- Rating
- 1
- Confidence
- 4
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 1
Read review
Paper summary
The paper focuses on the task of hallucination detection for LLMs. The authors proposed single forward pass label-free method that is based only on the hidden states of the model. The method is motivated by the geometry of the hidden states, with the finding that hallucinated and correct answers differ by a location shift. Finally, the authors provide results on 4 datasets and 3 models for both in-domain and out-of-domain setups.
Critical strengths
- The scope of the paper is well defined in the text, and the research direction is important. The method seems quite novel.
- The proposed method shows competitive performance compared to more complicated methods and even outperforms them without any label tuning.
Critical weaknesses
- My main concern is the importance of the individual components of the method. In particular, the norm is highly competitive and in some cases even outperforms the full method. This norm baseline (arguably the most important component) reduces to the distance between a test embedding and the median embedding of the training set. According to Table 7, the datasets contain only a small portion (<50% in most cases) of hallucinations, so the median should lie close to the mean of the correct-answer class, which dominates the sample. In other words, the method may largely measure how far an answer is from the typical correct one. This is supported by the AUC, which correlates strongly with the percentage of hallucinations across setups. Therefore, it is unclear whether the performance is a real advantage of the method or an artifact of the models, datasets, and choice of ROC-AUC as the metric. It is thus important to add a detailed ablation table covering all datasets and models, together with the average score, and to report other metrics, such as PR-AUC or PRR [1]. Additionally, since robust scaling seems to be so important, it would be useful to add further ablations: replacing with a random direction, an all-ones direction, or the first principal component, and using just the distance to the median as the final score.
- The experiment on transfer across benchmarks is questionable: it seems to require the source benchmark’s scale vector . If we can estimate from unlabelled states, we can estimate from the same states. I would suggest adding transfer results without any scaling. Also, following from weakness 1, I think it is important to add a transfer experiment from a dataset in which hallucinations are a minority to a dataset in which they are a majority.
- While the method is label-free, it still requires data. So the method still needs hundreds of in-distribution hidden states from the same model according to Table 13.
- The paper overlooks several important baselines: single-forward-pass methods, e.g., seq.prob [1], RAUQ [2], which are directly comparable to the proposed method, and more advanced sampling-based methods [3,4].
- Minor issues:
- I don't think it is fair to flip the sign of the score based on the orientation of ROC-AUC for each dataset. The sign should be fixed across all datasets, so please report raw AUC for all methods.
- Notation issues: is not defined in Section 2.2.
- Figure 1 is quite blurry and of low quality.
- The abbreviation HPSA is never explained.
[1] Benchmarking Uncertainty Quantification Methods for Large Language Models with LM-Polygraph (Vashurin et al., TACL 2025)
[2] Efficient Hallucination Detection for LLMs Using Uncertainty-Aware Attention Heads. (A. Vazhentsev et al., ICML 2026)
[3] Semantic Density: Uncertainty Quantification for Large Language Models through Confidence Measurement in Semantic Space. (Qiu et al., NeurIPS 2024)
[4] Uncertainty Quantification for LLMs through Minimum Bayes Risk: Bridging Confidence and Consistency (Vashurin et al., NeurIPS 2025)
Questions for authors
See weaknesses