5 papers
Trace-to-Architecture: Capability Bottleneck Diagnosis for Harness Revision
Wenfeng Wang, Jianrui Shi, Mingang Guo +5
Improving an LLM agent requires knowing what to change when it fails. Similar failures can require different fixes, and resolving one problem can expose another. We introduce Trace…
FPCO-Dialog: A Multi-Turn False-Premise Benchmark for Correction and Cooperation in Vision-Language Models
Jiayuan Ma, Yuqi Lu, Weiyang Guo +5
Vision-language models (VLMs) are increasingly deployed in multi-turn settings where users may describe visual content with incorrect assumptions. Yet existing evaluations rarely i…
E3-TIR: Enhanced Experience Exploitation for Tool-Integrated Reasoning
Weiyang Guo, Zesheng Shi, Liye Zhao +5
While Large Language Models (LLMs) have demonstrated significant potential in Tool-Integrated Reasoning (TIR), existing training paradigms face significant limitations: Zero-RL suf…
Lost in Pronunciation: Detecting Chinese Offensive Language Disguised by Phonetic Cloaking Replacement
Haotan Guo, Jianfei He, Jiayuan Ma +8
Phonetic Cloaking Replacement (PCR), defined as the deliberate use of homophonic or near-homophonic variants to hide toxic intent, has become a major obstacle to Chinese content mo…
Detecting Conversational Mental Manipulation with Intent-Aware Prompting
Jiayuan Ma, Hongbin Na, Zimu Wang +4
Mental manipulation severely undermines mental wellness by covertly and negatively distorting decision-making. While there is an increasing interest in mental health care within th…