Isolater - Feed

Ax Ta Duc Huy, Trang Nguyen, Townim Chowdhury, Ankit Yadav, Minh-Son To, Zhibin Liao, Johan W. Verjans, Vu Minh Hieu Phan 7/1/2026

Visual Semantic Entropy: Do Vision Language Models Recognize Visual Ambiguity?

Analysis of uncertainty estimation in vision-language models showing entropy methods underestimate confidence on ambiguous inputs.

Ax Karam Tomotaki-Dawoud, Anna Hilsmann, Peter Eisert, Sebastian Bosse 7/1/2026

Temporal Preservation over Processing: Diagnosing and Designing Spatiotemporal Single-Stage Video Detectors

Framework diagnosing whether video object detectors reason over temporal context or exploit single frames.

Ax Yizhe Liu, Shaolei Zhang, Ju Fan 7/1/2026

DA-Studio: An Agentic System for End-to-End Data Analysis

DA-Studio: Agentic system for autonomous multi-step data analysis with code execution, workflow orchestration, and interpretable traces.

Ax Kyomin Hwang, Hyeonjin Kim, Hyunho Lee, Nojun Kwak 7/1/2026

Team MKC at CLPsych 2026: Capturing and Characterizing Mental Health Changes through Social Media Timeline Dynamics

LLM application for mental health monitoring and early detection using social media timeline analysis.

Ax Yifei Dong, Zhanyi Sun, Lujie Yang, Manuel Baum, Kei Ikemura, Shuran Song, Florian T. Pokorny, Xianyi Cheng 7/1/2026

Robustness of Robotic Manipulation: Foundations and Frontiers

Systematic study of robustness in robotic manipulation, establishing unified framework across subfields for human-level performance.

Ax Muhammad Usman Safder (Steve), Ayesha Gull (Steve), Rania Elbadry (Steve), Fan Zhang (Steve), Yankai Chen (Steve), Xueqing Peng (Steve), Xue (Steve), Liu, Preslav Nakov, Zhuohan Xie 7/1/2026

FinPersona-Bench: A Benchmark for Longitudinal Psychometric Stability of Autonomous Financial Agents

FinPersona-Bench: benchmark measuring behavioral mandate decay in autonomous LLM financial agents over long deployment horizons.

Ax Xudong Wu, Pangpang Liu, Vaneet Aggarwal, Jiayu Chen 7/1/2026

On the Convergence of Self-Improving Online LLM Alignment

Theoretical analysis of Self-Improving Alignment (SAIL) algorithm convergence for online LLM alignment under distribution shift.

Ax Xu Yan, Huiqun Wang, Chen Wang, Lei Ren, Di Huang 7/1/2026

Mitigating Positional Leakage in 3D Masked Autoencoders for Robust Representation Learning

Research on mitigating positional leakage in 3D Masked Autoencoders for self-supervised point cloud learning.

Ax Zikang Yan, Xiao Wang, Qingquan Yang, Zhendong Yang, Gaoting Chen, Zehua Chen, Bo Jiang, Jin Tang, Guosheng Xu 7/1/2026

Temperature Field Reconstruction of Tungsten Monoblock Divertor on EAST using Physics-aware Neural Operator Transformer

Physics-aware Neural Operator Transformer for real-time temperature field reconstruction in fusion reactor divertor.

Ax Shun Kenney, Teppei Suzuki 7/1/2026

DPPE: Rethinking Camera-Based Positional Encoding for Scaling Multi-View Transformers

DPPE: camera-aware positional encoding technique for scaling multi-view Transformers in 3D computer vision.

Ax Asif Hanif, Mohammad Yaqub 7/1/2026

ZEBRA: Zero-Shot Entropy-Regularized Prompt Learning for Base-to-Novel Generalization in Audio-Language Models

ZEBRA: zero-shot prompt learning method for Audio-Language Models addressing base-to-novel generalization gap.

