Latent Space Chain-of-Embedding Enables Output-free LLM Self-Evaluation

Yiming Wang, Pei Zhang, Baosong Yang, Derek F. Wong, Rui Wang

Keywords: Large Language Models, Label-free Self-Evaluation, AI Reliability

TL;DR: We propose the Chain-of-Embedding method for LLM self-evaluation, which enables output-free response correctness estimation during inference time.

Abstract:

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 ensure 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.

Primary Area: alignment, fairness, safety, privacy, and societal considerations

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 255