reasoning-degeneration-dev/

t1-structured-reasoning-user-full-together_ai-qwen-qwen3-next-80b-a3b-thin-a18d7696

Structured reasoning evaluation: instead of injecting synthesized facts, this uses a static system prompt that teaches the model a heuristic search FORMAT with explicit structural markers ([STEP], [PRUNE], [BACKTRACK], [REVIEW OPTIONS], [SOLUTION FOUND]). Inspired by HandCraftedCountdownSearch — models SFT'd on structured search traces significantly outperform free-form CoT. This tests whether… See the full description on the dataset page: https://huggingface.co/datasets/reasoning-degeneration-dev/t1-structured-reasoning-user-full-together_ai-qwen-qwen3-next-80b-a3b-thin-a18d7696.