Recent Papers / arXiv:2512.16310
Agent Tools Orchestration Leaks More: Dataset, Benchmark, and Mitigation
Yuxuan Qiao, Dongqin Liu, Hongchang Yang, Wei Zhou, Songlin Hu
Abstract
TOP-Bench measures compositional privacy leakage from tool returns; average leakage rate 88.6% across six LLM agents.
TOP-Align (SFT+DPO) improves H-score by 16.2 points.
Tasks
editResults
No benchmark results recorded yet.
Benchmark results referencing this paper haven't been added to the registry yet. If you have a reproduction, submit it →
CodeSOTA extraction
Benchmark evidence
Link this paper to benchmark rows, datasets, model cards, and reproduced results as evidence is extracted.