- Rating
- 2
- Confidence
- 3
- Informativeness
- 1
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper studies safety issues related to jailbreaking in small language models (SLMs). The authors first study the mechanism underlying jailbreaking in SLMs and present their findings: small models can still clearly separate harmful from benign inputs in model representations. The failure to refuse jailbreaking requests comes from the lack of refusal behavior after identifying harmful inputs. They name this issue the recognition-to-refusal gap, as the bottleneck comes from the internal conversion from harmfulness recognition to refusal actions. They validate their hypothesis across model families and model sizes within the scope of SLMs. They also propose Coupling-Aware Direct Preference Optimization (CA-DPO), which focuses on adversarial data points for which the model can recognize harmful inputs at the latent level but fails to perform refusal actions. Their experiments demonstrate that the method can improve safety while preserving useful behavior on the tested datasets.
Critical strengths
(1) Important and interesting problem: The paper investigates LLM safety issues and the underlying mechanisms in small language models. Understanding both these threats and methods to reduce the risk is important for developing trustworthy LLMs.
(2) Experiments on diverse model families: The experiments cover models from the Qwen3.5, Llama3.2, and Gemma3 series. The broad coverage of model families provides evidence for good generalization of the proposed method.
Critical weaknesses
(1) Potentially biased evaluation of refusal answers. The main evaluation relies on refusal keywords listed in Figure 8 instead of using semantic checks or LLMs as judges for more precise evaluation. It is possible for LLMs to generate these words while still violating the safety requirements.
(2) Lack of recent baselines: In the experiments, the most recent baseline is Scans, which was proposed in 2025. The rest of the baselines were proposed in 2024 or earlier. The paper itself also discusses some recent methods proposed in 2026, such as HARC [1]. More discussion comparing these methods with recent approaches would improve the soundness of this paper.
Reference:
[1] HARC: Coupling Harmfulness and Refusal Directions for Robust Safety Alignment. Preprint, 2026.
Questions for authors
(1) Based on your experiments on larger open-source models such as Qwen3.5 4B, 9B, and 27B, do you observe that the recognition-to-refusal gap consistently correlates with refusal behaviors as model size increases?
(2) In Figure 2, the jailbreak points are more tightly grouped in a smaller area than the clean samples. I am thus curious to see whether there are any other features that can contribute to this difference, for example, input format and length. Could you provide some examples of the formats and a length analysis (mean and standard deviation of token length) for the different input sources?
2 comments
2. Missing comparison with HARC
We have added a direct HARC comparison on the same small models and test prompts, assessing semantic jailbreak refusal, false refusal on non-harmful inputs, and output collapse. CA-DPO outperforms HARC on each evaluated jailbreak set while also reducing false refusals and collapse. We will document implementation and training-setting differences in the revision.
| Model | Method | Clean ↓ | OKTest ↓ | OR ↓ | FR ↓ | Cipher ↑ | H4rm3l ↑ | Hand ↑ | Collapse ↓ |
|---|---|---|---|---|---|---|---|---|---|
| Qwen3.5-0.8B | Base | 0.11 | 9.83 | 35.50 | 15.50 | 0.83 | 27.83 | 17.50 | 3.07 |
| Qwen3.5-0.8B | HARC | 31.50 | 59.83 | 99.83 | 84.17 | 90.83 | 65.67 | 44.50 | 14.83 |
| Qwen3.5-0.8B | CA-DPO | 0.11 | 2.83 | 7.83 | 0.33 | 91.17 | 94.17 | 44.67 | 1.07 |
| Llama3.2-1B | Base | 1.11 | 3.33 | 12.33 | 21.50 | 20.33 | 33.50 | 48.33 | 6.47 |
| Llama3.2-1B | HARC | 12.17 | 24.67 | 97.33 | 71.83 | 89.33 | 85.17 | 74.83 | 9.41 |
| Llama3.2-1B | CA-DPO | 0.89 | 1.83 | 5.33 | 15.33 | 89.50 | 89.67 | 75.17 | 4.13 |
Here, the jailbreak columns are 100% minus the SORRY-Bench fulfillment rate; the non-harmful columns use keyword-based false-refusal rates. All values are percentages.
Follow-up 1. Scale trends in larger models
Appendix Figures 11–13 report the layer-wise harmfulness-related signal, refusal-related signal, and direction cosine for Qwen3.5-4B, 9B, and 27B, respectively. Across the larger scale range, the 27B model exhibits stronger directional coupling than the smaller models, consistent with the scale-related trend discussed in the paper. Adjacent model sizes are not strictly monotonic—for example, 4B is not uniformly more coupled than 2B—so we describe this as an overall scale trend rather than an improvement guaranteed at every size increment.
Directional coupling should also be distinguished from the per-example recognition-to-refusal gap and from the observed refusal rate. The appendix figures provide layer-wise signal statistics and do not alone establish monotonicity of either per-example gap or refusal rate across all model sizes. We will make this distinction explicit in the revision.
Follow-up 2. Input formats and token lengths in Figure 2
Figure 2 is a qualitative visualization, and its jailbreak samples are primarily cipher examples. Appendix Table 4 compares the three attack surface forms and shows a degree of consistency in their harmfulness-related directions, including positive leave-one-family-out separation. We will add an appendix visualization using the balanced, all-family configuration used by the analyses after Section 3.2. The subsequent quantitative analyses use balanced attack pools and an independent evaluation split; their data composition is described in Appendix A. The source-level token statistics below make the possible format and length differences explicit rather than treating Figure 2 as a controlled length-matched experiment.
| Source | Number of samples | Token length (mean ± std.) |
|---|---|---|
| Cipher jailbreak | 200 | 312.27 ± 65.71 |
| H4rm3l jailbreak | 200 | 240.60 ± 106.03 |
| Hand-crafted jailbreak | 200 | 161.71 ± 224.65 |
| Clean | 300 | 100.91 ± 151.11 |
| OKTest | 200 | 15.15 ± 4.06 |
| OR-Bench | 200 | 20.27 ± 5.12 |
| FalseReject | 200 | 24.85 ± 5.09 |
The reviewer is right that a response can include refusal wording yet still fulfill a harmful request. We have therefore added SORRY-Bench semantic evaluation of the jailbreak outputs: a response that fulfills the harmful request is counted as fulfillment even if it contains refusal words, and an unfulfilled harmful request is treated as a refusal. We retain the original keyword-based false-refusal metric for clean and over-refusal inputs, which are non-harmful requests, and report the metrics separately. The new semantic results support the safety improvement of CA-DPO over Base. We also inspected a subset of answers manually and found that keyword-based judgments were generally close to the observed behavior, with few apparent misclassifications.
Each entry is Base → CA-DPO semantic refusal rate, in percent. We will present these results alongside the original keyword-based evaluation and output collapse rates in the revision.