chemla, Benjamin Hoffman, Maddie Cusimano, Diane Kim, Felix Effenberger, Jane K. Lawton, Aza Raskin, Olivier Pietquin, Matthieu Geist 5/12/2026

Multi-layer attentive probing improves transfer of audio representations for bioacoustics

Study of multi-layer attentive probe design for improving transfer of audio representations in bioacoustics tasks.

Ax Heegeon Yoon, Heeyoung Kim 5/12/2026

Simultaneous Long-tailed Recognition and Multi-modal Fusion for Highly Imbalanced Multi-modal Data

Framework for long-tailed recognition in class-imbalanced multimodal data exploiting complementary information across modalities.

Ax Yousef A. Radwan, Yao Li, Qing Qing, Ziqi Xu, Qixin Zhang, Yongcheng Jing, Renqiang Luo, Xikun Zhang 5/12/2026

CMKL: Modality-Aware Continual Learning for Evolving Biomedical Knowledge Graphs

CMKL: continual learning framework for multimodal biomedical knowledge graphs with evolving structure and data.

Ax Ali Irzam Kathia, Yimika Erinle, Abylay Satybaldy, Paolo Tasca, Nikhil Vadgama, Marco Alberto Javarone 5/12/2026

SoK: A Systematic Bidirectional Literature Review of AI & DLT Convergence

Systematic review of AI and distributed ledger technology convergence, examining architectural interplay and application domains.

Ax Jaehoon Yoo, Wonjung Kim, Chanhyuk Lee, Seunghoon Hong 5/12/2026

Infinite Mask Diffusion for Few-Step Distillation

Infinite Mask Diffusion: distillation method for accelerating masked diffusion models to few-step generation.

Ax Yiqi Tian, Sangjoon Park, Bo Zeng, Pengfei Jin, Yujin Oh, Quanzheng Li 5/12/2026

DuetFair: Coupling Inter- and Intra-Subgroup Robustness for Fair Medical Image Segmentation

DuetFair: fairness mechanism for medical image segmentation addressing intra-group performance disparities across subgroups.

Ax Honghan Wu, Tianyan Wang, Jiacong Mi, Zhoyang Jiang, Yunsoo Kim 5/12/2026

HH-SAE: Discovering and Steering Hierarchical Knowledge of Complex Manifolds

Hybrid Hierarchical Sparse Autoencoders for discovering and steering hierarchical knowledge in complex high-dimensional manifolds.

Ax Davide Bruni, Carlo Bardazzi, Maurizio Tesconi 5/12/2026

ThreatCore: A Benchmark for Explicit and Implicit Threat Detection

ThreatCore: benchmark dataset for fine-grained threat detection distinguishing explicit threats, implicit threats, and non-threats in text.

Ax Phongsakon Mark Konrad, Toygar Tanyel, Serkan Ayvaz 5/12/2026

Acceptance Cards:A Four-Diagnostic Standard for Safe Fine-Tuning Defense Claims

Acceptance Cards: evaluation protocol and audit package for validating safe fine-tuning defense claims with statistical rigor.

Ax Chen Zhong, Xiao An, Jiaxing Sun, Zihan Gui, Guangyi Yang, Wei He 5/12/2026

SenseBench: A Benchmark for Remote Sensing Low-Level Visual Perception and Description in Large Vision-Language Models

SenseBench: benchmark for evaluating vision-language models on remote sensing image quality assessment and degradation description.

Ax Zheng Lin, Zhenxing Niu, Haoxuan Ji, Haichang Gao 5/12/2026

Guaranteed Jailbreaking Defense via Disrupt-and-Rectify Smoothing

Disrupt-and-Rectify Smoothing: guaranteed defense against jailbreak attacks for LLMs using two-stage prompt processing.

Ax Yiming Dong, Yacheng Kang, Junjie Zhao, Xinyuan Zhu, Ziming Wang, Lijing Shao 5/12/2026

An agentic framework for gravitational-wave counterpart association in the multi-messenger era

Agent framework for associating gravitational wave signals with electromagnetic counterparts in multi-messenger astronomy.

