- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This submission proposes VC-OPD, an on-policy self-distillation (OPSD) framework for small MLLMs. The teacher receives a training-only, question-conditioned textual visual cue as privileged information, while the student sees only the original image and question. A Visual Cue Agent uses a large MLLM (Gemini-3.5-Flash) to produce the cue, filtered by three gates: Difficulty (Eq. 1), Answer (Eq. 3) and Reliability (Eq. 4). These build a Visual Privilege Library (Eq. 5). A shared backbone is first GRPO-warmed into a frozen teacher (Eq. 6). Distillation then minimizes token-level JSD between the student and the cue-conditioned teacher on student rollouts (Eq. 8). Experiments fine-tune Qwen3-VL-Instruct at 2B, 4B and 8B on ViRL39K and evaluate on STEM and general visual-reasoning benchmarks (Table 1), with ablations over privileged-information type (Table 2) and GRPO warm-up (Table 3).
Critical strengths
- The Table 2 dissociation matters. By holding the student, teacher warm-up and configuration fixed and varying only the privileged information (§4.3, Table 2), the submission shows that maximizing teacher capability does not maximize transfer: the Answer/Reasoning teacher scores 95-98 against 72-82 for the Visual Cue teacher, yet Visual Cue wins downstream on average at every scale. This is the result that goes beyond showing that cues help, and it deserves the most weight in the assessment.
- Consistency across scales and a clean warm-up ablation. Table 1 and Table 3 report gains at 2B, 4B and 8B on two benchmark groups, and Table 3 isolates the GRPO warm-up contribution (for example +6.48/+3.91/+1.64 STEM at 2B/4B/8B against base+OPD). The pattern is monotone enough to be believable and is not a single lucky cell.
Critical weaknesses
-
I think the core conceptual claim overlaps substantially with a public work that is not cited, and the closest contemporaneous baseline is cited but not compared empirically. §1 (contributions) and §2 present question-conditioned visual cues as teacher-side privileged information that is task-relevant and recoverable as the novelty. However, the public preprint ViCuR (Visual Cues as Recoverable Privilege for Multimodal On-Policy Distillation, arXiv 2606.05718, submitted 4 June 2026) already proposes replacing answer-side privilege with query-related visual cues on recoverability grounds, and it does not appear in the reference list (§References). ViCuR is earlier than VAD, which the submission does cite. Separately, VAD (arXiv 2607.28590) is cited in §2 but is not compared empirically, even though VAD is the closest shared-framework method and reports a matched-budget result at 4B on its own six-benchmark suite. Both ViCuR (4 June 2026) and VAD (30 July 2026) predate the ICLR 2027 deadline of 25 September 2026, so a comparison with both was feasible. As written, the delta over ViCuR (an LLM-generated textual cue, three-gate filtering and API-only cue generation) is stated but not yet demonstrated. It would help to add an explicit differentiation section and a head-to-head comparison under a common evaluation suite.
-
Two controls would strengthen the argument. The recoverable property is not measured on the student, and run-to-run variability is not reported. The Reliability Gate (Eq. 4) shows only that the teacher answers correctly ≥7/8 with the cue; it does not test whether the student can recover the evidence from the original image, which is the mechanism the paper claims (§3.2-3.3, Eq. 8 discussion). Tables 1-3 report single point estimates with no seeds, no error bars and no significance tests. §4.1 does not say whether the baselines (OPSD, Vision-OPD, VGPO, ReMax, GRPO) were reimplemented or run from official code, nor their tuning budget. Some baseline cells fall below the untuned base (for example Vision-OPD-2B STEM average 40.41 against Qwen3-VL-2B base 43.46, Table 1), so it is difficult to tell whether the comparison is compute fair. It would help to report seeds and variance, an explicit baseline protocol and a student-side recovery probe.
-
The three gates are central to the method, but the paper does not appear to ablate them. The Difficulty, Answer and Reliability gates (Eqs. 1, 3 and 4; thresholds τd−=1, τd+=7 and τrel=7 in Appendix D) decide which samples and cues enter the Visual Privilege Library, so they change both the training distribution and the amount of supervision. Tables 2 and 3 ablate the privileged-information type and the GRPO warm-up, yet I could not find an experiment that removes or varies a gate, reports how many samples each gate filters, or tests whether the gains survive a simpler filter. The contribution of the gating pipeline, as opposed to the cue content or the reduced training set, is therefore difficult to identify. It would help to report per-gate retention rates and to ablate each gate and its thresholds.
