- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
The paper studies behavioral safety of tool-using agents under legitimate user requests. A linear probe predicts decision risk from hidden states before the next Thought is generated. A steering direction is constructed from differences between Thought representations of risky proposals and teacher-generated risk-aware proposals under matched contexts. During deployment, steering is applied to the subsequent Thought–Action generation when the predicted risk meets or exceeds a threshold, while the acting model remains frozen. The main additional design is Safety Preference Stability (SPS), which selects a fixed intervention layer and direction using worst-case utility over a finite set of local direction perturbations, lower-tail utility under case-level bootstrap resampling, and utility dispersion. Experiments evaluate three Qwen3 model sizes on in-distribution ToolEmu cases and newly constructed evolving scenarios, reporting safety, helpfulness, legitimate-goal completion, and per-invocation overhead.
Critical strengths
The paper addresses a practically important problem: harmful tool use while carrying out legitimate requests. Its emphasis on preserving safe task progress is valuable, and evaluating legitimate-goal completion against fixed, method-independent references helps reveal the utility costs of excessive refusal. The deployment mechanism is lightweight, requiring a risk probe and a fixed steering intervention without additional calls to a large guard model. The controlled fixed-context measurements support lower per-invocation overhead than the evaluated external defenses. The case-level bootstrap design also preserves the grouping of dependent decisions from the same case during resampling.
Critical weaknesses
-
The incremental contribution of SPS is not sufficiently isolated by the experiments. The mean activation-difference construction and additive intervention follow CAA (Rimsky et al., 2024). Conditional activation steering has been studied in CAST (Lee et al., 2025), and AgentLens (Luo et al., 2026) investigates risk detection and representation-level mitigation for multi-turn agents. I therefore view the main methodological contribution as the risk-aware paired construction and SPS-based selection criterion. SPS is reasonably motivated, but the evidence for its benefit over simpler selection criteria remains limited. Appendix C reports component ablations on Qwen3-8B, with partial Spearman correlations of 0.704 for the full score and 0.219, 0.139, and 0.602 after removing individual components. These results support the usefulness of the components for candidate ranking. However, the measured outcome is reference-tool alignment, which the paper explicitly distinguishes from argument-level safety and task completion. The paper does not clearly report a matched comparison between SPS and selection by nominal utility alone, using identical paired data, gating, intervention strength, and candidate layers. Nor does it report the SPS component ablations on final safety and completion metrics. Although Table 1 provides end-to-end results in evolving scenarios, these results do not isolate the contribution of SPS from the other framework components. Consequently, the practical benefit of the proposed stability criterion over simpler alternatives remains insufficiently established.
-
The reported percentages require clarification of the evaluation and aggregation protocol. Appendix B.1 states that the in-distribution test set contains 108 cases, while Appendix B.6 defines safety and helpfulness as binary pass rates over trajectories with valid scores. However, the Qwen3-4B safety rates in Table 1—52.89% for GuardAgent, 39.67% for CAA, and 61.98% for Thought-Aligner—cannot be obtained by rounding an integer pass count divided by an integer denominator of at most 108 to two decimal places.
Questions for authors
-
Please clarify how the percentages in Table 1 were computed, including the number of rollouts per case, any averaging across runs, and the pass counts, valid trajectory counts, and excluded evaluations for each method and metric. In particular, how do the 108 in-distribution cases yield the Qwen3-4B safety rates of 52.89%, 39.67%, and 61.98%?
-
With paired data, gating, intervention strength, and candidate layers held fixed, how does full SPS compare with selection by nominal utility alone on final safety and legitimate-goal completion, particularly in evolving scenarios? Please provide these outcomes for the component ablations, together with uncertainty estimates. To assess the specific contribution of Thought-based construction, is there a controlled comparison with directions extracted from Action or the full proposal?
3 comments
Response to Reviewer omEh (Part II)
W1 & Q2(a). Full SPS versus nominal utility
With paired data, gating, intervention strength, and candidate layers held fixed, how does full SPS compare with selection by nominal utility (U(\kappa)) alone on final safety and legitimate-goal completion, particularly in evolving scenarios?
To isolate the selector, we held the paired proposals, candidate layers, probe, gate (), steering strength () and decoding fixed on Qwen3-8B. CAA is an ungated baseline. Safety covers 108 ID and 72 Evolving tasks. LGC covers the 99 and 66 tasks with legitimate goals.
| Selector | ID safety / LGC (%) | Evolving safety / LGC (%) | Evolving vs. nominal (safety / LGC, points) |
|---|---|---|---|
| CAA (un-gated baseline) | 60.19 / — | 33.33 / 49.24 | -20.83 / -10.61 |
| Nominal utility only | 56.48 / 69.02 | 54.17 / 59.85 | 0.00 / 0.00 |
| Full SPS (R_dir + R_tail + R_stab) | 62.04 / 73.91 | 69.44 / 68.18 | +15.28 / +8.33 |
Compared with nominal utility, full SPS gains 5.56 safety points and 4.88 LGC points in ID, and 15.28 and 8.33 points in Evolving.
On Evolving, nominal vs. full SPS safety / LGC (%) was 50.00 / 53.03 vs. 63.89 / 56.82 on Qwen3-4B, and 59.72 / 52.27 vs. 75.00 / 59.09 on Qwen3-14B. The safety gains were 13.89 and 15.28 points, respectively.
We will add the matched nominal-utility comparison to Section 3.3 and report its ID and Evolving safety and LGC results in Appendix C.
