- Rating
- 1
- Confidence
- 4
- Informativeness
- 1
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper conducts a benchmarking study of LLM uncertainty quantification (UQ) methods across four datasets with three models. The three models are open-weight models, and the datasets cover text generation in the context of RAG, short-form answers, and long-text answers. They focus on standard UQ methods (black-box, white-box, reflexive) and implement their experiments using LM-polygraph. They find mixed results across models, datasets, and UQ methods, similar to findings in prior studies.
Critical strengths
-
Readability: The paper is very well-written. It is entirely human-readable, and I did not need to re-read any part of the paper due to lack of clarity of writing.
-
Reproducibility: The paper implements experiments using an open-source library (LM-polygraph), open-weights models, and publicly available datasets. All necessary reproduction details are provided in the paper and easy to understand.
Critical weaknesses
The paper has several important weaknesses:
- Insufficient breadth for a benchmarking study: This study is not offering any new methods, and is primarily a benchmarking study. This is not itself a problem, but for an ICLR-caliber benchmarking study, one would expect far broader breadth and scope coverage than what this paper covers. This concern applies to several aspects of the paper's scope:
- No evaluation of supervised ensembles: Although they do not evaluate supervised UQ ensembles, which have been shown to generally outperform individual scorer baselines (Bakman et al., ACL 2025; Bouchard et al., TMLR 2025) and have greater robustness across datasets and evaluation settings (Chauhan et al., AACL 2026)
- Insufficient model and dataset breadth: The authors evaluate UQ on only four datasets and three models. They motivate this by claiming to require human annotations made available by the RAGTruth dataset with the three open-weight models they used. However, obtaining faithfulness-style ground-truth labels can be easily obtained with a competent LLM in the same way that long-form labels are obtained in LongWiki. Using an strong LLM to obtain RAGTruth labels and validating with human review on a reasonable sample would enable use of additional models, which is absolutely necessary for a benchmarking study.
- Long-form setting missing well-established long-form scorers: Only one long-text adapted (claim-level) scorer (LUQ) is used on the long-text dataset, despite well-established literature on this, and it is used incorrectly. Per appendix B.6.2, as the method is designed to operate claim-wise or sentence-wise and check entailment of claims/sentences of original response against full sampled responses, but the authors instead compute full-response comparisons. The name of the scorer is Long-Text Uncertainty Quantification (LUQ), hence the claim/sentence-wise scoring, it was designed to compute. For other well-established methods designed for long-text UQ, see claim-level graph-based scorers (Jiang et al., NeurIPS 2024), long-form semantic entropy (Farquhar et al., Nature 2024), atomic calibration (Zhang et al., AACL 2025), and generalizations of these (Bouchard et al., TMLR 2026). Further, the authors use LUQ incorrectly,
-
Insufficient motivation for the paper and lack of positioning relative to prior work: How is this paper different from other benchmarking studies such as Vashurin et al., TACL 2025; Bouchard et al., TMLR 2025; Bakman et al., ACL 2025? This appears to evaluate the same scorers as in LM-Polygraph (Vashurin et al., TACL 2025) but uses far fewer models and datasets than their TACL study. It also misses the claim-level and code generation settings in the recent benchmarking study evaluated by Chauhan et al. (AACL 2026).
-
Main results missing from main text There is no results table in main text or exhaustive list of which specific scorers are used in the study, only scorer families.
References:
- Reconsidering LLM Uncertainty Estimation Methods in the Wild. Bakman et al., ACL 2025: https://aclanthology.org/2025.acl-long.1429/
- Uncertainty Quantification for Language Models: A Suite of Black-Box, White-Box, LLM Judge, and Ensemble Scorers. Bouchard et al., TMLR 2025: https://openreview.net/forum?id=WOFspd4lq5
- When Do Supervised UQ Ensembles Improve LLM Hallucination Detection? A Robustness Study. CHauhan et al., AACL 2026. https://arxiv.org/abs/2608.24492
- Benchmarking Uncertainty Quantification Methods for Large Language Models with LM-Polygraph. Vashurin et al., TACL 2025: https://aclanthology.org/2025.tacl-1.11/
- Uncertainty Propagation on LLM Agent. Zhao et al., ACL 2025 : https://aclanthology.org/2025.acl-long.302/
- UProp: Investigating the Uncertainty Propagation of LLMs in Multi-Step Agentic Decision-Making: Duan et al., 2025: https://arxiv.org/abs/2506.17419
- Uncertainty Calibration for Tool-Using Language Agents. Liu et al. EMNLP Findings 2024: https://aclanthology.org/2024.findings-emnlp.978/
- Graph-based Uncertainty Metrics for Long-form Language Model Generations. Jiang et al., NeurIPS 2024: https://proceedings.neurips.cc/paper_files/paper/2024/hash/39c5871aa13be86ab978cba7069cbcec-Abstract-Conference.html
- Detecting hallucinations in large language models using semantic entropy. Farquhar et al., Nature 2024: https://www.nature.com/articles/s41586-024-07421-0
- Atomic Calibration of LLMs in Long-Form Generations. Zhang et al., AACL-IJCNLP 2025: https://aclanthology.org/2025.findings-ijcnlp.9/
- Fine-Grained Uncertainty Quantification for Long-Form Language Model Outputs: A Comparative Study. Bouchard et al., TMLR 2026: https://openreview.net/forum?id=gngp4Zz9Sj
Note: some of the provided references are quite recent and provided only to help authors in future versions of the paper. I do not hold this against the authors for not including them in their evaluation.
