yavuz-ai/self-reward-collapse-anchored download history

yavuz-ai/self-reward-collapse-anchored is a text generation dataset on the Hugging Face Hub. In the last 30 days it was downloaded 109 times (21 in the last 7 days), and 331 times in total. It ranks #107,861 among datasets by monthly downloads.

self-reward-collapse-anchored Per-round trajectory and held-out samples from an iterative self-training loop on GSM8K (Qwen2.5-7B-Instruct, LoRA DPO, 6 rounds). Same loop every arm; only the preference-label source differs. This dataset is the anchored arm: pairs labelled by the gold

Open yavuz-ai/self-reward-collapse-anchored on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.