-
The paper states that the final answer is excluded from the cue to prevent leakage (§3.2), and Fig. 3(b) shows a cue from the tuned model that ends with "The lowest accuracy value reported is 1." This appears to be the final answer. I could not find an automated check for leakage at scale, and the Answer Gate verifies only that the large MLLM's answer is correct, not that the cue omits it. If cues occasionally contain the answer, the teacher's supervision can encode the same shortcut the paper aims to remove, which also complicates the Table 2 comparison with Answer privilege. It would help to run an automated leakage detector over the full Visual Privilege Library, report the rate, and either filter leaking cues or show that the downstream result is unchanged.
Questions for authors
- The public VAD release (arXiv 2607.28590) includes an internal run identifier beginning with VC-OPD whose optimization configuration (batch 96, PPO mini-batch 96, 8 rollouts per prompt, lr 2×10⁻⁶, symmetric JSD α=0.5, top-100 student support) matches Table 4 of this submission. The public VAD repository describes
target_mode=vcandvc_target_variant=budgeted_asvcas internal identifiers. Please clarify the relationship between this submission and the public VAD (arXiv 2607.28590) and V-Zero (arXiv 2606.25319) work. Is it a shared codebase, a shared data pipeline, or independent work? This materially affects how I weigh the contribution. - ViCuR (arXiv 2606.05718) proposes query-related visual cues as recoverable privilege for multimodal OPD, and it is not cited. What exactly does VC-OPD add over ViCuR? Is it the LLM-generated textual form of the cue, the three-gate filtering, black-box cue generation, or the scale of the empirical study?
- VAD (arXiv 2607.28590) is cited in §2 but is not compared empirically, and V-Zero (arXiv 2606.25319), which VAD itself compares against, does not appear in the submission. Why is there no empirical comparison, and would VC-OPD retain its advantage under a common benchmark suite and a matched budget?
- See Weakness 4: Could you report the leakage rate and the automated check used at scale? 5.Add seeds and variance, a student-side recoverability measurement, and a train/test contamination analysis between ViRL39K and the evaluation benchmarks.
4 comments
Response to Reviewer 8P4Y (Q1 and Q2)
Relationship to VAD code and experimental settings
VC-OPD was developed independently using the publicly available Vision-OPD codebase, without using VAD's or V-Zero's code or data-processing pipelines. Similar training configurations may arise from the shared Vision-OPD foundation. We cannot explain the VC-OPD identifier used in VAD's public experiments, as it was assigned by its authors. We will clarify our code provenance and methodological differences in the revision.
Empirical comparison with V-Zero, VAD, and ViCuR
We reproduced V-Zero, VAD, and ViCuR and evaluated them alongside VC-OPD on the same benchmark suite. Results appear in Table R5 (STEM visual reasoning) and Table R6 (general visual understanding).
Vision-OPD, V-Zero, and VAD all use local visual crops to support distillation, but in different ways:
Vision-OPD: The student sees the original image; the teacher uses enlarged crops of key regions.
V-Zero: It compares teacher probabilities under relevant- and irrelevant-region views to estimate students' reliance on visual evidence and adjust distillation weights.
VAD: It uses changes in the teacher distribution when key regions are present versus removed to estimate evidence-driven corrections and reconstruct distillation targets.
By contrast, VC-OPD supplies the teacher with question-relevant cues rather than only enlarged local regions, making it suitable for tasks requiring both detailed perception and complex visual reasoning.
Additional experiments support VC-OPD's advantages. Tables R5–R6 show that VC-OPD outperforms concurrent methods V-Zero, VAD, and ViCuR on both STEM and general-vision tasks at 2B, 4B, and 8B. These consistent results indicate improvements in both fine-grained perception and complex reasoning and support the method's effectiveness and generalization.
