Ax Wuyang Zhang, Shichao Pei 27d ago

Your LLM Agent Can Leak Your Data: Data Exfiltration via Backdoored Tool Use

Security study of data exfiltration attacks via backdoored tool-use LLM agents, presenting Back-Reveal attack with semantic triggers.

Ax Gwanghyun Kim, Junghun James Kim, Suh Yoon Jeon, Jason Park, Se Young Chun 27d ago

Human Interaction-Aware 3D Reconstruction from a Single Image

3D human reconstruction from single images in multi-person scenes with interaction awareness.

Ax Anes Abdennebi, Nadjia Kara, Laaziz Lahlou, Hakima Ould-Slimane 27d ago

LanG -- A Governance-Aware Agentic AI Platform for Unified Security Operations

Open-source governance-aware agentic platform for security operations, addressing alert fatigue and cross-source event correlation with LLM assistance.

Ax Qiyuan Chen, Hongsen Huang, Jiahe Chen, Qian Shao, Jintai Chen, Hongxia Xu, Renjie Hua, Chuan Ren, Jian Wu 27d ago

Learning What Matters: Dynamic Dimension Selection and Aggregation for Interpretable Vision-Language Reward Modeling

Vision-language reward model framework dynamically decomposing evaluation into interpretable dimensions via gating mechanism.

Ax Md Shamimul Islam, Luis G. Jaimes, Ayesha S. Dina 27d ago

MA-IDS: Multi-Agent RAG Framework for IoT Network Intrusion Detection with an Experience Library

Multi-agent RAG framework using agents for IoT network intrusion detection with experience library, improving interpretability over ML approaches.

Ax Jiachun Li, David Simchi-Levi, Will Wei Sun 27d ago

LLM Evaluation as Tensor Completion: Low Rank Structure and Semiparametric Efficiency

Statistical framework treating LLM evaluation as tensor completion problem, addressing uncertainty quantification in pairwise comparison leaderboards.

Ax Melika Sepidband, Hung Viet Pham, Hadi Hemmati 27d ago

On the Role of Fault Localization Context for LLM-Based Program Repair

Empirical study on fault localization's role in LLM-based automated program repair, evaluating context requirements across 500 SWE-bench instances.

Ax Pu Wang, Zhixuan Mao, Jialu Li, Zhuoran Zheng, Dianjie Lu, Youshan Zhang 27d ago

Unifying VLM-Guided Flow Matching and Spectral Anomaly Detection for Interpretable Veterinary Diagnosis

Diagnostic framework combining vision-language models with flow matching and spectral detection for veterinary pneumothorax diagnosis.

Ax Ljupcho Milosheski, Fedja Mo\v{c}nik, Mihael Mohor\v{c}i\v{c}, Carolina Fortuna 27d ago

Learned Elevation Models as a Lightweight Alternative to LiDAR for Radio Environment Map Estimation

Learned elevation models as alternative to LiDAR for radio environment map estimation in wireless networks.

Ax Zhetao Hu, Yiquan Zhou, Wenyu Wang, Zhiyu Wu, Xin Gao, Jihua Zhu 27d ago

Controllable Singing Style Conversion with Boundary-Aware Information Bottleneck

Singing voice conversion system using boundary-aware information bottleneck for fine-grained style control.

Ax Zhongxin Yang, Chun Bao, Yuanwei Bin, Xiang I. A. Yang, Shiyi Chen 27d ago

Turbulence-like 5/3 spectral scaling in contextual representations of language as a complex system

Analysis of transformer embedding trajectories exhibiting turbulence-like 5/3 power-law spectral scaling across languages.

Ax Dat Nguyen-Cong, Tung Kieu, Hoang Thanh-Tung 27d ago

FastDiSS: Few-step Match Many-step Diffusion Language Model on Sequence-to-Sequence Generation--Full Version

FastDiSS improves few-step diffusion language models for sequence-to-sequence generation by addressing self-conditioning approximation gaps.

Ax Junan Hu, Shudan Guo, Wenqi Liu, Jianhua Yin, Yinwei Wei 27d ago

Context-Agent: Dynamic Discourse Trees for Non-Linear Dialogue

Context-Agent framework using dynamic discourse trees for hierarchical non-linear dialogue management in LLMs.

Ax Jan Gruber, Jan-Niclas Hilgert 27d ago

Foundations for Agentic AI Investigations from the Forensic Analysis of OpenClaw

Empirical forensic analysis of OpenClaw agentic AI system, examining internal state reconstruction and action logging for digital investigations.