Ax Jason R. Brown, Patrick Leask, Lev McKinney 7/1/2026

Evil Spectra: How Optimisers can Amplify or Suppress Emergent Misalignment

Analysis of how optimizers affect emergent misalignment in LLMs, showing fine-tuning on narrow misaligned tasks affects unrelated behavior.

Ax Rana Alharbi, Chuadhry Mujeeb Ahmed 7/1/2026

Comparative Analysis of Machine Learning based Intrusion Detection in Realistic IoT Networks

Comparative study of machine learning methods for intrusion detection in resource-constrained IoT networks.

Ax Kaitao Chen, Weiqian Zhao, Jiamin Wu, Qihao Zheng, Shangquan Sun, Chunfeng Song, Xiaosong Wang, Mu Zhou, Mianxin Liu 7/1/2026

Token-Sparse Medical Multimodal Reasoning via Dual-Stream Reinforcement Learning

Medical VLM reasoning via dual-stream RL for active visual token pruning on sparse medical imagery to improve clinical decision-making.

Ax Nikolai R\"ohrich, Julian Glei{\ss}ner, Ahmed H. A. Ibrahim, Silvan Mertes, Tobias Huber 7/1/2026

Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models

Data augmentation technique using diffusion models with uncertainty guidance to generate synthetic training data while preserving hard examples.

Ax Ali Zia, Muhammad Umer Ramzan, Abdelwahed Khamis, Usman Ali, Abdul Rehman 7/1/2026

Learning Structurally Consistent Representations for Multi-View Radar Semantic Segmentation

Method for multi-view radar semantic segmentation using higher-order relational structure to handle sparse and noisy measurements.

Ax Javal Vyas, Milapji Singh Gill, Mehmet Mercang\"oz 7/1/2026

Automating Cause-Effect Specification with Knowledge Graphs and Large Language Models

Framework automating cause-effect specification generation for process control using knowledge graphs combined with constrained LLM inference.

Ax Javal Vyas, Milapji Singh Gill, Artan Markaj, Felix Gehlhoff, Mehmet Mercang\"oz 7/1/2026

A Tutorial on Autonomous Fault-Tolerant Control Using Knowledge-Grounded LLM Agents

Tutorial on LLM agents for autonomous fault-tolerant control in process plants, with agents interpreting procedures and process data for recovery decisions.

Ax Seyed Bagher Hashemi Natanzi, Bo Tang 7/1/2026

A Lifecycle and Application-Stack Survey of Large Language Model Vulnerabilities: Attacks, Risks, Defenses, and Open Problems

Comprehensive survey of LLM vulnerabilities across lifecycle and application stack including attacks, risks, defenses for agents and tool-using systems.

Ax Zijun Xie, Binbin Zheng, Enlei Gong, Jihua Liu, Yuyang You, Lingfeng Liu, Jiayao Tang, Guanqun Zhao, Aoqi Hu, Zeyu Chen 7/1/2026

ECHO: Prune to act, trace to learn with selective turn memory in agentic RL

Framework for LLM agents in RL with selective turn memory management, addressing context window constraints through pruning and tracing mechanisms.

Ax Matteo Melis, Jesus Martinez Del Rincon, Vishal Sharma 7/1/2026

Improving Certified Robustness via Adversarial Distillation

Method for improving certified robustness in neural networks via adversarial distillation combining tight relaxation bounds with adversarial training.

Ax Dimitrios Koutsianos, Ladislav Mo\v{s}ner, Yannis Panagakis, Themos Stafylakis 7/1/2026

Sparsity-Inducing Divergence Losses for Biometric Verification

Research on sparsity-inducing divergence losses for biometric face and speaker verification extending margin-penalty approaches.

Ax Ting-Bing Xu, Jiacheng Sui, Zhe Gao, Kewei Shi, Wenjin Yang, Zhicheng Liu, Zhaoxu Sun, Mingchao Sun, Hongyu Pan, Fan Jiang, Mu Xu, Qi Fan, Yong Li, Baoquan Chen 7/1/2026

WorldRoamBench: An Open-World Benchmark for Long-Horizon Stability of Interactive World Models

Open-world benchmark for evaluating long-horizon stability of interactive world models across action, vision, and memory dimensions.