Table R3 (response to Reviewer uF2w) further compares VC-OPD and Vision-OPD on fine-grained perception and global spatial reasoning. Vision-OPD is stronger on some local-detail tasks but weaker on spatial tasks. VC-OPD maintains strong local perception while improving SpatialEval and PuzzleVQA performance, yielding higher overall averages at all three model scales and better adaptation across tasks.
In summary, the new experiments show better overall performance for VC-OPD across model sizes, supporting our cue-generation and distillation design. The revision will expand the related-work discussion, implementation details, and full comparisons.
Q2. Baseline reproduction details and Qwen3-VL-2B performance degradation
Thank you for the suggestion. We clarify our baseline reproduction protocol and the performance drop observed for the 2B model.
Baseline reproduction and fairness
All baselines share the same evaluation protocol. Their training settings are:
Vision-OPD and VGPO: We used the official open-source code and original hyperparameters, replacing only the base model with the corresponding Qwen3-VL size.
OPSD: We matched VC-OPD's training hyperparameters for fairness.
GRPO and ReMax: We implemented both from official verl examples and matched VC-OPD's base model, batch size, learning rate, and rollout settings.
Using official implementations or aligned training settings helps make the reproduced baselines reliable and comparable.
Why Qwen3-VL-2B performance drops
Some baselines, including Vision-OPD, degrade on Qwen3-VL-2B, whereas VC-OPD improves performance. We view this contrast as an advantage of VC-OPD.
Qwen3-VL-2B has limited initial visual-reasoning ability. Under direct OPSD, the teacher may not use visual privileged information effectively, leading to unchanged or lower performance.
In the original paper, Table 1 shows that Vision-OPD-2B lowers the STEM average from 43.46 to 40.41 (−3.05 points). The ablation in Table 3 similarly shows a drop from 43.46 to 41.66 (−1.80 points) with direct OPD.
The GRPO-warmed model achieves a STEM average of 44.99. Distillation from this teacher with visual cues yields a student score of 48.14, compared with 43.46 for the base student.
These results suggest that direct OPD can hurt weaker 2B models, while GRPO warm-up makes subsequent distillation more effective.
Response to Reviewer 8P4Y (Q3 and Q4)
Q3. Ablation of the gating mechanism
Thank you for the suggestion. We added per-stage sample pass rates and an ablation of the complete gating mechanism.
(1) Sample pass rates at each gate. See Table R7.
The Difficulty Gate passes 30.53%, 24.02%, and 24.06% of samples for 2B, 4B, and 8B, respectively. It filters overly easy or hard cases, focusing training near each model's capability boundary.
The Answer Gate passes 91.93%, 90.04%, and 88.85%, suggesting that Gemini-3.5-Flash solves most candidate questions correctly and provides an initial check on cue correctness.
The Reliability Gate passes only 58.34%, 70.89%, and 73.91%, lower than the Answer Gate. A correct answer therefore does not ensure cues consistently help teacher reasoning, justifying additional filtering. The higher pass rates for larger models also suggest more reliable use of cues.
Together, the three gates complement one another by checking sample difficulty, answer correctness, and cue reliability, improving the Visual Privilege Library's quality.
(2) Full gating ablation. Tables R8–R9 compare models with and without gating. Gating improves the STEM averages by 5.73, 3.81, and 2.19 points at 2B, 4B, and 8B, and the general-vision averages by 2.43, 2.07, and 1.08 points, respectively. Gains are consistent across scales and task types.
These experiments support the effectiveness of multistage gating. We will include them in the revision.
Table R7. Sample pass rates for the three gates.
| Model | Difficulty Gate | Answer Gate | Reliability Gate |
|---|---|---|---|
| 2B | 30.53% | 91.93% | 58.34% |
| 4B | 24.02% | 90.04% | 70.89% |
| 8B | 24.06% | 88.85% | 73.91% |
Table R8. Gating ablation on STEM visual-reasoning benchmarks. The number of samples without gate matches that with gate, randomly drawn from the dataset.