Ax Isaac David, Arthur Gervais 5/12/2026

CrackMeBench: Binary Reverse Engineering for Agents

CrackMeBench: benchmark for evaluating AI agents on binary reverse engineering tasks without source code access.

Ax Mieke Wilms, Christoph Heitz 5/12/2026

Fairness vs Performance: Characterizing the Pareto Frontier of Algorithmic Decision Systems

Investigation of fairness-performance trade-offs in algorithmic decision systems as multi-objective optimization problem.

Ax Benjamin Icard, Lila Sainero, Alice Breton, Evangelia Zve, Jean-Gabriel Ganascia 5/12/2026

Measuring Embedding Sensitivity to Authorial Style in French: Comparing Literary Texts with Language Model Rewritings

Analysis of authorial style encoding in language model embeddings using French literary texts and LLM rewritings.

Ax Zheng Lin, Zhenxing Niu, Haoxuan Ji, Yuzhe Huang, Haichang Gao 5/12/2026

Re-Triggering Safeguards within LLMs for Jailbreak Detection

Defense method against jailbreak attacks that detects and re-activates built-in LLM safeguards through embedding disruption.

Ax Bruno Gatti, Giuliano Martinelli, Roberto Navigli 5/12/2026

Interpretable Coreference Resolution Evaluation Using Explicit Semantics

New evaluation methodology for coreference resolution using explicit semantic categories instead of aggregate metrics for better diagnostic insights.

Ax Krishak Aneja, Manas Mittal, Anmol Goel, Ponnurangam Kumaraguru, Vamshi Krishna Bonagiri 5/12/2026

Intrinsic Guardrails: How Semantic Geometry of Personality Interacts with Emergent Misalignment in LLMs

Study of how LLM personality geometry relates to emergent misalignment vulnerabilities when fine-tuned on narrow data, using psychometric profiles.

Ax Haoyu Wang, Yifan Shang, Zhongxiang Sun, Weijie Yu, Xiao Zhang, Jun Xu 5/12/2026

Towards Understanding Continual Factual Knowledge Acquisition of Language Models: From Theory to Algorithm

Theoretical framework and algorithm for continual factual knowledge acquisition in language models, addressing how LMs integrate new knowledge without catastrophic forgetting.

Ax Nikolaos Gkalelis, Vasileios Mezaris 5/12/2026

LLaVA-CKD: Bottom-Up Cascaded Knowledge Distillation for Vision-Language Models

LLaVA-CKD applies cascaded knowledge distillation to compress vision-language models for practical deployment.

Ax Zheng Li, Feng Xie, Shenglan Nie, Xichen Guo, Ruxin Wang, Hao Zhang 5/12/2026

A Recursive Decomposition Framework for Causal Structure Learning in the Presence of Latent Variables

Recursive decomposition framework for causal structure learning with latent variables extending divide-and-conquer strategies.

Ax Mumin Jia, Yilin Chen, Divya Sharma, Jairo Diaz-Rodriguez 5/12/2026

When Can Digital Personas Reliably Approximate Human Survey Findings?

Study testing whether LLM-powered digital personas can reliably approximate human survey responses using LISS panel data.

Ax Michal Byra, Pawel Olszowiec, Grzegorz Stefanski, Grzegorz Gruszczynski, Alberto Presta 5/12/2026

bViT: Investigating Single-Block Recurrence in Vision Transformers for Image Recognition

bViT architecture exploring single-block recurrence in Vision Transformers to reduce depth while maintaining performance.

Ax Diancheng Kang, Zheyuan Liu, Ningshan Ma, Yue Huang, Zhaoxuan Tan, Meng Jiang 5/12/2026

Prompt-Activation Duality: Improving Activation Steering via Attention-Level Interventions

Gated Cropped Attention-Delta steering improves activation steering in dialogue by addressing KV-cache contamination in language models.

