ryan-superman/alfworld-teacher-sampling-measurements download history
ryan-superman/alfworld-teacher-sampling-measurements is a reinforcement learning dataset on the Hugging Face Hub. In the last 30 days it was downloaded 38 times (15 in the last 7 days), and 72 times in total. It ranks #227,732 among datasets by monthly downloads.
ALFWorld teacher-guided sampling measurements Raw measurement artefacts from a study of whether inference-time sampling methods (power sampling / best-of-n / teacher-tilted resampling) transfer from math and code to multi-turn LLM agents. Student: an OPD-distilled Qwen2.5-3B. Teacher:
Open ryan-superman/alfworld-teacher-sampling-measurements on Hugging Face
Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.