| Model | Gate | MathVerse$_m^v$ | MathVerse$_m^t$ | DynaMath | WeMath | MathVista$_m$ | MathVision$_m$ | Average |
|---|---|---|---|---|---|---|---|---|
| 2B | without gate | 34.39 | 48.60 | 51.44 | 29.62 | 63.10 | 27.30 | 42.41 |
| 2B | with gate | 45.05 | 58.12 | 54.13 | 38.19 | 67.00 | 26.32 | 48.14 |
| 4B | without gate | 51.27 | 67.77 | 65.05 | 46.00 | 74.80 | 42.11 | 57.83 |
| 4B | with gate | 59.39 | 72.08 | 65.89 | 51.62 | 76.80 | 44.08 | 61.64 |
| 8B | without gate | 57.23 | 70.94 | 68.28 | 51.52 | 76.10 | 43.42 | 61.25 |
| 8B | with gate | 61.55 | 73.10 | 68.82 | 55.33 | 76.80 | 45.07 | 63.44 |
Table R9. Gating ablation on general visual-reasoning benchmarks.
| Model | Gate | MMMU$_{pro}^v$ | MMMU$_{pro}^{std}$ | HallBen | Average |
|---|---|---|---|---|---|
| 2B | without gate | 27.92 | 35.43 | 68.66 | 44.00 |
| 2B | with gate | 31.73 | 37.63 | 69.93 | 46.43 |
| 4B | without gate | 42.72 | 47.63 | 71.82 | 54.06 |
| 4B | with gate | 45.49 | 49.71 | 73.19 | 56.13 |
| 8B | without gate | 50.58 | 52.54 | 74.76 | 59.29 |
| 8B | with gate | 52.49 | 53.76 | 74.87 | 60.37 |
Q4. Answer leakage in visual cues
Thank you for the careful observation.
(1) Answer leakage in Figure 3(b). The cues in Figure 3(b) were generated by the trained student, solely to examine whether it could recover key visual information from the original image. They were not teacher privileges used during training. The student may include a final answer in its description due to limited ability, but this is unrelated to the training cues.
(2) Leakage in training data. Gemini-3.5-Flash generated the training cues under explicit instructions to describe only question-relevant visual facts, excluding reasoning steps and final answers. We tested the entire Visual Privilege Library for leakage (Table R4).
An LLM judge is used to assesse whether a cue directly revealed answer-bearing information; the resulting leakage rate was 4.86%.
Masking-based robustness check. To assess whether the small fraction of potentially answer-leaking training cues affected the results, we masked answers in all flagged cues by replacing answer-bearing keywords or fields with *** while keeping the ground-truth labels unchanged, and reran VC-OPD with the masked cues. Performance was comparable to that with the original unmasked cues, with no notable difference. This indicates that the limited potential answer leakage in the original cue library is not the primary source of VC-OPD’s gains. Together with the leakage analysis in Table R4, it shows that the improvements persist even after suspected answer-revealing content is masked.
These results suggest that our cue-generation approach reduces answer leakage: most training cues do not contain the answer directly, limiting answer-shortcut learning during distillation.
Response to Reviewer 8P4Y (Q5)
Q5. Random-seed variance, student-side recoverability, and train–test contamination
Thank you for the suggestion. We added random-seed experiments, an analysis of student-side recoverability, and checks for train–test contamination.
1. Random-seed variance experiments
To evaluate the stability of VC-OPD, we conducted five independent runs with different random seeds (0–4) on Qwen3-VL-4B, as shown in Tables R10(a)–R10(b). The standard deviations of the aggregate scores are 0.51 and 0.15 for STEM and general visual reasoning, respectively. Most benchmarks have standard deviations below 1.00, indicating that VC-OPD is robust to random-seed variation.
Table R10(a): Random-seed standard deviation on STEM visual-reasoning benchmarks (Qwen3-VL-4B).
| Method | MathVerse$_m^v$ | MathVerse$_m^t$ | DynaMath | WeMath | MathVista$_m$ | MathVision$_m$ | Average |
|---|---|---|---|---|---|---|---|
| VC-OPD | 0.64 | 0.32 | 0.19 | 1.07 | 0.63 | 2.00 | 0.51 |
Table R10(b): Random-seed standard deviation on general visual-reasoning benchmarks (Qwen3-VL-4B).
| Method | MMMU$_{pro}^v$ | MMMU$_{pro}^{std}$ | HallBen | Average |
|---|---|---|---|---|
| VC-OPD | 0.57 | 0.40 | 0.77 | 0.15 |
2. Student-side recoverability
As Figures 3(b) and 5 show, after VC-OPD training, the student can recognize and use visual information it previously struggled to interpret without access to visual cues. This provides qualitative evidence of student-side recoverability.
