4 papers
CypherTurn: A Multi-Turn Benchmark for Conversational Text-to-Cypher Evaluation and the Autonomy Divergence
Yuzhe Zhang, Weijie Zhu, Haolin Yang +5
Graph databases are increasingly queried through natural language, yet every existing benchmark evaluates isolated single-turn queries rather than the multi-turn sessions through w…
ExpThink: Experience-Guided Reinforcement Learning for Adaptive Chain-of-Thought Compression
Tingcheng Bian, Yuzhe Zhang, Jing Jin +5
Large reasoning models (LRMs) achieve strong performance via extended chain-of-thought (CoT) reasoning, yet suffer from excessive token consumption and high inference latency. Exis…
Silo-Bench: A Scalable Environment for Evaluating Distributed Coordination in Multi-Agent LLM Systems
Yuzhe Zhang, Feiran Liu, Yi Shan +8
Large language models are increasingly deployed in multi-agent systems to overcome context limitations by distributing information across agents. Yet whether agents can reliably co…
Don't Act Blindly: Robust GUI Automation via Action-Effect Verification and Self-Correction
Yuzhe Zhang, Xianwei Xue, Xingyong Wu +8
Autonomous GUI agents based on vision-language models (VLMs) often assume deterministic environment responses, generating actions without verifying whether previous operations succ…