atfortes/Awesome-LLM-Reasoning

A curated collection of papers and resources covering reasoning techniques in LLMs and multimodal systems, from foundational prompting methods to advanced inference-time scaling approaches. Organizes research by technique categories—including unimodal reasoning, multimodal reasoning, and scaling smaller models—alongside analysis studies on CoT faithfulness, token bias, and reasoning stability. Complements a companion benchmarking project (LLMSymbolicReasoningBench) for empirically testing symbolic reasoning capabilities.

3,558 stars. No commits in the last 6 months.