- Rating
- 3(+1)
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper addresses a well-known failure mode of steering as steering vectors estimated from contrastive data (e.g., difference-of-means) entangle the target concept with incidental variation (topic, style, length), and injecting such vectors degrades generation quality. The authors propose a new method Contrastive Covariance Steering (CCS) which centers all class activations on the grand mean of class means and builds per-class scatter matrices about this common center. No need for training, the CCS beats every activation-steering baseline in 3 LLMs (Llama-3.1-8B, Qwen3-8B, Gemma-2-9B) × 4 domains (AxBench, BIG5/NPTI, Emotion, TruthfulQA).
Critical strengths
- The entanglement account of why steering vectors hurt fluency is intuitive, and the paper consistently ties design choices back to this story. This makes the method easy to follow and understand. Also, the human evaluation is more rigorous than most steering papers.
- Solid experiments and evaluation. The multi-turn evaluation is a genuinely useful contribution. The author conduct an important experiments on 20-turn steering with a joint concept-fluency metric and the finding that CCS overtakes system prompting by turn 20 (80% vs. 71%) is an interesting counterpoint to the "prompting is unbeatable" narrative from AxBench.
Critical weaknesses
- The paper is well structured but the evaluation is confined to concept expression and text quality; capability and safety side effects are unmeasured. This makes the contribution limited. There is no measurement of what steering does to general capabilities (e.g., MMLU, reasoning) or safety-relevant behavior, yet recent multi-dimensional evaluations (e.g., SteerScope-style suites) show that steering efficacy and side effects are strongly coupled, and that method rankings can change once side effects are priced in.
- Each component is a classical tool: grand-mean centering, the total-scatter decomposition (the standard LDA scatter identity), a contrastive variance objective in the spirit of contrastive PCA, and projection filtering via eigendecomposition. The paper's contribution is primarily a careful adaptation and thorough validation of these ideas for steering-vector construction. From my perspective, the paper is good but not that inspiring. I would like to raise my score if I can see more interesting and promising results for current reasoning and agentic era.
- Some writing and presentation comments.
- is used in Eq. 3 (l. 183) but only introduced in the following paragraph (l. 190). Moreover, the column scaling is part of 's definition, yet Eq. 3 already relies on it to identify \tilde{H}_c\tilde{H}_c^\top. Please define \tilde{H} before Eq. 3.
- in Eq. 4 (l. 192) is never formally introduced (e.g., " with orthonormal columns spanning the sought subspace"). is also an unfortunate choice of symbol, as it conventionally denotes model weight matrices; consider or .
- The target-class index , the contrastive weight , and the projector all appear in Figure 2 before their formal definitions (l. 190–192, Eq. 6).
- Please introduce every mathematical symbol near its first appearance, including inside figure panels.
- Figure 1 is hard to parse and Figure 2 is clearer in structure but has consistency problems. It appears in the Introduction (p. 2), before any notation exists, yet uses , , , , , without any legend; the caption does not define them either. Also, the plane, the contrastive subspace, i.e., the central object of the paper, is unlabeled. The covariance-ellipse convention ("") is never explained.
- In figure 2, Terminology drifts across the paper: panel 1 says "Competing", Figure 1 says "Nuisance", the text says "nuisance classes" and "non-target scatter". Please unify. Panel 3 states the intervention as , which matches neither Eq. 1 (unit-normalized ) nor the actual intervention used (Eq. 7, per-token norm scaling). Please align the figure with the real steering rule. Panel 2 draws the nuisance ellipse centered on and the target ellipse offset from it; it is unclear whether this asymmetry is intentional (scatter about the common center includes the displacement term , so both should in general be off-center).
Questions for authors
- Does CCS preserve general capabilities (e.g., MMLU or a comparable benchmark) at the steering strengths selected in Table 1, and how does this compare to CAA/Linear-AcT at matched concept-expression levels?
- Was the intervention layer (15 for Llama, 21 for Qwen/Gemma) re-validated for CCS, or inherited from prior work on other methods?
5 comments
Response to Reviewer p9ag (2/4)
3. Steering under capability constraints. To see how much can we steer with each method if we don't want to drop the model's general capabilities significantly, we evaluate the highest steering effectiveness achievable while limiting the average accuracy loss to 2 or 5 points. We choose the strength for each concept that maximize the average effectiveness, subject to the average accuracy loss staying within the budget, and average over the three models. We also report the best effectiveness without a budget.
