Benchmark
UC-Bench
A human-annotated benchmark dataset designed to evaluate the ability of large language models to detect implicit user-side conflicts in human–computer dialogues.
Timeline (last 90 days)
Sep 18
UC-Bench is a human-annotated benchmark specifically designed to evaluate the ability to detect user-side conflicts.
Unverified50%
Sep 18
On UC-Bench, Qwen3.5-4B trained on UC-Data outperformed larger general-purpose models such as Claude Opus 4.8.
Unverified50%