Share your thoughts, 1 month free Claude Pro on usSee more

Home/Benchmarks

Instruction Tuning on Anthropic HH (test)

56.3Win Rate

SCAR

Updated 1mo ago

Evaluation Results

Method	Links
SCAR 2025.12		56.3
SCAR 2025.12		54.9

SOTA Paper

SCAR

Towards better dense rewards in Reinforcement Learning Applications

Dataset

Anthropic-HH

Follow for update

@wizwand_team Discord

Related Benchmarks

Text Summarization on TL;DR (test)Instruction Tuning on Anthropic HH-RLHF (test)

© 2026 wizwand

Blog Contact Changelog Swarm

Privacy Policy Terms of Service FAQs Swarm Docs