通过隐空间推理路径实现大模型无需输出的自我评估
Latent Space Chain-of-Embedding Enables Output-free LLM Self-Evaluation
- 利用推理过程中的隐状态序列构建思维路径，捕捉正确与错误回答的差异
- 在七种大模型上验证有效，毫秒级延迟支持实时反馈
- 无需训练、无需标签，适合大规模部署场景
大模型自我评估依赖于模型自身对回答正确性的判断，可显著提升部署可靠性。本文提出在隐空间中构建链式嵌入（Chain-of-Embedding, CoE），使大模型实现无需输出的自我评估。CoE由推理过程中产生的所有渐进隐藏状态构成，可视作大模型的隐式思考路径。我们发现，当模型回答正确或错误时，其CoE特征存在明显差异，这些差异可用于估计回答正确性。在四个不同领域和七种大模型上的实验充分验证了该方法的有效性。其无标签、免训练设计及毫秒级计算成本，确保了在大规模场景下的实时反馈。更重要的是，本研究从大模型内部隐藏状态变化角度，提供了关于回答正确性的新洞察。
原文摘要 · Abstract (English)
LLM self-evaluation relies on the LLM's own ability to estimate response correctness, which can greatly improve its deployment reliability. In this research track, we propose the Chain-of-Embedding (CoE) in the latent space to enable LLMs to perform output-free self-evaluation. CoE consists of all progressive hidden states produced during the inference time, which can be treated as the latent thinking path of LLMs. We find that when LLMs respond correctly and incorrectly, their CoE features differ, these discrepancies assist us in estimating LLM response correctness. Experiments in four diverse domains and seven LLMs fully demonstrate the effectiveness of our method. Meanwhile, its label-free design intent without any training and millisecond-level computational cost ensures real-time feedback in large-scale scenarios. More importantly, we provide interesting insights into LLM response correctness from the perspective of hidden state changes inside LLMs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。