Grounding the Score: Explicit Visual Premise Verification for Reliable Vision-Language Process Reward Models

Junxin Wang; Dai Guan; Weijie Qiu; Zhihang Li; Yongbo Gai; Zhengyi Yang; Mengyu Zhou; Erchao Zhao

cs.CVMar 17, 2026

Grounding the Score: Explicit Visual Premise Verification for Reliable Vision-Language Process Reward Models

Authors: Junxin Wang, Dai Guan, Weijie Qiu, Zhihang Li, Yongbo Gai, Zhengyi Yang, Mengyu Zhou, Erchao Zhao,

Organizations: Qwen Large Model Application Team, Alibaba · Institute of Automation, Chinese Academy of Sciences · Beijing University of Posts and Telecommunications

Abstract

Vision-language process reward models (VL-PRMs) are increasingly used to score intermediate reasoning steps and rerank candidates under test-time scaling. However, they often function as black-box judges: a low step score may reflect a genuine reasoning mistake or simply the verifier's misperception of the image. This entanglement between perception and reasoning leads to systematic false positives (rewarding hallucinated visual premises) and false negatives (penalizing correct grounded statements), undermining both reranking and error localization. We introduce Explicit Visual Premise Verification (EVPV), a lightweight verification interface that conditions step scoring on the reliability of the visual premises a step depends on. The policy is prompted to produce a step-wise visual checklist that makes required visual facts explicit, while a constraint extractor independently derives structured visual constraints from the input image. EVPV matches checklist claims against these constraints to compute a scalar visual reliability signal, and calibrates PRM step rewards via reliability gating: rewards for visually dependent steps are attenuated when reliability is low and preserved when reliability is high. This decouples perceptual uncertainty from logical evaluation without per-step tool calls. Experiments on VisualProcessBench and six multimodal reasoning benchmarks show that EVPV improves step-level verification and consistently boosts Best-of-N reranking accuracy over strong baselines. Furthermore, injecting controlled corruption into the extracted constraints produces monotonic performance degradation, providing causal evidence that the gains arise from constraint fidelity and explicit premise verification rather than incidental prompt effects. Code is available at: https://github.com/Qwen-Applications/EVPV-PRM

Figures & tables

Figure 1: EVPV for reliable multimodal PRMs. (A) A motivating failure case (hallucinated visual premise). (B) Step errors are dominated by visual misinterpretation on VisualProcessBench. (C) EVPV-PRM improves step-level verification (Macro-F1). (D) EVPV-PRM improves Best-of- $8$ reranking (BoN@8) across InternVL2.5 scales.

Figure 2: Overview of EVPV-PRM. Given an image $I$ and question $q$ , the policy model generates a step-by-step solution and, for each step, declares whether it depends on visual evidence, forming a visual checklist of explicit claims. In parallel, a constraint extractor predicts a structured set of visual facts $C$ (numeric readings, geometric relations, and compositional structure). We compute a visual reliability score $r$ by matching checklist claims against $C$ to obtain support scores and aggregating them into a single confidence signal. A step verifier then produces base step rewards, which are calibrated by reliability gating: rewards for non-visual steps are kept unchanged, while rewards for visually dependent steps are down-weighted when $r$ is low and preserved when $r$ is high. The resulting reliability-gated step rewards are aggregated for Best-of- $N$ reranking and process diagnosis.

Figure 3: Training and inference pipeline of EVPV. EVPV trains a constraint extractor $E_{\phi}$ and a step verifier $V_{\theta}$ . The policy is not trained and is only prompted to produce step-wise solutions with a visual checklist at inference time.

Proprietary Models
Model	EVPV	DynaMath	MMMU	MathVerse	MathVision	WeMath	Overall
gpt-4o-mini	✗	$56.57$	$54.08$	$52.53$	$51.42$	$56.74$	$53.57$
	✓	$58.13$	$53.20$	$54.09$	$52.07$	$54.62$	$54.29$
	$\Delta$	+1.56	-0.88	+1.56	+0.65	-2.12	+0.72
doubao-seed-1.6-vision	✗	$66.19$	$59.47$	$63.12$	$61.07$	$62.74$	$62.77$
	✓	$68.66$	$61.86$	$65.57$	$62.51$	$64.62$	$64.91$

Table 1: VisualProcessBench Macro-F1 (%). For external judge models, ✗ uses the judge alone; ✓ attaches the EVPV plug-in (constraints-based reliability + gating, with the same constraints provided as evidence).

