Evidence-Aligned Multimodal On-Policy Self-Distillation for Fine-Grained Visual Understanding

Nanxing Hu; Qiwei Yan; Jinchao Zhang; Guoliang Kang

cs.CVSep 28, 2026

Evidence-Aligned Multimodal On-Policy Self-Distillation for Fine-Grained Visual Understanding

Authors: Nanxing Hu, Qiwei Yan, Jinchao Zhang, Guoliang Kang

Organizations: Beihang University · University of the Chinese Academy of Sciences

Abstract

Fine-grained visual understanding requires models to recognize small details within complex images. Multimodal on-policy self-distillation (OPSD) addresses this challenge by using a teacher conditioned on evidence-centered crops to supervise a student conditioned on original images along student-generated trajectories. Ideally, teacher corrections, the distributional changes from the student toward the privileged teacher, should be driven by task-relevant visual evidence. However, the designs that make the teacher effective also introduce other interference. Using a lagged or frozen teacher improves training stability but introduces a model-state gap from the evolving student, while cropping enhances task-relevant evidence but also loses the visual context. These two sources of interference make the teacher corrections not purely rely on the visual evidence. We introduce Evidence-Aligned multimodal on-policy self-Distillation (EAD), which retains the crop-conditioned teacher as the target but constructs a separate evidence reference for weighting the corrections. To exclude the effect of lagged model-state from this reference, EAD measures prediction changes using the current student. To avoid crop-induced context changes, EAD masks the evidence region in the original image while preserving the other visual context. The change from the student's masked-image prediction to its original-image prediction provides a controlled reference for the direction in which the visual evidence shifts the student's prediction. EAD weights each teacher correction by its cosine alignment with the reference, i.e., retaining aligned corrections and downweighting the rest. Retaining only 6% of the supervision mass of dense OPSD, EAD consistently outperforms previous state-of-the-art methods.

Figures & tables

Figure 1: Motivation and overview of EAD. The privileged teacher correction in multimodal OPSD contains both the desired evidence-driven signal and interference arising from model-state gap and crop-induced context changes (top left). EAD constructs a separate student-side evidence reference from the current student’s prediction shift induced by evidence restoration, which keeps the model state fixed and preserves the surrounding visual context (top right). EAD then weights teacher corrections by the alignment with this evidence reference, concentrating supervision on visually supported corrections (bottom).

Figure 2: Diagnosing interference in teacher corrections. (a) Total Variation (TV) distance between the normalized token-wise JS distributions of $D_{t}^{\mathrm{teacher}}$ and $D_{t}^{\mathrm{crop}}$ over 512 images, showing the effect of model-state gap across training. (b) TV distance between $D_{t}^{\mathrm{crop}}$ and $D_{t}^{\mathrm{mask}}$ under the current student, showing the difference between crop-induced changes and evidence removal. (c) The corresponding token-wise JS distributions for one example. $D_{t}^{\mathrm{mask}}$ places greater emphasis on tokens associated with the task-relevant visual evidence. $D_{t}^{\mathrm{crop}}$ shows large divergence on less vision-relevant tokens.

Figure 3: Overview of EAD. The crop-conditioned EMA teacher provides the privileged target $p_{\bar{\theta},t}^{C}$ , while the current student produces $p_{\theta,t}^{R}$ and $p_{\theta,t}^{R^{-}}$ on the original and masked images, respectively. The Original–Mask prediction change $p_{\theta,t}^{R}-p_{\theta,t}^{R^{-}}$ forms the evidence reference. EAD compares the teacher correction and evidence reference after a shared JS-motivated normalization, and uses their clipped cosine similarity to weight the distillation loss.

Closed-Source Models (Single Forward Pass)
Model	Params	V*Bench	ZoomBench	HR-Bench		MME-RealWorld		Avg.
				4K	8K	EN	CN
GPT-5.2 ( OpenAI, 2025 )	–	79.06	50.89	81.12	78.38	72.60	68.80	71.81
GPT-5.4 ( OpenAI, 2026 )	–	76.96	52.66	84.00	77.88	74.20	70.93	72.77
Gemini-3.1-Pro ( Google, 2026a )	–	87.96	61.18	89.63	86.88	76.53	73.31	79.25
Gemini-3.5-Flash ( Google, 2026b )	–	89.01	61.42	89.12	86.62	75.31	73.97	79.24

Table 1: Fine-grained visual understanding. We report results on six fine-grained visual benchmarks under the single-forward-pass evaluation setting. Results for closed-source and open-weight models are taken from Yuan et al. (2026) , while all Qwen3.5-4B/9B-based methods are obtained from our evaluation. All post-training methods use the same backbone and training data, are trained for one epoch, and are evaluated using the final checkpoint. Avg. is the unweighted mean over the six benchmarks. Best results within each Qwen3.5 scale are shown in bold.

