Judy Hanwen Shen (@heyyjudes.bsky.social)

How do you know when one preference dataset is better than another? Come check out our work: Towards Data-Centric RLHF: Simple Metrics for Preference Dataset Comparison ATTRIB Workshop @ Neurips West Room 205-207 Sat 12:30-2 pm www.arxiv.org/abs/2409.09603

loading . . .

Towards Data-Centric RLHF: Simple Metrics for Preference Dataset Comparison The goal of aligning language models to human preferences requires data that reveal these preferences. Ideally, time and money can be spent carefully collecting and tailoring bespoke preference data t... https://www.arxiv.org/abs/2409.09603