6 papers
Practice Makes Unsafe: Skill Misevolution in Self-Improving LLM Agents
Xutao Mao, Liangjie Zhao, Xiang Zheng +1
LLM agents increasingly save experience from past tasks as reusable skills through skill evolution. When an agent completes an unsafe task, skill evolution can keep the unsafe proc…
Agent Hacks Agents: Autoresearch Discovers Vulnerabilities in Production Agents
Xutao Mao, Xiang Zheng, Rui Qian +1
Production LLM agents such as Claude Code and Codex can modify files and execute commands, so safety failures become real destructive actions. Automatic red-teaming lets safety tea…
What Happens Inside Agent Memory? Circuit Analysis from Emergence to Diagnosis
Xutao Mao, Jinman Zhao, Gerald Penn +1
Agent memory failures are silent: an LLM-based agent can produce a fluent response even when it fails to extract, retain, or retrieve the information needed across sessions. The wr…
STARE: Step-wise Temporal Alignment and Red-teaming Engine for Multi-modal Toxicity Attack
Xutao Mao, Liangjie Zhao, Tao Liu +3
Red-teaming Vision-Language Models is essential for identifying vulnerabilities where adversarial image-text inputs trigger toxic outputs. Existing approaches treat image generatio…
SteinerSQL: Graph-Guided Mathematical Reasoning for Text-to-SQL Generation
Xutao Mao, Tao Liu, Hongying Zan
Large Language Models (LLMs) struggle with complex Text-to-SQL queries that demand both sophisticated mathematical reasoning and intricate schema navigation. Existing methods often…
LogicCat: A Chain-of-Thought Text-to-SQL Benchmark for Complex Reasoning
Tao Liu, Xutao Mao, Hongying Zan +12
Text-to-SQL is a critical task in natural language processing that aims to transform natural language questions into accurate and executable SQL queries. In real-world scenarios, t…