SoftCoT++：基于软思维链推理的测试时扩展
SoftCoT++: Test-Time Scaling with Soft Chain-of-Thought Reasoning
AI总结：
SoftCoT++通过扰动潜在思维和对比学习，将SoftCoT扩展到测试时扩展范式，实现多样化思维路径探索，在多个推理基准上显著提升性能。
摘要：
测试时扩展（TTS）指通过在推理阶段分配额外计算来提升推理性能的方法，且不改变模型参数。现有TTS方法通常在离散词元空间中通过生成更多中间步骤来运作，但近期Coconut和SoftCoT的研究表明，在连续潜空间中思考可进一步增强推理性能。此类潜在思维编码了信息丰富的思考过程，避免了自回归词元生成带来的信息损失，从而引发了对连续空间推理的日益关注。与离散解码不同（离散解码可通过重复采样探索多样化的推理路径），连续空间中的潜在表示对于给定输入是固定的，这限制了多样化的探索，因为所有解码路径都源自同一潜在思维。为克服这一局限，我们提出SoftCoT++，将SoftCoT扩展至测试时扩展范式，通过实现思维路径的多样化探索。具体而言，我们通过多个专门的初始词元对潜在思维进行扰动，并应用对比学习以促进软思维表示之间的多样性。在五个推理基准和两种不同LLM架构上的实验表明，SoftCoT++显著提升了SoftCoT的性能，并优于采用自一致性扩展的SoftCoT。此外，它还与自一致性等传统扩展技术表现出强兼容性。源代码可在https://github.com/xuyige/SoftCoT获取。
英文摘要：
Test-Time Scaling (TTS) refers to approaches that improve reasoning performance by allocating extra computation during inference, without altering the model's parameters. While existing TTS methods operate in a discrete token space by generating more intermediate steps, recent studies in Coconut and SoftCoT have demonstrated that thinking in the continuous latent space can further enhance the reasoning performance. Such latent thoughts encode informative thinking without the information loss associated with autoregressive token generation, sparking increased interest in continuous-space reasoning. Unlike discrete decoding, where repeated sampling enables exploring diverse reasoning paths, latent representations in continuous space are fixed for a given input, which limits diverse exploration, as all decoded paths originate from the same latent thought. To overcome this limitation, we introduce SoftCoT++ to extend SoftCoT to the Test-Time Scaling paradigm by enabling diverse exploration of thinking paths. Specifically, we perturb latent thoughts via multiple specialized initial tokens and apply contrastive learning to promote diversity among soft thought representations. Experiments across five reasoning benchmarks and two distinct LLM architectures demonstrate that SoftCoT++ significantly boosts SoftCoT and also outperforms SoftCoT with self-consistency scaling. Moreover, it shows strong compatibility with conventional scaling techniques such as self-consistency. Source code is available at https://github.com/xuyige/SoftCoT.
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。
