Diagnosing and Correcting Reward Hacking in Verifiable Instruction Following Paper • 2610.19525 • Published 5 days ago • 13
Diagnosing and Correcting Reward Hacking in Verifiable Instruction Following Paper • 2610.19525 • Published 5 days ago • 13
sangyon/llama31_tulu3_8b_dpo_grpo_nonthink_intentcheck_tulu3_8b_dpo Text Generation • 8B • Updated Sep 9 • 32