The Molecular Structure of Thought: Mapping the Topology of Long Chain-of-Thought Reasoning
This addresses a bottleneck in improving reasoning capabilities for AI systems, though it appears incremental as it builds on existing chain-of-thought fine-tuning approaches.
The paper tackles the problem of large language models failing to learn effective long chain-of-thought reasoning by proposing that stable molecular-like structures are key, and introduces Mole-Syn, a method that boosts performance and reinforcement learning stability across benchmarks.
Large language models (LLMs) often fail to learn effective long chain-of-thought (Long CoT) reasoning from human or non-Long-CoT LLMs imitation. To understand this, we propose that effective and learnable Long CoT trajectories feature stable molecular-like structures in unified view, which are formed by three interaction types: Deep-Reasoning (covalent-like), Self-Reflection (hydrogen-bond-like), and Self-Exploration (van der Waals-like). Analysis of distilled trajectories reveals these structures emerge from Long CoT fine-tuning, not keyword imitation. We introduce Effective Semantic Isomers and show that only bonds promoting fast entropy convergence support stable Long CoT learning, while structural competition impairs training. Drawing on these findings, we present Mole-Syn, a distribution-transfer-graph method that guides synthesis of effective Long CoT structures, boosting performance and RL stability across benchmarks.