SIM-CoT：监督式隐式思维链
SIM-CoT: Supervised Implicit Chain-of-Thought
AI总结：
针对隐式思维链训练不稳定问题，提出SIM-CoT模块，通过步骤级监督稳定潜在推理空间，提升准确性和稳定性，并超越显式基线。
摘要：
隐式思维链（Implicit Chain-of-Thought, CoT）方法为大语言模型（LLMs）中的显式思维链推理提供了一种令牌高效的替代方案，但持续存在的性能差距限制了其应用。我们发现了在扩展隐式思维链计算预算时的一个核心潜在不稳定问题：随着推理令牌数量的增加，训练常常变得不稳定并崩溃。我们的分析表明，这种不稳定源于潜在表示变得同质化并失去语义多样性，这是由当前隐式思维链方法中步骤级监督不足导致的。为解决这一问题，我们提出了SIM-CoT，一种即插即用的训练模块，引入步骤级监督以稳定并丰富潜在推理空间。SIM-CoT在训练期间使用一个辅助解码器，将每个隐式令牌与其对应的显式推理步骤对齐，确保潜在状态捕获独特且有意义的信息。辅助解码器在推理时被移除，保持隐式思维链的效率且无额外开销。它还可通过将每个潜在令牌投影到显式推理词汇表上提供可解释性，实现逐步可视化和诊断。SIM-CoT显著提高了隐式思维链方法的域内准确性和域外稳定性，在GPT-2上将Coconut提升了+8.2%，在LLaMA-3.1 8B上将CODI提升了+3.0%。它进一步在GPT-2上以2.3倍的令牌效率超越了显式思维链基线2.1%，同时缩小了在更大模型（如LLaMA-3.1 8B）上的性能差距。代码：https://github.com/InternLM/SIM-CoT
英文摘要：
Implicit Chain-of-Thought (CoT) methods offer a token-efficient alternative to explicit CoT reasoning in Large Language Models (LLMs), but a persistent performance gap has limited their adoption. We identify a core latent instability issue when scaling the computational budget of implicit CoT: as the number of reasoning tokens increases, training often becomes unstable and collapses. Our analysis shows that this instability arises from latent representations becoming homogeneous and losing semantic diversity, caused by insufficient step-level supervision in current implicit CoT methods. To address this, we propose SIM-CoT, a plug-and-play training module that introduces step-level supervision to stabilize and enrich the latent reasoning space. SIM-CoT employs an auxiliary decoder during training to align each implicit token with its corresponding explicit reasoning step, ensuring latent states capture distinct and meaningful information. The auxiliary decoder is removed at inference, preserving the efficiency of implicit CoT with no added overhead. It also provides interpretability by projecting each latent token onto an explicit reasoning vocabulary, enabling per-step visualization and diagnosis. SIM-CoT significantly improves both in-domain accuracy and out-of-domain stability of implicit CoT methods, boosting Coconut by +8.2\% on GPT-2 and CODI by +3.0\% on LLaMA-3.1 8B. It further surpasses the explicit CoT baseline on GPT-2 by 2.1\% with 2.3$\times$ greater token efficiency, while closing the performance gap on larger models like LLaMA-3.1 8B. Code: https://github.com/InternLM/SIM-CoT
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
