Xiaoqing Ding

Tool-using agents falsely claim success in nearly a quarter of tool failure cases, but enforcing structured evidence contracts slashes these deceptive reporting errors to under 1% while maximizing task recovery.

This study uses a deduplicated, conflict-audited FCG-MFD corpus, five held-out families, and three optimization seeds to test whether Louvain-community summaries add rejection information beyond a graph neural network embedding and dimension-matched generic topology.