UXBench
UXBench: Benchmarking User Experience in AI Assistants UXBench is the first LLM benchmark grounded in real user satisfaction signals — built from 70K+ thumbs-up / thumbs-down conversations from Tencent Yuanbao.
How to run
$ npm i -g lettokenburn-cli$ ltb bench info uxbench
External harness. No Harbor adapter has been generated for this benchmark yet. `ltb bench info` prints the known links.
Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs
Scores
0 sourced entriesNo sourced scores recorded yet.
Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.
Reproduced by the community
All runsAbout
UXBench: Benchmarking User Experience in AI Assistants
UXBench is the first LLM benchmark grounded in real user satisfaction signals — built from 70K+ thumbs-up / thumbs-down conversations from Tencent Yuanbao. It evaluates LLMs across three UX tasks covering 8 interaction scenarios and 83 domains.
Dataset Files
File Task Split Size Description
uxbench_task1_judge_bad_1k.jsonl Task 1 · UX Judge BAD 1,000 Disliked conversations (10 failure dimensions)…