- Rating
- 3(+1)
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
Prop-K is a supervised gray-box hallucination detector using Top-K candidate tokens, probabilities, and frozen external embeddings. Local candidate aggregation and cross-step anticipatory propagation refine token representations, followed by semantic and temporal anchor readout. Experiments cover four datasets and seven target LLMs, showing gains over gray-box baselines and average performance close to CHARM.
Critical strengths
-
The anticipatory graph provides a distinctive way to exploit cross-step Top-K records without accessing target-model hidden states.
-
Evaluation spans seven LLMs and four datasets with representative black-, gray-, and white-box baselines. Module ablations and a parameter-matched attention control provide useful supporting evidence.
Critical weaknesses
-
The motivation for the local “support” interpretation and the necessity of the architecture need clearer justification. Similarity between independently encoded tokens is not question-conditioned agreement: two related city names may be mutually exclusive answers. Learned compatibility and gating may mitigate this, but their ability to do so is not directly established. The paper explains the cues individually more clearly than why rank, probability gap, entropy, cosine similarity, and an extra compatibility network must be combined in this particular way. The most distinctive contribution appears to be the cross-step anticipatory graph; local aggregation and dual-anchor fusion need stronger justification of their added value. CCP should be discussed to distinguish prior semantic/probabilistic candidate modeling from this contribution.
-
Existing comparisons and module ablations are useful but insufficiently isolate that contribution. Stronger, reasonably tuned classifiers using identical candidate embeddings and probabilities are needed to separate richer inputs from propagation. Matched edge-perturbation controls would test anticipatory connectivity, and cue-wise and semantic/temporal/fusion ablations would test component necessity; the main ablations cover only one model and two datasets. Please clarify baseline tuning budgets and white-box layer selection. Finally, a nonsignificant difference from CHARM supports competitiveness on the tested settings, rather than establishing general equivalence to white-box detection.
Questions for authors
-
How should semantic relatedness between independently encoded tokens be interpreted as support under a specific question? What evidence shows the learned gates handle related but competing candidates, and why are all five local cues and the dual-anchor components needed?
-
How does Prop-K compare with well-tuned classifiers using identical embeddings/probabilities? Do gains persist under degree- and distance-matched edge perturbations and component-wise ablations beyond the current model/dataset pair? Please clarify tuning budgets and white-box layer selection.
-
Please distinguish the contribution from CCP and qualify the claim of white-box-level performance: the CHARM comparison does not establish statistical equivalence.
10 comments
Response to W2 and Q2
We thank the reviewer for these valuable suggestions. We’ve extended our ablations to both Llama3-8B and Gemma3-12B on HaluEval and GSM8K, as in the following.
(1) Comparison with a tuned classifier using identical inputs.
We augmented HALT with the same external encoder and Top-20 candidate tokens/probabilities as Prop-K (candidate embeddings projected, fused with HALT’s probability features).
We tuned the augmented HALT’s learning rate, dropout rate, and hidden dimension over 11 configurations per model–dataset setting.
AUROC (%)
| Method | Llama3-8B / HaluEval | Llama3-8B / GSM8K | Gemma3-12B / HaluEval | Gemma3-12B / GSM8K | Mean |
|---|---|---|---|---|---|
| HALT + Top-20 embeddings | 73.93 | 86.51 | 80.25 | 83.48 | 81.04 |
| Prop-K | 82.40 | 91.30 | 84.60 | 84.80 | 85.78 |
As can be seen, Prop-K shows a significant advantage over this tuned classifier (+ 4.73% AUROC averaged over all four settings). This means simply combining external candidate embeddings and probability features is not sufficient, and resolving the mis-match between target LLM and external embedding is more important.
(2) Do gains persist after matching node degree and positional distance?
Following your suggestion, we retrained a variant with graph edges rewired while matching node degree and positional distance. Compared with the full model, matched rewiring reduces AUROC by 1.59–4.47 % across the four settings, supporting the value of the specific cross-step connections beyond the matched graph statistics.
AUROC (%)
| Setting | Llama3-8B / HaluEval | Llama3-8B / GSM8K | Gemma3-12B / HaluEval | Gemma3-12B / GSM8K |
|---|---|---|---|---|
| Full | 82.40 ± 0.50 | 91.30 ± 1.10 | 84.60 ± 0.70 | 84.80 ± 1.40 |
| Matched rewiring | 80.02 ± 0.45 | 86.83 ± 0.69 | 83.01 ± 1.02 | 81.98 ± 0.59 |
| AUROC (pp) | +2.38 | +4.47 | +1.59 | +2.82 |
(3) Do propagation and readout gains persist across settings?
Below we report new ablation results on the contribution of the propagation/readout modules across settings (mean/std of AUROC for 3 random seeds).