\Delta

= ✓

-

✗ (points); positive

\Delta

is shown in blue and negative

\Delta

in red.

Proprietary Models
Model	MathVista	MathVision	MathVerse-VO	WeMath	LogicVista	MMMU	Overall
GPT-4o	60.00	31.20	40.60	45.80	52.80	70.70	47.90
Gemini-2.0-Flash	70.40	43.60	47.80	47.40	52.30	69.90	53.40
Claude-3.5-Sonnet	65.30	35.60	46.30	44.00	60.40	66.40	50.50
Open-source Models
InternVL2.5-8B	64.50	17.00	22.80	23.50	36.38	56.20	32.84

Table 2: Downstream Best-of-8 reranking with InternVL2.5 policies. BoN@8 accuracy (%) after reranking with different PRMs; numbers denote

\Delta_{8}

(BoN@8

-

Pass@1) for our PRM.

Variant	DynaMath	MMMU	MathVerse	MathVision	WeMath	Overall	$\Delta$
Full Method
Full (EVPV + gating)	69.57	68.86	67.09	65.27	69.11	67.46	+0.00
Evidence / structure ablations
w/o structured facts (caption-only)	$67.75$	$58.09$	$63.48$	$60.68$	$67.10$	$63.38$	-4.08
w/o constraints (facts = $\varnothing$ )	$66.66$	$55.80$	$62.61$	$59.13$	$65.81$	$62.11$	-5.35
w/ shuffled facts (structure corrupted)	$62.86$	$52.57$	$59.81$	$58.52$	$64.77$	$59.82$	-7.64

Table 3: Key ablations on VisualProcessBench (Macro-F1; higher is better).

\Delta

is relative to the full method.

Figure 4: Constraint quality–performance causal curves under controlled noise.

Variant	DynaMath	MMMU	MathVerse	MathVision	WeMath	Overall	$\Delta$
Full Method
Full (EVPV + gating)	69.57	68.86	67.09	65.27	69.11	67.46	+0.00
Evidence / structure ablations
w/o structured facts (caption-only)	$67.75$	$58.09$	$63.48$	$60.68$	$67.10$	$63.38$	-4.08
w/o constraints (facts = $\varnothing$ )	$66.66$	$55.80$	$62.61$	$59.13$	$65.81$	$62.11$	-5.35
w/ shuffled facts (structure corrupted)	$62.86$	$52.57$	$59.81$	$58.52$	$64.77$	$59.82$	-7.64

Table 4: Key ablations on VisualProcessBench (Macro-F1; higher is better).

\Delta

is relative to the full method.

Appendix figures & tables18 assets

Supplementary material from the paper’s appendix.

Appendix

Panel of Figure 5 — Figure 5: A representative structured constraint set $\mathcal{C}$ serialized as JSON.

Category	Key fields	Description
numeric	entity , attribute , value , unit , confidence *	A measurable fact associated with a named visual entity. entity is a label or description of the object (e.g., "segment AB" ); attribute names the quantity (e.g., "length" , "angle" , "count" ); value is a numeric literal; unit is optional (e.g., "cm" , "degrees" ).
relation	type , entities , direction , confidence	A geometric or logical relationship between two or more entities. type encodes one of { parallel , perpendicular , equal , subset , incident , adjacent , greater , less }. entities is an ordered list of participants; direction is optional (e.g., "AB $\to$ CD" ).
structure	type , parts , attachment , adjacency , confidence	Compositional or topological description of a multi-part figure. type is one of { composite , graph , table , sequence }. parts lists sub-components; attachment and adjacency are optional relational lists specifying how parts connect.

Appendix

Table 5: Top-level fields for each constraint category. * confidence is a model-estimated reliability weight in

[0,1]

and is used during matching (Appendix B ).

$\tau$	0.3	0.4	0.5	0.6	0.7
Macro-F1	66.91	67.23	67.46	67.18	66.74

Appendix

Table 6: VisualProcessBench overall Macro-F1 (%) under varying reliability threshold

\tau

(

\beta=10

fixed).

Panel of Figure 6 — Figure 6: Policy prompt used to generate step-by-step reasoning traces with explicit visualdependency checklist items.

Panel of Figure 7 — Figure 7: Teacher prompt used to distill pseudo-gold structured constraints $C^{\star}$ for training $E_{\phi}$ .

