Kawin Ethayarajh
PulseAugur coverage of Kawin Ethayarajh — every cluster mentioning Kawin Ethayarajh across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
New KTO method aligns LLMs using prospect theory, matching DPO performance
Researchers have introduced KTO (Kahneman-Tversky Optimization), a novel method for aligning large language models with human feedback. KTO is based on prospect theory, a framework developed by Kahneman and Tversky that…
-
New RIPA measure offers improved assessment of word embedding bias
A new paper published on arXiv introduces RIPA, a novel measure for assessing undesirable associations in word embeddings. The research demonstrates that common debiasing techniques, like subspace projection, can be equ…
-
New paper questions standard human evaluation methods for NLG models
A new paper published on arXiv critiques the standard protocol for human evaluation of natural language generation (NLG) systems. The authors argue that common practices, particularly the use of Likert scales, can lead …