Ax Haoming Liu, Yuanhe Guo, Yijia Cao, Shenji Wan, Hongyi Wen 7/1/2026

Histogram-constrained Image Generation

Method for controlling diffusion models with histogram constraints to generate outputs aligned with user intent balancing global and local precision.

Ax Jesus S. Aguilar-Ruiz 7/1/2026

When to Truncate a Feature Ranking: A Residual-Overlap Stopping Rule for Subset Selection

Statistical method for determining optimal feature ranking subset size using residual-overlap stopping rule in feature selection.

Ax Jiacheng Chen, Tao Zhang, Manxi Lin, Dunxian Huang, Teng Shi, Honghao Fu, Mengyan Li, Xinming Zhang, Chenchi Zhang, Xuan Lu, Xiaoxiong Du, Haibin Chen, Shaolin Ye, Hao Chang, Xiaoqi Li, Shuwen Xiao, Yujin Yuan, Jingxuan Feng, Shaopan Xiong, Huimin Yi, Ju Huang, Qiu Shen, Ying Chen, Junjun Zheng, Xiangheng Kong, Yuning Jiang 7/1/2026

ShopX: A Foundation Model for Intent-to-Item Fulfillment in Agentic Shopping

Foundation model for LLM-agent-driven shopping that bridges language understanding to item fulfillment via generative recommendation instead of search interfaces.

Ax Jingbo He, Michael F\"arber, Roberto Calandra 7/1/2026

RCT: A Robot-Collected Touch-Vision-Language Dataset for Tactile Generalization

Robot-collected multimodal dataset of 29K tactile frames from presses on 122 materials for tactile generalization in robotic manipulation.

Ax Enrico Cassano, Riccardo Renzulli, Rayyan Ahmed, Marco Grangetto, Stephan Alaniz 7/1/2026

Look But Don't Touch with Sparse Autoencoders for Unlearning in Diffusion Models

Study of sparse autoencoders for concept manipulation in diffusion models, testing assumptions about feature isolation for object erasure and steering.

Ax Dragos-Mitrut Vasile, Elena-Simona Apostol, Stefan-Adrian Toma, Adrian Paschke, Ciprian-Octavian Truica 7/1/2026

Cross-lingual Relation Extraction with Large Language Models: Zero-Shot, Few-Shot, and Fine-Tuned Evaluation on Romanian

Cross-lingual relation extraction for Romanian using LLM inference with automatic dataset translation and fine-tuning approaches on low-resource language tasks.

Ax Nan Li, Albert Gatt, Massimo Poesio 7/1/2026

Seeing Is Not Sharing: Some Vision-Language Models Overestimate Common Ground in Asymmetric Dialogue

Research on whether vision-language models can distinguish shared vs. unshared information in dialogue tasks, evaluated on 13K annotated reference expressions.

Ax Rafael Rivera Soto, Anna Wegmann, Cristina Aggazzotti 7/1/2026

STEB: Style Text Embedding Benchmark

Standardized benchmark for evaluating text style embeddings across 96 datasets and 7 languages.

Ax Maximilian Andreas Hoefler, Karsten Mueller, Wojciech Samek 7/1/2026

FedXDS: Leveraging Model Attribution Methods to counteract Data Heterogeneity in Federated Learning

Federated learning method using explainable AI attribution to mitigate data heterogeneity across distributed clients.

Ax Dohyeon Kwon, Youngjin Park 7/1/2026

CHERRY: Compressed Hierarchical Experts with Recurrent Representational Yield

Compute-efficient language model training using selective supervision and token-level auxiliary tasks for 15% semantic tokens.