Ax Igor Slinko, Ilia Zavidnyi, Egor Bogomolov, Yaroslav Zharov 5/12/2026

Step Rejection Fine-Tuning: A Practical Distillation Recipe

Step Rejection Fine-Tuning method for training LLM agents that leverages partially correct trajectories instead of discarding unresolved paths.

Ax Xiao Tian, Jue Fan, Rachael Hwee Ling Sim, Zixuan Wang, Nancy F. Chen, Bryan Kian Hsiang Low 5/12/2026

Is Data Shapley Not Better than Random in Data Selection? Ask NASH

NASH framework examining effectiveness of Data Shapley values for data selection, comparing against random selection baselines.

Ax Dahlia Shehata, Ming Li 5/12/2026

The Bystander Effect in Multi-Agent Reasoning: Quantifying Cognitive Loafing in Collaborative Interactions

Study of cognitive loafing in multi-agent LLM systems showing collaboration can degrade reasoning quality due to social pressure effects.

Ax Fred Philippy, Siwen Guo, Jacques Klein, Tegawend\'e F. Bissyand\'e 5/12/2026

Why Low-Resource NLP Needs More Than Cross-Lingual Transfer: Lessons Learned from Luxembourgish

Research on limitations of cross-lingual transfer for low-resource NLP using Luxembourgish as case study, showing need for language-specific approaches.

Ax Suvi De Silva, Alfreds Lapkovskis, Alaa Saleh, Sasu Tarkoma, Praveen Kumar Donta 5/12/2026

An Uncertainty-Aware Resilience Micro-Agent for Causal Observability in the Computing Continuum

AURORA micro-agent framework for diagnosing and mitigating grey failures in edge computing with causal reasoning under uncertainty.

Ax David F. Ramirez, Tim Overman, Kristen Jaskie, Andreas Spanias 5/12/2026

Geospatial-Temporal Sensemaking of Remote Sensing Activity Detections with Multimodal Large Language Model

Visual question answering dataset from satellite imagery for spatiotemporal analysis of construction activity.

Ax Li Shen, Xiaolei Hao, Qinglun Li, Xiaochun Cao, Zhifeng Hao, Xun Yang 5/12/2026

Provable Sparse Inversion and Token Relabel Enhanced One-shot Federated Learning with ViTs

One-shot federated learning framework with token relabeling for vision transformers under non-IID data distributions.

Ax Mohamed Eltahir, Lama Ayash, Ali Habibullah, Tanveer Hussain, Naeemullah Khan 5/12/2026

GridProbe: Posterior-Probing for Adaptive Test-Time Compute in Long-Video VLMs

Adaptive frame selection method for efficient long-video understanding in vision-language models via posterior probing.

Ax Mengqi He, Xinyu Tian, Xin Shen, Shu Zou, Jinhong Ni, Zhaoyuan Yang, Weikang Li, Xuesong Li, Jing Zhang 5/12/2026

Break the Brake, Not the Wheel: Untargeted Jailbreak via Entropy Maximization

Entropy maximization approach for untargeted jailbreaks on vision-language models with improved transferability.

Ax Tao Hu, Da-Wei Zhou 5/12/2026

Dynamic Cross-Modal Prompt Generation for Multimodal Continual Instruction Tuning

Cross-modal prompt generation framework for multimodal continual instruction tuning to mitigate catastrophic forgetting.

Ax Ziyi Wang, Xianping Ma, Ziyao Wang, Hongyang Zhang, Man On Pun 5/12/2026

MPerS: Dynamic MLLM MixExperts Perception-Guided Remote Sensing Scene Segmentation

Dynamic mixture-of-experts MLLM for remote sensing scene segmentation with multimodal caption guidance.

Ax David F. Ramirez, Tim L. Overman, Kristen Jaskie, Marv Kleine, Andreas Spanias 5/12/2026

Towards a Large Language-Vision Question Answering Model for MSTAR Automatic Target Recognition

Application of large language-vision models to remote sensing automatic target recognition tasks.