EP322: Why cliff tokens break AI math

Title: Cliff Tokens: Identifying Single-Token Failure Triggers in LLM Mathematical ReasoningSource: http://arxiv.org/abs/2606.25524v1Summary:This paper identifies 'cliff tokens' as the exact single-token triggers that cause large language models to diverge into reasoning failures during multi-step mathematical tasks. By introducing a taxonomy of these failures and a targeted preference optimization method (Cliff-DPO), it establishes a foundational approach to diagnosing and improving LLM reasoning reliability.