Q2(b). Contribution of each SPS component
Please provide these outcomes for the component ablations, together with uncertainty estimates.
| Qwen3-8B selector | ID safety / LGC (%) | Evolving safety / LGC (%) | Evolving difference vs. full SPS (points) | Safety standard error (SE), ID / Evolving (points) |
|---|---|---|---|---|
| Full SPS | 62.04 / 73.91 | 69.44 / 68.18 | 0.00 / 0.00 | 4.67 / 5.43 |
| Without directional term | 58.33 / 70.88 | 59.72 / 62.88 | −9.72 / −5.30 | 4.74 / 5.78 |
| Without lower-tail term | 57.41 / 69.70 | 58.33 / 61.36 | −11.11 / −6.82 | 4.76 / 5.81 |
| Without dispersion term | 61.11 / 72.73 | 66.67 / 66.67 | −2.78 / −1.52 | 4.69 / 5.56 |
Removing the lower-tail term produces the largest observed Evolving safety drop (eight fewer passes), followed by the directional term (seven) and dispersion (two). The last column gives binomial standard errors for individual safety rates, , treating tasks as independent draws. Paired uncertainty for the differences and for LGC requires the per-task outcomes.
In Appendix C, we will add a rollout table alongside Table 4 showing how omitting each SPS term changes safety and LGC on both test sets.
Q2(c). Direction extraction span
To assess the specific contribution of Thought-based construction, is there a controlled comparison with directions extracted from Action or the full proposal?
On Qwen3-8B we varied only the offline span in Eq. 5. The online intervention in Eq. 8 still covered the subsequent Thought and Action tokens whenever the gate opened ().
| Offline extraction span | Selected layer | ID safety / LGC (%) | Evolving safety / LGC (%) | Evolving difference vs. Thought (points) |
|---|---|---|---|---|
| Thought only (Ours) | 28 | 62.04 / 73.91 | 69.44 / 68.18 | Baseline |
| Action only (tool name + arguments in JavaScript Object Notation (JSON)) | 26 | 52.78 / — | 44.44 / 52.27 | -25.00 / -15.91 |
| Full proposal (Thought + Action) | 28 | 58.33 / 69.70 | 61.11 / 62.88 | -8.33 / -5.30 |
On unseen toolkits, Action alone scores 25.00 safety points below Thought alone. Pooling both spans narrows the gap to 8.33 points. We observed malformed arguments with Action-only directions. A risky action can call for the same precaution across toolkits even when names and argument formats change. A direction estimated from Thought may capture that shared reason, whereas including Action tokens can make the direction more sensitive to toolkit-specific syntax.
We will distinguish the extraction and intervention spans in Section 2.3 and add the three-span comparison to Appendix C, including the selected layer and rollout outcomes for each span.
Dear Reviewer,
Thank you again for your thoughtful comments. In the revised manuscript, we expanded the reporting of evaluation counts and missing-score handling in Appendix C (Tables 6–7), including pass/fail/missing counts for Qwen3-4B and SPS pass counts for Qwen3-8B and 14B. We also added matched end-to-end comparisons between SPS and nominal-utility selection, with component ablations, in Section 3.4 (Table 2) for Qwen3-8B and Appendix E.1 (Table 13) for Qwen3-4B and 14B. Finally, Appendix D.3 (Table 10) compares Thought-only, Action-only, and Thought+Action extraction with the online intervention span held fixed. We would appreciate your feedback on whether these additions address your main concerns, and we would be happy to clarify any remaining questions.
Best regards, The authors
Response to Reviewer omEh (Part I)
We thank the reviewer for their careful audit of our evaluation. In the response below, we provide detailed counts for Table 1 (Part I) and compare Safety Preference Stability (SPS) with nominal utility, component ablations, and alternative extraction spans (Part II).
W2 & Q1(a). Rollouts, valid scores, and excluded evaluations
For each metric (safety or helpfulness), the valid-score pass rate is the fraction of scored trajectories with a ToolEmu score of 2 or 3 (Appendix B.6). A malformed response, a timeout or an
unknownjudgment can leave a metric unscored. For example, GuardAgent passes safety on 57/106 scored tasks (53.77%), or 57/108 designated tasks (52.78%) when missing scores count as failures.Each case contributes one trajectory per method. For Qwen3-4B, the table gives safety and helpfulness over valid scores and over all 108 in-distribution (ID) tasks (missing counts as failure). Legitimate-goal completion (LGC) applies to 99 tasks with at least one legitimate goal. We average each task’s completed-goal fraction over scored tasks, then over all 99 with missing scores set to zero.
With missing scores counted as zero, SPS leads Thought-Aligner by 7.41 safety points on the same 108 ID tasks (75 vs. 67 passes) and 7.07 LGC points on the same 99 eligible tasks (64.31% vs. 57.24%).
For SPS at 8B and 14B, the ID safety counts are 67/108 and 78/108, and helpfulness counts are 63/108 and 71/108, respectively. The complete method-by-metric tally here covers 4B. We will add the per-metric counts and denominator conventions to Appendix B.6, define a safety pass in the Table 1 caption, and report a follow-up comparison averaging repeated rollouts within each case.
Q1(b). Why the three published Qwen3-4B rates used 121 cases
We designated 108 of the 144 ToolEmu cases for ID evaluation and 36 for offline development. We also ran 13 offline calibration cases to check that we could reliably extract safety and helpfulness scores from the evaluator outputs. The combined log contained 121 records. On the 108 designated test cases, the safety pass rates among valid scores were 57/106 (53.77%) for GuardAgent and 67/107 (62.62%) for Thought-Aligner.
The published 4B safety rates used 64/121 for GuardAgent, 48/121 for CAA and 75/121 for Thought-Aligner. For CAA, the corrected safety rate is 43/108 (39.81%). Qwen3-8B SPS helpfulness changes from 71/121 (58.68%) to 63/108 (58.33%).
We apologize for these reporting errors. We will correct the affected Table 1 entries and their reported gains and document the test-ID filter in Appendix B.1.