Model State	V*Bench	ZoomBench	HR-Bench		MME-RealWorld		Avg.
			4K	8K	EN	CN
Cross-state	89.53	59.05	85.12	81.38	71.05	69.75	75.98
Teacher-side	89.01	59.76	85.25	83.62	73.18	71.35	77.03
Student-side (EAD)	91.62	60.83	86.38	82.88	72.39	71.10	77.53

Table 2: Effect of model state on the evidence reference. All variants use the same Original–Mask image pair. Cross-state uses different model states for the two predictions, whereas Teacher-side and Student-side (EAD) use the EMA teacher and current student alone, respectively. Avg. denotes the unweighted mean over the six benchmarks. Best results in each column are shown in bold.

Visual Contrast	V*Bench	ZoomBench	HR-Bench		MME-RealWorld		Avg.
			4K	8K	EN	CN
Original–Crop	88.48	60.00	83.00	80.25	70.61	68.62	75.16
Original–Mask (EAD)	91.62	60.83	86.38	82.88	72.39	71.10	77.53
Crop–Downsampled Crop (VAD)	91.62	58.82	86.50	83.00	69.33	68.73	76.33

Table 3: Effect of visual contrast on the evidence reference. The first two variants use the current student and differ only in the visual inputs, while the VAD-style variant uses the EMA teacher on the cropped image and its downsampled counterpart. All variants use the same EAD objective. Avg. denotes the unweighted mean over the benchmarks. Best results in each column are shown in bold.

Variant	V*Bench	ZoomBench	HRBench-4K	HRBench-8K	MME-EN	MME-CN	Avg.
No JS Scaling	89.01	59.29	85.12	82.38	72.37	71.29	76.58
Shifted-Cosine	90.58	61.18	83.50	81.25	72.22	70.05	76.46
EAD	91.62	60.83	86.38	82.88	72.39	71.10	77.53

Table 4: Effect of local JS scaling and clipped cosine similarity. No JS Scaling removes the local normalization before computing directional alignment, while Shifted-Cosine retains negatively aligned teacher corrections.

Figure 4: Sparsity and concentration of EAD supervision. (a) Fraction of the original distillation-loss mass retained by EAD throughout training. (b) Concentration of the retained loss across response tokens, compared with the original dense distillation loss on the same responses. (c) On-policy rollout accuracy of EAD and Vision-OPD during training. Results are shown for Qwen3.5-4B and Qwen3.5-9B.

Appendix figures & tables7 assets

Supplementary material from the paper’s appendix.

Appendix

Method	MMVP	CV-Bench	MMStar	POPE	Avg.
Qwen3.5-4B Post-Training Strategies
Vanilla	79.00	86.97	74.00	89.33	82.33
GRPO ( Shao et al., 2024 )	78.00	85.78	69.33	87.87	80.25
Vision-OPD ( Yuan et al., 2026 )	76.67	85.48	70.80	88.99	80.49
OPD-V ( Bi et al., 2026 )	77.00	84.38	71.40	89.43	80.55
VAD ( Zhang et al., 2026 )	79.67	87.30	71.80	88.79	81.89

Appendix

Table 5: Preservation of broader visual capabilities. We evaluate the same 4B and 9B post-training checkpoints from Table 1 on four held-out benchmarks without additional training. Avg. denotes the unweighted mean over the four benchmarks. Best results within each model scale are shown in bold.

Benchmark	Vision-OPD 4B
V*Bench	90.05
ZoomBench	59.29
HRBench-4K	82.25
HRBench-8K	79.25
MME-RealWorld-EN	70.71
MME-RealWorld-CN	69.43

Appendix

Table 6: Released Vision-OPD 4B checkpoint under our evaluation pipeline. All results are obtained using the same revised evaluator applied to the Qwen3.5-based models in this work.

Statistic	4B	9B
Cumulative retained loss mass	5.87	6.15
Top-10% loss mass: EAD	95.98	96.81
Top-10% loss mass: dense	61.10	62.27
Training rollout accuracy: EAD	59.53	64.31
Training rollout accuracy: Vision-OPD	55.06	57.49

Appendix

Table 7: Summary of supervision retention, concentration, and rollout accuracy. All values are percentages. Top-10% loss mass measures how much of each response’s total loss comes from its 10% highest-loss tokens, averaged across responses. Rollout accuracy is computed over all responses from training steps 1–65.

Evidence Reference	MMVP	CV-Bench	MMStar	POPE	Holdout Avg.	Fine Avg.	All-10
Cross-state masking	77.00	85.49	73.53	86.66	80.67	75.98	77.86
Crop-induced change	78.33	86.61	70.87	89.11	81.23	75.16	77.59
Teacher-side masking	78.67	87.59	72.67	89.24	82.04	77.03	79.03
Teacher crop degradation	76.33	84.15	71.87	83.30	78.91	76.33	77.37
Current student (EAD)	80.33	86.95	73.07	87.70	82.01	77.53	79.33

Appendix

Table 8: Held-out generalization of evidence-reference variants. The same checkpoints evaluated in Tables 2 and 3 are evaluated on MMVP, CV-Bench, MMStar, and POPE. Holdout Avg. is the unweighted mean over the four held-out benchmarks, Fine Avg. is the mean over the six fine-grained benchmarks, and All-10 averages all ten benchmarks.