Questions for authors
-
why not present results table in main text?
-
how do you distinguish this study from other benchmarking studies that have studied the same question with the same scorers?
-
why not consider code generation, agentic settings, or more claim level scorers for long text settings?
-
I would encourage the authors to position the paper relative to the papers mentioned in weaknesses to further clarify how their work differs and compare this work to those.
2 comments
Narrow scope, no long-text scorers, insufficient positioning
Thank you for your replies. While I appreciate the author's clarifying comments on their work, I do not believe the breadth/scope is sufficient to meet the bar for an ICLR-caliber empirical/benchmarking paper.
I am choosing to keep my original score for the following reasons:
- text generation only (no code generation or agent settings)
- no supervised/ensembled uncertainty scorers, which has been shown in the literature to perform strongly across settings, in contrast to individual, unsupervised UQ scorers
- no proper long-text scorer despite using long-text datasets and existence of a well-established long-text UQ literature
- insufficient positioning relative to existing literature and general lack of novel contribution
We thank the reviewer for their constructive, detailed feedback and for pointing us to relevant studies. We address the four questions below, combining Q2 and Q4.
Q1. Why not present a results table in the main text?
We chose figures because our main message concerns variation across evaluation settings, rather than a ranking of estimators. Figure 1 shows each estimator's mean AUROC against its ranking variability, illustrating how average performance can hide substantial differences across settings. The complete results are available in Appendix D.1 (Tables 6–9). We also consider explicitly naming all evaluated estimators in §2, where we introduce representative estimators by family.
Q2 & Q4. How does our study differ from existing benchmarks?
We study how uncertainty estimators align with explicitly defined hallucination criteria, and how this relationship changes with the evaluation criterion. We agree that this focus should be positioned more clearly relative to previous benchmarks.
Bouchard and Chauhan (2025) and Chauhan et al. (2026) study supervised ensembles of uncertainty scores. Their focus is on combining estimators, whereas we evaluate individual methods. Ensembling is a natural extension, not our present objective.
Bouchard et al. (2026), Jiang et al. (2024), and Zhang et al. (2025) focus on fine-grained uncertainty estimation for long-form generation, particularly at the sentence or claim level. Our comparison uses query- and response-level estimators and response-level evaluation targets.
Bakman et al. (2025) investigate practical deployment conditions while keeping the correctness criterion fixed. They study sensitivity to decision thresholds, input transformations, long-form adaptation, and ensembling. In contrast, we focus on variation in the evaluation target, including changing the error definition on the same responses.
Vashurin et al. (2025, LM-Polygraph) are the closest to our study. We use their library and many of the same estimators, while their benchmark covers more tasks and models. The main distinction lies in the evaluation targets. Their benchmark primarily uses general response-quality measures: AlignScore against reference answers, accuracy, COMET, and ROUGE-L against reference summaries. It also evaluates hallucination detection through claim-level fact-checking of generated biographies. We do not claim to introduce hallucination detection into UQ benchmarking.
Our difference lies in how we operationalize and compare hallucination criteria. Following Bang et al. (2025), we consider responses unsupported by the supplied context (RAGTruth, with human span annotations) and responses unsupported by knowledge expected from pretraining (HalluLens). We examine short- and long-form generation, as well as failure to refuse false premises. We also investigate how rankings change with the error definition and severity while keeping responses fixed (Appendix D.6–D.7). Another difference is that they use base models for white-box estimation while we consider instruction-tuned models.
Thus, our contribution is not broader benchmark coverage, but a focused investigation of the relationship between uncertainty and hallucination under different evaluation criteria, using query- and response-level estimators rather than claim-level scorers or supervised ensembles.
We acknowledge the scope could have been more explicit.
Q3. Why not consider code generation, agentic settings, or more claim-level estimators?
Code generation and agentic settings involve different failure modes and evaluation requirements, such as execution correctness or multi-step decisions. They fall outside our present scope, which focuses on hallucinations in text generation.
Regarding LUQ, the reviewer is correct. We use the simplified LM-Polygraph implementation, which computes entailment between complete responses. The original LUQ (Zhang et al., 2024) instead evaluates the entailment of individual sentences against sampled responses. We may rename the methods and clearly distinguish it from the original method.