Ax N. D. Tantaroudas, A. J. McCracken, I. Karachalios, E. Papatheou 27d ago

AI-Driven Modular Services for Accessible Multilingual Education in Immersive Extended Reality Settings: Integrating Speech Processing, Translation, and Sign Language Rendering

Modular platform combining speech recognition, translation, emotion classification, and sign language rendering using open-source AI services.

Ax Nikolaos D. Tantaroudas, Andrew J. McCracken, Ilias Karachalios, Evangelos Papatheou 27d ago

INTERACT: An AI-Driven Extended Reality Framework for Accesible Communication Featuring Real-Time Sign Language Interpretation and Emotion Recognition

Extended reality framework integrating AI services for sign language interpretation and emotion recognition in video conferencing.

Ax Dustin Eisenhardt, Timothy Schauml\"offel, Alperen Kantarci, Gemma Roig 27d ago

Evaluation of Randomization through Style Transfer for Enhanced Domain Generalization

Study evaluating style transfer randomization for domain generalization in computer vision synthetic-to-real transfer.

Ax Chenyu Xue, Yiran Liu, Mian Zhou, Jionglong Su, Zhixiang Lu 27d ago

Semantic-Topological Graph Reasoning for Language-Guided Pulmonary Screening

Multimodal model for medical image segmentation guided by clinical text using semantic-topological graph reasoning.

Ax Peixi Peng (Institute of Medical Robotics, School of Biomedical Engineering, Shanghai Jiao Tong University), Housheng Xie (Institute of Medical Robotics, School of Biomedical Engineering, Shanghai Jiao Tong University), Yanling Wei (Daping Hospital, Army Medical University), Guangcong Ruan (Daping Hospital, Army Medical University), Xiaoyang Zou (Institute of Medical Robotics, School of Biomedical Engineering, Shanghai Jiao Tong University), Qian Cao (Sir Run Run Shaw Hospital, Zhejiang University School of Medicine), Yongjian Nian (Daping Hospital, Army Medical University), Guoyan Zheng (Institute of Medical Robotics, School of Biomedical Engineering, Shanghai Jiao Tong University) 27d ago

Analogical Reasoning as a Doctor: A Foundation Model for Gastrointestinal Endoscopy Diagnosis

Foundation model for gastrointestinal endoscopy diagnosis using analogical reasoning to improve generalizability and robustness.

Ax Prashant Kumar, Rajesh Ranjan 27d ago

Multiscale Physics-Informed Neural Network for Complex Fluid Flows with Long-Range Dependencies

Physics-informed neural networks for modeling multiscale fluid dynamics with long-range dependencies in Navier-Stokes equations.

Ax Lihao Sun, Hang Dong, Bo Qiao, Qingwei Lin, Dongmei Zhang, Saravan Rajmohan 27d ago

LLM Reasoning as Trajectories: Step-Specific Representation Geometry and Correctness Signals

Research characterizing LLM chain-of-thought reasoning as trajectories through representation space, showing step-specific subspaces become more separable with layer depth.

Ax Wuyang Luan, Junhui Li, Weiguang Zhao, Wenjian Zhang, Tieru Wu, Rui Ma 27d ago

SnapFlow: One-Step Action Generation for Flow-Matching VLAs via Progressive Self-Distillation

SnapFlow self-distillation method converting 10-step flow-matching VLA models to one-step action generation for real-time robotic manipulation.

Ax Wuyang Luan, Junhui Li, Weiguang Zhao, Wenjian Zhang, Tieru Wu, Rui Ma 27d ago

Rectified Schr\"odinger Bridge Matching for Few-Step Visual Navigation

Rectified Schrödinger Bridge matching approach for few-step visual navigation in embodied AI agents reducing denoising iterations.

Ax Shaofei Huang, Christopher M. Poskitt, Lwin Khin Shar 27d ago

From Incomplete Architecture to Quantified Risk: Multimodal LLM-Driven Security Assessment for Cyber-Physical Systems

Multimodal LLM-based security assessment method for cyber-physical systems with incomplete architectural documentation and legacy systems.

Ax Zhen Cheng, Hao-Bo Yang, Wan-Yi Huang, Jin-Long Li 27d ago

Attention Editing: A Versatile Framework for Cross-Architecture Attention Conversion

Framework for converting attention mechanisms between architectures (MLA, SWA) to reduce KV cache memory and bandwidth in LLM inference.

Ax Xuecong Liu, Mengzhu Ding, Zixuan Sun, Zhang Li, Xichao Teng 27d ago

CRFT: Consistent-Recurrent Feature Flow Transformer for Cross-Modal Image Registration

CRFT transformer-based framework using feature flow learning for robust cross-modal image registration in coarse-to-fine approach.