Ax Ruijia Zhang, Jiacheng Zhu, Hanqing Zhu, Laixi Shi 7/1/2026

Geometry-Preserving Orthonormal Initialization for Low-Rank Adaptation in RLVR

Low-rank adaptation initialization strategy for RL-based fine-tuning of large language models with verifiable rewards.

Ax Junha Jung, Minbyul Jeong, Suhyeon Lim, Sungwook Jung, Jaehoon Yun, Taeyun Roh, Mujeen Sung, Jaewoo Kang 7/1/2026

Breaking Failure Cascades: Step-Aware Reinforcement Learning for Medical Multimodal Reasoning

Step-aware RL for medical multimodal reasoning addressing cascading errors in clinical image interpretation with MLLMs.

Ax Lang Cao, Renhong Chen, Luyi Li, Peng Wang, Mofan Peng, Yitong Li 7/1/2026

Z-1: Efficient Reinforcement Learning for Vision-Language-Action Models

Reinforcement learning post-training for vision-language-action robotic manipulation models to improve from policy failures.

Ax Zhaoyang Luo, Runmin Dong, Miao Yang, Fan Wei, Yushan Lai, Bin Luo, Haohuan Fu 7/1/2026

Attend, Transform, or Silence: Operator-Level Visual Skipping for Efficient Multimodal LLM Inference

Operator-level visual token skipping for efficient multimodal LLM inference reducing computation while preserving evidence.

Ax Qingyun Liu, Jiwen Zhang, Jingyi Hu, Siyuan Wang, Zhongyu Wei 7/1/2026

MECoBench: A Systematic Study of Multimodal Agent Collaboration in Embodied Environments

Benchmark evaluating multimodal LLM collaboration in embodied environments across diverse real-world tasks and cooperation modes.

Ax Yufei Li (Yongkang), Zaiwei Zhang (Yongkang), Mingfu Liang (Yongkang), Kavosh Asadi (Yongkang), Jay Xu (Yongkang), Jimmy Kim (Yongkang), Chongyang Bai (Yongkang), Jieyi Zhang (Yongkang), Hongye Xie (Yongkang), Prachi Agrawal (Yongkang), Dian Yu (Yongkang), Tianyi Chen (Yongkang), Jean-Pascal Billaud (Yongkang), Garret Buell (Yongkang), YK (Yongkang), Zhu (Yang), Sachin Patil (Yang), Brooke Bian (Yang), Zhou Fang (Yang), Kevin Huang (Yang), Shiva Sudanagunta (Yang), Yuzhen Huang (Yang), Emma Lu (Yang), Chris O'Brien (Yang), Yang Song (Yang), Lihong Li (Yang), Jacob Tao (Yang), Zhicheng Zhu (Yang), Chao Li (Yang), Gaoxiang Liu (Yang), Neil Wu (Yang), Zhongyin Hu (Yang), Li Han (Yang), Loki Chen (Yang), Ming Lei (Yang), Greg Rehm (Yang), Siyuan Song (Yang), Tianwei Zhang (Yang), Li Li (Yang), Ketan Singh (Yang), Yavuz Yetim (Yang), Ilyas Atishev (Yang), Satendra Gera (Yang), Ashkan Sadeghi (Yang), Rachel Yan (Yang), Nikko Mizutani (Yang), Shuaiwen Wang (Yang), Song Yang (Yang), Zhijing Li (Yang), Jiang Liu (Yang), Mengying Sun (Yang), Fei Tian (Yang), Xiaohan Wei (Yang), Chonglin Sun (Yang), Parish Aggarwal (Yang), Kaushik Rangadurai (Yang), Zhi Hua (Yang), Frank Shyu (Yang), Ruchit Sharma (Yang), Liyuan Li (Yang), Shike Mei (Yang), Wenlin Chen (Yang), Santanu Kolay (Yang), Ben Schulte (Yang), Deepak Chandra (Yang), Adam (Yang), Song, Sandeep Pandey, Xi Liu, Hamed Firooz, Luke Simon 7/1/2026