ultrafeedback

Here are 2 public repositories matching this topic...

RLHF preference data curation pipeline: HH-RLHF + UltraFeedback + OASST1 → quality filter → MinHash dedup → DPO-ready JSONL

python minhash data-curation post-training dpo llm rlhf preference-data hh-rlhf ultrafeedback

Measuring verbosity bias in UltraFeedback (61K GPT-4-annotated preference pairs): statistically unambiguous, but a small effect. Python + Streamlit.

python data-quality streamlit rlhf rlaif preference-data llm-alignment ultrafeedback

To associate your repository with the ultrafeedback topic, visit your repo's landing page and select "manage topics."