Ax Guan-Yan Yang, Wei-Ling Wen, Shu-Yuan Ku, Farn Wang, Kuo-Hui Yeh 27d ago

SemLink: A Semantic-Aware Automated Test Oracle for Hyperlink Verification using Siamese Sentence-BERT

SemLink tool using Siamese Sentence-BERT for semantic-aware automated test oracles detecting hyperlink rot and semantic drift in web applications.

Ax Jiaren Peng, Zeqin Li, Chang You, Yan Wang, Hanlin Sun, Xuan Tian, Shuqiao Zhang, Junyi Liu, Jianguo Zhao, Renyang Liu, Haoran Ou, Yuqiang Sun, Jiancheng Zhang, Yutong Jiao, Kunshu Song, Chao Zhang, Fan Shi, Hongda Sun, Rui Yan, Cheng Huang 27d ago

Hackers or Hallucinators? A Comprehensive Analysis of LLM-Based Automated Penetration Testing

Systematic analysis and benchmark comparing LLM-based automated penetration testing frameworks for autonomous security testing.

Ax Amit Vaisman, Gal Pomerants, Raz Lapid 27d ago

On the Robustness of Diffusion-Based Image Compression to Bit-Flip Errors

Analysis of diffusion-based image compressors' robustness to bit-flip errors compared to classical and learned codecs.

Ax Tim Lukas Adam, Phongsakon Mark Konrad, Riccardo Terrenzi, Florian Girardo Lukas, Rahime Yilmaz, Krzysztof Sierszecki, Serkan Ayvaz 27d ago

CAKE: Cloud Architecture Knowledge Evaluation of Large Language Models

CAKE benchmark with 188 expert-validated questions evaluating LLMs' understanding of cloud-native software architecture across Bloom's taxonomy levels.

Ax Joosung Lee, Hwiyeol Jo, Donghyeon Ko, Kyubyung Chae, Cheonbok Park, Jeonghoon Kim 27d ago

What Models Know, How Well They Know It: Knowledge-Weighted Fine-Tuning for Learning When to Say "I Don't Know"

Fine-tuning technique using instance-level knowledge scores to reduce LLM hallucinations by aligning pre-training and fine-tuning knowledge.

Ax Fernando L\'opez, Paula Delgado-Santos, Pablo G\'omez, David Solans, Jordi Luque 27d ago

"OK Aura, Be Fair With Me": Demographics-Agnostic Training for Bias Mitigation in Wake-up Word Detection

Demographics-agnostic training method for mitigating bias in wake-up word detection across diverse speaker populations.

Ax Panagiotis Andrikopoulos, Siamak Mehrkanoon 27d ago

EEG-MFTNet: An Enhanced EEGNet Architecture with Multi-Scale Temporal Convolutions and Transformer Fusion for Cross-Session Motor Imagery Decoding

EEG-MFTNet deep learning architecture combining multi-scale temporal convolutions and transformers for cross-session motor imagery decoding in BCIs.

Ax Junsoo Park, Youssef Medhat, Htet Phyo Wai, Ploy Thajchayapong, Ashok K. Goel 27d ago

Evaluating Learner Representations for Differentiation Prior to Instructional Outcomes

Representation-level evaluation metric for learner representations in educational AI systems measuring distinctiveness between students.

Ax Osama Orabi, Artur Zagitov, Hadi Salloum, Viktor A. Lobachev, Kasymkhan Khubiev, Yaroslav Kholodov 27d ago

Neural Network Pruning via QUBO Optimization

Neural network pruning formulated as QUBO optimization problem with principled objective formulations capturing filter interactions.

Ax Fatih Uenal 27d ago

Swiss-Bench 003: Evaluating LLM Reliability and Adversarial Security for Swiss Regulatory Contexts

Swiss-Bench 003 benchmark extending HAAS framework to evaluate LLM reliability and adversarial security in Swiss regulatory and financial contexts.

Ax Antonio D. Villegas-Yeguas, Xavier Abreau-Freire, Guillermo R-Garc\'ia, Andrea Valsecchi, Teresa Pinho, Daniel P\'erez-Mongiovi, Oscar Ib\'a\~nez, Oscar Cord\'on 27d ago

Automatic dental superimposition of 3D intraorals and 2D photographs for human identification

Method for automated dental superimposition comparing 3D intraoral scans and 2D photos for human identification in forensic contexts.

Ax Jungwon Park, Jungmin Ko, Dongnam Byun, Wonjong Rhee 27d ago

Selective Aggregation of Attention Maps Improves Diffusion-Based Visual Interpretation

Technique for improving text-to-image diffusion model interpretability through selective aggregation of cross-attention maps from relevant attention heads.

