Arena-Hard

A high-quality DPO derivative of LDJnr's Capybara, with chosen/rejected pairs synthesized and rated using Argilla's distilabel pipeline.

NVIDIA's permissively-licensed human-annotated preference dataset with 5-axis Likert ratings - engineered to train high-quality reward models.

A self-synthesis method (and family of datasets) that elicits high-quality instructions directly from an aligned LLM using only its chat template - no seed prompts required.

Berkeley NEST's seven-way ranked preference dataset built from GPT-4 rankings over responses from a diverse model pool, used to train Starling.

OpenBMB's 64k-prompt preference dataset built with GPT-4 critiques across instruction-following, truthfulness, honesty, and helpfulness - the de facto open DPO baseline.

1M real user-chatbot conversations collected by Allen AI / UW via a free GPT proxy - a window into how real users actually prompt LLMs.

LMSYS's automated pipeline for distilling high-quality LLM benchmarks from crowdsourced chat data (e.g. Chatbot Arena, WildChat), producing the Arena-Hard-Auto benchmark.

Introduces Arena-Hard, a 500-prompt benchmark auto-curated from Chatbot Arena traffic that correlates ~0.9 with Arena Elo using LLM-as-a-Judge.

Introduces Arena-Hard, a 500-prompt benchmark auto-curated from Chatbot Arena traffic that correlates ~0.9 with Arena Elo using LLM-as-a-Judge.