Student-side recoverability is difficult to measure directly, and no unified reliable metric is yet available. Concurrent state-of-the-art methods, including VAD, Vision-OPD, and ViCuR, also lack an established quantitative measure. We therefore assess it indirectly through pre- versus post-training capability changes.
3. Train–test contamination analysis
We examined potential overlap between ViRL39K and the evaluation benchmarks by matching both image content and question semantics (Table R11). Among 22,301 evaluation samples, only 115 showed potential overlap (0.5157%), indicating limited train–test overlap.
Overall, the low overlap rate suggests a limited risk of train–test contamination and minimal potential impact on the reported results.
Table R11. Potential data contamination between ViRL39K and evaluation benchmarks. Both image content and question semantics are matched.
| Benchmarks | Benchmark Samples | Similar Samples | Proportion of Similar Samples |
|---|---|---|---|
| MathVerse_MINI | 3,940 | 21 | 0.5330% |
| DynaMath | 5,010 | 70 | 1.3972% |
| WeMath | 1,740 | 7 | 0.4023% |
| MathVista_MINI | 1,000 | 17 | 1.7000% |
| MathVision_MINI | 304 | 0 | 0.0000% |
| MMMU_Pro_V | 1,730 | 0 | 0.0000% |
| HallusionBench | 951 | 0 | 0.0000% |
| VStarBench | 191 | 0 | 0.0000% |
| HRBench4K | 800 | 0 | 0.0000% |
| SpatialEval | 4,635 | 0 | 0.0000% |
| PuzzleVQA | 2,000 | 0 | 0.0000% |
| All | 22,301 | 115 | 0.5157% |
Response to Reviewer 8P4Y (Q1)
We thank the reviewer for highlighting our central finding—that transferability of visual cues matters more than teacher improvement alone—and for recognizing consistent gains across model sizes.
Relationship to concurrent work and novelty
We thank the reviewer for pointing out ViCuR, V-Zero, and VAD. These works appeared 1.9 to 3.7 months before the ICLR 2027 deadline and were concurrent with our independent research. We acknowledge that our initial submission lacked sufficient discussion and comparison. We therefore reproduced all three baselines. For ViCuR, we used its released visual cue data and trained with the same configuration as VC-OPD. For V-Zero and VAD, we used their official implementations and original training datasets, changing only the base model. All methods were evaluated under a unified protocol (Tables R5–R6).
We acknowledge that ViCuR has explored question-relevant visual cues as recoverable privileged information. VC-OPD differs in its general cue-generation framework, quality gating, answer leakage control, student architecture compatibility, and pre-distillation teacher enhancement, as detailed below.
1. Cue generation. ViCuR provides limited details on cue generation. VC-OPD uses a Visual-Cue Agent with a unified API for diverse VQA tasks, supported by automated quality checks and gating. Table R4 shows higher teacher scores with VC-OPD cues than with ViCuR cues across 2B, 4B, and 8B models.
2. Answer leakage. According to ViCuR's public issue, cue generation uses full solution steps and answers. In contrast, VC-OPD generates cues using only the image and question, reducing leakage risk by design. Table R4 reports leakage rates of 20.60% for ViCuR and 4.86% for VC-OPD, a 15.74 percentage-point reduction.
3. Student architecture compatibility. ViCuR introduces a SinkTrack-based visual cue recovery module that modifies the student architecture and remains active during inference. This adds inference overhead and complicates adaptation and deployment across different base models. In contrast, VC-OPD transfers the teacher's visual knowledge through on-policy distillation without architectural changes. At inference, the student requires only the original image and question, with no additional modules or associated inference overhead.
4. Teacher enhancement. VC-OPD applies GRPO warm-up before on-policy distillation to improve the teacher's visual STEM reasoning, whereas ViCuR has no corresponding step.
Table R4. Visual-cue quality. Trainset Score is the teacher's training-set accuracy when given visual privileged information.
Table R5. Results on STEM visual-reasoning benchmarks.
Table R6. Results on general visual-understanding benchmarks.