Ax Mamoona Ghafoor, Tatsuya Akutsu 27d ago

ReLU Networks for Exact Generation of Similar Graphs

Neural network method using ReLU networks for generating graphs constrained by specified graph edit distance for cheminformatics and data augmentation.

Ax Naen Xu, Jiayi Sheng, Changjiang Li, Chunyi Zhou, Yuyuan Li, Tianyu Du, Jun Wang, Zhihui Fu, Jinbao Li, Shouling Ji 27d ago

"I See What You Did There": Can Large Vision-Language Models Understand Multimodal Puns?

Benchmark evaluating vision-language models' ability to understand multimodal puns combining visual and textual elements.

Ax Jingbo Sun, Qichao Zhang, Songjun Tu, Xing Fang, Yupeng Zheng, Haoran Li, Ke Chen, Dongbin Zhao 27d ago

Saliency-Guided Representation with Consistency Policy Learning for Visual Unsupervised Reinforcement Learning

Successor representation method for zero-shot unsupervised RL in visual environments using saliency-guided representations and consistency policy learning.

Ax Kai Yu, Zhenhao Zhou, Junhao Zeng, Ying Wang, Xueying Du, Zhiqiang Yuan, Junwei Liu, Ziyu Zhou, Yujia Wang, Chong Wang, Xin Peng 27d ago

Does Pass Rate Tell the Whole Story? Evaluating Design Constraint Compliance in LLM-based Issue Resolution

Evaluation method for LLM-based issue resolution agents beyond pass rates, assessing compliance with implicit design constraints and architectural conventions.

Ax Nirajan Acharya, Gaurav Kumar Gupta 27d ago

A Formal Security Framework for MCP-Based AI Agents: Threat Taxonomy, Verification Models, and Defense Mechanisms

Formal security framework for MCP-based AI agents, including threat taxonomy, verification models, and defense mechanisms for tool-connected LLM systems.

Ax Xiaojie Gu, Ziying Huang, Weicong Hong, Jian Xie, Renze Lou, Kai Zhang 27d ago

The Model Agreed, But Didn't Learn: Diagnosing Surface Compliance in Large Language Models

Study on surface compliance in LLMs: models agree with knowledge edits but don't internalize changes, affecting reliability of edited parametric memory.

Ax Uloma Okoro, Tammy Mckenzie, Branislav Radeljic 27d ago

Governance and Regulation of Artificial Intelligence in Developing Countries: A Case Study of Nigeria

Qualitative case study examining legal professionals' perceptions on AI governance, regulatory gaps, and institutional readiness in Nigeria.

Ax Gustav Keppler, Moritz Gst\"ur, Veit Hagenmeyer 27d ago

CritBench: A Framework for Evaluating Cybersecurity Capabilities of Large Language Models in IEC 61850 Digital Substation Environments

CritBench: evaluation framework for cybersecurity capabilities of LLM agents in operational technology (OT) environments like IEC 61850 digital substations.

Ax Maria Mahbub, Gregory M. Dams, Josh Arnold, Caitlin Rizy, Sudarshan Srinivasan, Elliot M. Fielstein, Minu A. Aghevli, Kamonica L. Craig, Elizabeth M. Oliva, Joseph Erdos, Jodie Trafton, Ioana Danciu 27d ago

A Multi-Stage Validation Framework for Trustworthy Large-scale Clinical Information Extraction using Large Language Models

Multi-stage validation framework for trustworthy clinical information extraction using LLMs at scale without annotation-intensive reference standards.

Ax Ben Wigler, Maria Tsfasman, Tiffany Matej Hrkalovic 27d ago

Stories of Your Life as Others: A Round-Trip Evaluation of LLM-Generated Life Stories Conditioned on Rich Psychometric Profiles

Evaluation of LLM personality simulation using psychometric profiles and life story generation, comparing model outputs against real human psychological data.

Ax Junbin Zhang, Meng Cao, Feng Tan, Yikai Lin, Yuexian Zou 27d ago

Graph-PiT: Enhancing Structural Coherence in Part-Based Image Synthesis via Graph Priors

Framework using graph priors to improve structural coherence in part-based image synthesis by modeling spatial and semantic relationships.

Ax Juekai Lin, Yun Zhu, Honglin Lin, Sijing Li, Tianwei Lin, Zheng Liu, Xiaoyang Wang, Wenqiao Zhang, Lijun Wu 27d ago

Scientific Graphics Program Synthesis via Dual Self-Consistency Reinforcement Learning

Method using dual self-consistency reinforcement learning to synthesize TikZ graphics code from images, addressing precision challenges in multimodal LLM code generation.