AUROC (%)
| Setting | Llama3-8B / HaluEval | Llama3-8B / GSM8K | Gemma3-12B / HaluEval | Gemma3-12B / GSM8K |
|---|---|---|---|---|
| Full | 82.40 ± 0.50 | 91.30 ± 1.10 | 84.60 ± 0.70 | 84.80 ± 1.40 |
| w/o local propagation | 79.65 ± 0.34 | 83.45 ± 0.37 | 79.10 ± 0.70 | 79.80 ± 0.40 |
| w/o global propagation | 79.25 ± 0.06 | 88.67 ± 1.03 | 82.70 ± 0.50 | 82.10 ± 0.50 |
| w/o both propagation stages | 72.63 ± 0.43 | 81.78 ± 0.84 | 71.40 ± 0.60 | 72.60 ± 0.80 |
| Semantic anchors only | 81.77 ± 0.08 | 88.23 ± 0.85 | 83.17 ± 0.50 | 83.33 ± 1.78 |
| Temporal anchors only | 81.01 ± 0.37 | 87.73 ± 1.33 | 83.20 ± 0.01 | 80.14 ± 1.43 |
As noted, removing either propagation stage or either readout component leads to drop of average AUROC by 1.65%–5.28%. These show the important gains of these modules.
(4) Tuning budgets and white-box layer selection.
For clarity, we summarize the tuning settings and white-box layer-selection protocol below.
Prop-K. (Appendix C.3 and Table 8) learning rates , dropout rates , local and global propagation rounds each in , and window sizes .
White-box layers. (Appendix C.4) SAPLMA, HaMI, and CHARM follow their original feature extraction procedures and select among layers at 40%, 60%, 80%, and 100% of model depth, with indices rounded up. Layers were selected by validation AUROC.
Response to W1, W2, and Q3
We thank the reviewer for these valuable comments, our responses are as follows.
(1) Difference between Prop-K and CCP
(1) CCP mainly considers local, stepwise evidence, including candidate probabilities and semantic relations determined by a natural language inference (NLI) model at each step. In comparison, Prop-K models both local candidate consistency and global, multi-step anticipatory consistency through longer-range info. propagation.
(2) CCP exploits token semantic information through a scoring function, whereas Prop-K uses fine-grained representation learning via information propagation. Specifically, CCP computes a scalar confidence score at each position by aggregating and normalizing candidate probabilities based on NLI judgments. In contrast, Prop-K performs message passing over high-dimensional embedding vectors with adaptive gating based on consistency signals to control weights.
(2) The CHARM comparison does not establish statistical equivalence.
We highly agree with the reviewer that the statistically nonsignificant difference between Prop-K and CHARM on the tested datasets does not constitute general evidence for method equivalence and cannot be generalized to white-box or grey-box settings. We have revised the manuscript consistently from introduction to conclusion to use more conservative descriptions and avoid overclaiming. Thank you for this rigorous and valuable comment.”
Response to W1 and Q1
We thank the reviewer for these insightful comments.
We fully agree with the reviewer that due to the inherent limitations of the gray-box setting, we have to use independent external token embeddings, whose similarity in the local module may not fully capture question-conditioned agreement. This can be partially mitigated by candidate-token probabilities from the target LLM, which retain context (question) information.
Besides, the local gating function is jointly learned with the global propagation module, which captures longer-range token dependencies, so this could also help question-conditioned information to be indirectly injected to local estimation. We hope this could alleviate your concern.
We further investigate whether the learned gate helps LAC distinguish candidates that are semantically related to Top-1 but express competing answers. Specifically, we test whether the Top-1 representation learned by LAC with gating can better encode the support/conflict relationship between the local candidate set and Top-1 prediction.
We obtain a ground-truth consistency label as follows. (1) For each generation step, given the question, the answer information, and the current generation context, we use an LLM to judge whether each Top-K candidate expresses the same or an equivalent answer as the generated Top-1 (consistent) or a different/competing answer (inconsistent). (2) We then aggregate these candidate-level consistency labels into a step-level consistency label by majority voting.
After obtaining step-level consistency labels, we train a linear probe classifier to predict them using the Top-1 representations learned by LAC as features. We examine two settings: (1) with gating, and (2) without gating. A higher probe accuracy indicates that LAC better encodes whether the candidate sets support or conflicts with the current Top-1 prediction.
We use 1,000 model-generated responses and randomly sample five generation positions from each response. We further construct a hard subset of inconsistent positions where most alternatives belong to the same semantic category as the Top-1 token but express different answers. For example, the Top-1 token may name one city while most alternatives name other cities. Such cases are particularly challenging (candidates semantically related to the Top-1 token but competing with it).
Below, we report both the overall accuracy and the hard-subset accuracy:
As can be seen, the LAC-gating substantially improves probe accuracy from 54.05% to 83.20% overall and from 4.09% to 69.01% on the hard subset. This shows that gating helps LAC better encode whether semantically related candidates support or compete with the Top-1 prediction, even when they belong to the same semantic category but express different answers.
We use ablations to show that the five local cues and two anchor types all make important contributions (using Gemma3-12B and Llama3-8B, on HaluEval and GSM8K).
AUROC (%)
As noted, removing any individual cue lowers AUROC by 0.44–2.34%. Retaining only compatibility reduces AUROC by 1.15–2.85%. Using only semantic or temporal anchors leads to mean AUROC drops of 1.65-2.76%. These results demonstrate the importance of each component/module.