Table R3. Highest average AxBench effectiveness within each accuracy budget (average accuracy change in parentheses); strengths in {0.2, 0.4, 0.6, 0.8, 1.0}.
| Budget | Unsteered | CCS | CAA | RepE | PCA | Linear-AcT | SEA | Angular |
|---|---|---|---|---|---|---|---|---|
| ≤ 2 points | 0.078 | 0.366 (−1.9) | 0.216 (−1.8) | 0.210 (−1.6) | 0.303 (−2.0) | 0.180 (−1.9) | 0.118 (0.0) | 0.160 (−0.3) |
| ≤ 5 points | 0.078 | 0.474 (−4.5) | 0.255 (−4.7) | 0.223 (−3.1) | 0.362 (−3.9) | 0.261 (−4.9) | 0.123 (−0.9) | 0.160 (−0.3) |
| No budget | 0.078 | 0.646 (−21.0) | 0.297 (−18.0) | 0.256 (−16.9) | 0.398 (−12.3) | 0.408 (−20.7) | 0.127 (−2.6) | 0.166 (−2.8) |
The scores are computed on the 10 sweep concepts and therefore differ from Table 1. Under both budgets CCS achieves the highest effectiveness: with an average accuracy loss of at most 5 points it reaches 0.474, compared with 0.362 for PCA and 0.255 for CAA. CCS therefore retains its advantage when capability preservation is explicitly constrained.
4. Safety side effects. We compare safety changes at matched steering effectiveness on AxBench, averaged over the three models (baseline / CCS; higher is worse).
Table R4.
| Baseline | HarmBench (harmful compliance) | XSTest (over-refusal) | XSTest (unsafe compliance) |
|---|---|---|---|
| CAA | −1.4 / −0.9 | +3.2 / +4.6 | +0.1 / −1.5 |
| RepE | −1.5 / −0.2 | +2.9 / +2.6 | +0.1 / −0.7 |
| PCA | −1.6 / −1.6 | +4.8 / +5.1 | −0.9 / −1.4 |
| Linear-AcT | −1.4 / −1.4 | +5.0 / +5.3 | +10.6 / −1.2 |
| SEA | +0.1 / +0.1 | −0.1 / +0.4 | −0.1 / +0.1 |
| Angular | +0.1 / 0.0 | +0.4 / +0.9 | +2.2 / −0.3 |
Harmful compliance on HarmBench changes little for all methods (between −1.6 and +0.1 points). Over-refusal on XSTest increases for all additive methods, by similar amounts for CCS and its matched baselines (+2.6 to +5.3 vs. +2.9 to +5.0 points). Unsafe compliance decreases for CCS in almost all comparisons, whereas Linear-AcT increases it by 10.6 points on average. Overall, CCS does not show systematically larger safety side effects in these evaluations, although these results do not establish general safety guarantees.
Response to Reviewer p9ag (3/4)
5. Reasoning models: steering with thinking enabled. We evaluate Qwen3-8B with thinking enabled, applying steering either only to the final answer or to both the reasoning trace and the answer, and compare against the same model with thinking disabled. All settings use the same steering directions, 10 AxBench concepts, evaluation prompts and capability questions; accuracy changes are relative to each setting's unsteered baseline. Due to time constraints, the thinking results use a subset of about half of the 1,000 questions, the same for all three settings. As in Table R3, each concept uses its own strength, chosen to maximize average effectiveness within a 2- or 5-point accuracy budget, or without a budget.
Table R5. Qwen3-8B: highest average AxBench effectiveness within each budget (average accuracy change in parentheses).
