Chi-Kuang Yeh
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
Stats ML×1ML×1Stats Comp.×1Stats Method.×1
Frequent co-authors
Research Timeline
2026
Quantifying and Auditing LLM Evaluation via Positive--Unlabeled Learning
The paper proposes a method to correct biased language models by aligning human-verified positives with unlabelled outputs in a fixed embedding space.
Highlighted terms show continued research focus across papers