TRAP: Tool-result Replay Attack for Probing Misbinding in LLM Agents

Yutong Zhi et al.

#arXiv #cs.AI

ScreenASR-Bench: A Tiered Benchmark for Speech Recognition with Visual Context

Mingwei Fu et al.

#arXiv #cs.AI

Large Language Model Assisted Transportation Planning through a Differentiable User Equilibrium Layer

Eunseung Lee, Gyeongjun Kim, Keemin Sohn

#arXiv #cs.AI

Revocation Is Not Complete at Receipt: An Executable Outcome Contract for Distributed Agent-Tool Gateways

Lifei Liu, Haoran Yu, Danping Zhang

#arXiv #cs.AI #cs.CR

LURE: a Phase-Labelled Synthetic Corpus of Online-Grooming Exchanges

Emanuele Florindi et al.

#arXiv #cs.AI

MedHarnessBench: A Unified Benchmark for Healthcare Agent Evaluation

Rui Tang, Hongqiu Wang

#arXiv #cs.AI

Radioactive Feedback: Tracing Models Trained by Label-Only LLM Supervision

Yuejia Li et al.

#arXiv #cs.AI

Knowledge Interaction in Hypernetwork-based Continual Learning: Conflict and Composition

Jiaqi Zheng et al.

#arXiv #cs.AI

MemeLogos: Knowledge-Driven Retrieval-Augmented Reasoning for Harmful Meme Detection

Hexiang Gu et al.

#arXiv #cs.AI

SpecFit: Not All Draft Tokens Cost the Same in Offloaded MoE Inference

Hongliang Cao et al.

#arXiv #cs.AI

Analyzing the Dynamics of WSBM-based Echo State Networks using InputDSA

Huy Mai, Robert Clarke

#arXiv #cs.AI

Understanding Unified VLMs: From Attention Probing to Guidance Tuning for Multimodal Generation

Zhengyang Luo et al.

#arXiv #cs.AI

Continuous Risk-Based Scheduling for Food Safety Inspection: Integrating Historical Decay with Event-Driven Consumer Review Signals

Akriti Bhat, Wayne Zhu, Ishan Gupta

#arXiv #cs.AI #cs.CY #cs.MA

SkillForgeBench: Evaluating Agent Reliability Grounded in Agent Skill Specifications

Zhengxi Li et al.

#arXiv #cs.AI #cs.CL

ASIWorld: Benchmarking Autonomous Agents in a Persistent Survival Sandbox

Jianfeng Lin et al.

#arXiv #cs.AI

GateSight: Recurrent Memory Anticipates Attention Memory for Query-Agnostic KV Eviction in Hybrid LLMs

Junjie Li, Jiong Lou, Jie Li

#arXiv #cs.AI

arXiv paper "LabInstruct" benchmarks situated instructional video generation for lab procedures, aiming to help AI plans guide human operators in complex self-driving laboratory settings where fine-grained manipulation is hard to…

#Robotics #AI #Automation #ArXiv
https://arxiv.org/abs/2610.14020

A new arXiv paper compares human analysts and nine LLMs on two reverse-engineering tasks in IDA Pro, with top agents like GPT-5.5 Pro and Claude Opus 4.8 reaching statistically human-competitive understanding. The authors…

#AI #ReverseEngineering #DevTools #arxiv
https://arxiv.org/abs/2610.14960

A Hybrid Kinematic GCN and Elastic Transformer for Efficient Indian Sign Language Recognition

Pushp Garg, Himangshu Sarma, Mrinmoy Ghorai

#arXiv #cs.AI #cs.CV

ScienceBuddy-PV: A Verifier-Governed Recursive Improvement Protocol for Panvascular Intervention Research

Lingsen You et al.

#arXiv #cs.AI #cs.CV