UltraFeedback, every mention
3 scenes · ← back to UltraFeedback
tap a year for its mentions
every year anyone Nathan Lambert 3
Verbatim, from the transcripts: the passages where UltraFeedback comes up
The RLVR Revolution — with Nathan Lambert (AI2, Interconnects.ai)
- ▶ 3:15 Nathan Lambert The academic community had been using this one data set since like all the way back in the hugging face models of like Zephyr beta is when this ultra feedback data set got popular.
- ▶ 11:54 Nathan Lambert And it, it's for example, like this ultra feedback that I mentioned is just has a lot of models in it.
The Origin and Future of RLHF: the secret ingredient for ChatGPT - with Nathan Lambert
- ▶ 29:20 Nathan Lambert Like everyone's using this ultra feedback data set and it boosts AlpacaVal, MTBench, TruthfulQA, and like the qualitative model a bit.