Benchmark	Vanilla Cosine	EAD
V*Bench	89.01	91.62
ZoomBench	59.29	60.83
HRBench-4K	85.12	86.38
HRBench-8K	82.38	82.88
MME-RealWorld-EN	72.37	72.39
MME-RealWorld-CN	71.29	71.10

Appendix

Table 9: Effect of local JS scaling. Vanilla cosine directly compares the unscaled probability differences, whereas EAD applies the shared local normalization

1/\sqrt{\max(p_{\theta,t}^{R},\epsilon)}

before measuring directional alignment. Both variants otherwise use the same evidence reference and training procedure.

Benchmark	Shifted-Cosine	EAD
V*Bench	90.58	91.62
ZoomBench	61.18	60.83
HRBench-4K	83.50	86.38
HRBench-8K	81.25	82.88
MME-RealWorld-EN	72.22	72.39
MME-RealWorld-CN	70.05	71.10

Appendix

Table 10: Effect of excluding negatively aligned corrections. Shifted-Cosine retains negatively aligned corrections by mapping cosine similarity from

[-1,1]

[0,1]

. All other settings are unchanged.

Figure 5: Additional token-level case studies. Each example shows the original, cropped, and masked images together with the normalized token-wise JS divergence of $D_{t}^{\mathrm{teacher}}$ , $D_{t}^{\mathrm{crop}}$ , and $D_{t}^{\mathrm{mask}}$ . Across different fine-grained visual tasks, $D_{t}^{\mathrm{mask}}$ tends to emphasize tokens more closely associated with the task-relevant visual evidence, while the other comparisons can assign large divergence to less directly related tokens.

Explore similar work

VAD: Attributing Visual Evidence for Target Reconstruction in Multimodal On-Policy Distillation
Jul 30, 2026Kangning Zhang, Yixing Li, Shuai Shao +9Dataset Distillation Visual Evidence
Shanghai Jiao Tong University · Xiaohongshu Inc. · The Chinese University of Hong Kong +1
Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation
May 18, 2026Qianhao Yuan, Jie Lou, Xing Yu +4Recent Vision-Language Models Multimodal Large Language Models
Chinese Information Processing Laboratory, Institute of Software, Chinese Academy of Sciences · University of Chinese Academy of Sciences · Xiaohongshu Inc.
OPD-Aha: From Linguistic Momentum to Visual Reflection in Multimodal On-Policy Distillation
Sep 15, 2026Chenhao Qiu, Dawei Li, Yechao Zhang +2Multimodal Reasoning Teacher
Arizona State University · University of Virginia · Stevens Institute of Technology

Jul 30, 2026, cs.CVJ/K move · Enter open · S save

VAD: Attributing Visual Evidence for Target Reconstruction in Multimodal On-Policy Distillation

Kangning Zhang, Yixing Li, Shuai Shao, Qingyao Li+8

Shanghai Jiao Tong University · Xiaohongshu Inc. · The Chinese University of Hong Kong+1

Dataset Distillation Visual Evidence Teacher

Open paper AskPDF arXiv

Multimodal on-policy distillation (OPD) transfers fine-grained visual knowledge by supervising student-generated trajectories with a privileged-view teacher. Yet its next-token corrections are source-mixed, combining visual signals with linguistic priors and teacher-specific effects. The key challenge is to estimate which corrections are supported by visual evidence, not merely where or how strongly to distill. We introduce Visual Attribution Distillation (VAD), a counterfactual target-reconstruction algorithm that estimates the visually attributable part of a teacher correction. At each student-generated prefix, VAD evaluates the same fixed teacher with the relevant evidence present and removed. The corresponding change in centered log-probabilities defines ut, a signed proxy for the visual evidence direction that estimates how revealing the evidence supports or refutes candidate tokens. VAD projects the original correction onto this proxy to obtain an intervention-aligned component and a proxy-unexplained residual, then reconstructs a student-anchored target from the former. During training, this reconstructed target supplies the primary supervision signal, while the privileged teacher contributes a weak regularizer. Across six fine-grained visual benchmarks at 4B and 9B scales, VAD outperforms direct privileged-view distillation and visual-advantage weighting. Token- level and controlled-target analyses show that the proxy-aligned component is enriched in task-relevant visual corrections and yields stronger target shifts, especially when evidence refutes a mistaken answer. These results support counterfactual target reconstruction as an effective alternative to source-mixed supervision.

Abstract

Explore similar work

VAD: Attributing Visual Evidence for Target Reconstruction in Multimodal On-Policy Distillation

Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation

OPD-Aha: From Linguistic Momentum to Visual Reflection in Multimodal On-Policy Distillation