| Setting | Method | ≤ 2 points | ≤ 5 points | No budget |
|---|---|---|---|---|
| Thinking, answer only | CAA | 0.41 (−1.7) | 0.41 (−3.7) | 0.41 (−3.7) |
| PCA | 0.39 (+1.2) | 0.39 (+1.2) | 0.39 (+1.2) | |
| CCS | 0.66 (−1.3) | 0.66 (−1.3) | 0.66 (−1.3) | |
| Thinking, reasoning + answer | CAA | 0.33 (−2.0) | 0.37 (−4.7) | 0.39 (−14.6) |
| PCA | 0.40 (−2.0) | 0.43 (−3.3) | 0.43 (−3.3) | |
| CCS | 0.59 (−1.7) | 0.74 (−4.8) | 0.74 (−4.8) | |
| No thinking | CAA | 0.21 (−1.7) | 0.25 (−4.6) | 0.40 (−26.7) |
| PCA | 0.30 (−1.6) | 0.34 (−4.9) | 0.35 (−10.1) | |
| CCS | 0.38 (−2.0) | 0.53 (−4.9) | 0.58 (−13.0) |
Restricting steering to the final answer of a thinking-enabled model substantially improves the effectiveness–capability trade-off. Under a 2-point budget, CCS achieves 0.66 effectiveness with answer-only steering, compared with 0.38 without thinking. Without a budget, answer-only steering reaches 0.66 at a 1.3-point accuracy loss, whereas the model without thinking reaches 0.58 at a 13.0-point loss. Since the reasoning trace that solves the question is left unsteered, reasoning accuracy is largely preserved while the concept is still expressed in the answer. Steering both the reasoning trace and the answer reaches the highest unconstrained effectiveness (0.74), at a 4.8-point accuracy loss. Across all settings and budgets, CCS achieves the highest effectiveness. This table reports Qwen3-8B only, whereas Table R3 averages three models over separate generations. These preliminary findings suggest that applying interventions only during final-answer generation can substantially mitigate capability degradation while retaining effective control over the expressed behavior.
Summary and changes to the paper. CCS, like other additive methods, can substantially degrade general capabilities at strong intervention settings. However, at matched steering effectiveness and under explicit accuracy-loss budgets, it provides a better effectiveness–capability trade-off than competing activation-steering methods, without systematically worse safety outcomes. For reasoning models, steering only the final answer largely preserves reasoning accuracy while keeping concept effectiveness high. We have added these results to the revised paper: the capability and safety results in Appendix I (Tables 23–26) and the thinking-model results in Appendix J (Table 28).
Response to Reviewer p9ag (1/4)
W1 + Q1: Capabilities and safety side effects
We thank the reviewer for highlighting the importance of capability and safety side effects; we agree that steering effectiveness alone is insufficient. We ran additional evaluations of general capabilities and safety when steering towards AxBench concepts, for all three models, both at the original Table 1 strengths and at matched steering effectiveness. We also study a reasoning model with thinking enabled.
Evaluation setup.
We additionally sweep each activation-based method over 10 steering strengths, evaluating capability on the same 1,000 questions and safety on all 650 prompts at every strength, which allows comparing methods at matched steering effectiveness. Due to time constraints, the sweeps use 10 of the 50 AxBench concepts, sampled at random with the same genre mix as the full set (7 text, 2 code, 1 math).
1. Capability at the original steering strengths. Table R1 reports the AxBench effectiveness from Table 1 together with the accuracy change at the same operating points, averaged over the three models and all 50 concepts.
Table R1.
At its Table 1 strengths, CCS reaches by far the highest effectiveness among activation-based methods (0.83 vs. at most 0.49), and its accuracy loss (−18.6 points) is in the same range as CAA, PCA and Linear-AcT (−16.9 to −19.2), which steer much less. Strong interventions therefore degrade capabilities for all additive methods, and the Table 1 operating points differ in strength across methods. This motivates comparing methods at matched effectiveness.
2. Capability at matched steering effectiveness. For each baseline, we take its Table 1 effectiveness and compare its capability change with that of CCS at the same effectiveness, using the strength sweeps.
Table R2. Accuracy change (points) at matched effectiveness, baseline / CCS, averaged over three models.
At matched effectiveness, CCS loses less accuracy than every baseline, e.g. 4.4 vs. 18.9 points for CAA and 7.5 vs. 18.8 points for Linear-AcT. These values use the 10 sweep concepts and therefore differ from Table R1.
References
Clark, P., et al. Think you have solved question answering? Try ARC, the AI2 reasoning challenge. arXiv:1803.05457, 2018.
Hendrycks, D., et al. Measuring massive multitask language understanding. ICLR, 2021.
Mazeika, M., et al. HarmBench: A standardized evaluation framework for automated red teaming and robust refusal. ICML, 2024.
Röttger, P., et al. XSTest: A test suite for identifying exaggerated safety behaviours in large language models. NAACL, 2024.