Panel of Figure 8 — Figure 8: Optional inference-time prompt for generating constraints in JSON form (used only when $E_{\phi}$ is instantiated via prompting).

Panel of Figure 9 — Figure 9: Optional prompt for external step judges in the EVPV plug-in setting. The judge’s output is treated as $R_{t}^{\mathrm{base}}\in\{-1,+1\}$ and then calibrated by EVPV gating.

Panel of Figure 10 — Figure 10: Prompt used for error-type classification of incorrect steps in VisualProcessBench.

	MathVista			MathVision			MathVerse-VO			WeMath			LogicVista			Overall
Strategy	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$
Geometric Mean	64.5	76.3	+11.8	17.0	22.1	+5.1	22.8	29.5	+6.7	23.5	37.5	+14.0	36.4	45.3	+8.9	32.8	41.7	+8.9
Correctness Rate	64.5	75.1	+10.6	17.0	21.4	+4.4	22.8	28.9	+6.1	23.5	36.8	+13.3	36.4	44.6	+8.2	32.8	41.0	+8.2
Streak Score	64.5	74.8	+10.3	17.0	21.9	+4.9	22.8	28.6	+5.8	23.5	36.5	+13.0	36.4	44.3	+7.9	32.8	40.7	+7.9
Weighted Correctness	64.5	73.2	+8.7	17.0	20.5	+3.5	22.8	27.4	+4.6	23.5	35.1	+11.6	36.4	43.1	+6.7	32.8	39.5	+6.7
First-Error Position	64.5	75.7	+11.2	17.0	22.0	+5.0	22.8	29.0	+6.2	23.5	37.1	+13.6	36.4	44.9	+8.5	32.8	41.3	+8.5

Appendix

Table 7: Best-of-8 reranking under five aggregation strategies, InternVL2.5-8B policy. Pass@1 is the same across strategies; BoN@8 and

\Delta_{8}

vary.

	MathVista			MathVision			MathVerse-VO			WeMath			LogicVista			Overall
Strategy	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$
Geometric Mean	68.2	79.6	+11.4	23.4	28.1	+4.7	24.0	32.5	+8.5	30.9	42.1	+11.2	39.6	51.7	+12.1	37.2	46.8	+9.6
Correctness Rate	68.2	78.4	+10.2	23.4	27.5	+4.1	24.0	31.8	+7.8	30.9	41.3	+10.4	39.6	50.9	+11.3	37.2	45.8	+8.6
Streak Score	68.2	78.0	+9.8	23.4	27.2	+3.8	24.0	31.4	+7.4	30.9	41.0	+10.1	39.6	50.5	+10.9	37.2	45.4	+8.2
Weighted Correctness	68.2	76.5	+8.3	23.4	26.0	+2.6	24.0	30.1	+6.1	30.9	39.6	+8.7	39.6	49.1	+9.5	37.2	44.0	+6.8
First-Error Position	68.2	79.0	+10.8	23.4	27.9	+4.5	24.0	32.2	+8.2	30.9	41.7	+10.8	39.6	51.2	+11.6	37.2	46.3	+9.1

Appendix

Table 8: Best-of-8 reranking under five aggregation strategies, InternVL2.5-26B policy.

	MathVista			MathVision			MathVerse-VO			WeMath			LogicVista			Overall
Strategy	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$	P@1	B@8	$\Delta_{8}$
Geometric Mean	71.9	83.5	+11.6	32.2	37.6	+5.4	36.9	47.7	+10.8	38.3	50.0	+11.7	47.9	58.7	+10.8	45.4	55.2	+9.8
Correctness Rate	71.9	82.3	+10.4	32.2	36.8	+4.6	36.9	46.8	+9.9	38.3	49.1	+10.8	47.9	57.8	+9.9	45.4	54.3	+8.9
Streak Score	71.9	81.9	+10.0	32.2	36.4	+4.2	36.9	46.5	+9.6	38.3	48.8	+10.5	47.9	57.5	+9.6	45.4	54.0	+8.6
Weighted Correctness	71.9	80.4	+8.5	32.2	35.0	+2.8	36.9	45.1	+8.2	38.3	47.4	+9.1	47.9	56.1	+8.2	45.4	52.5	+7.1
First-Error Position	71.9	83.0	+11.1	32.2	37.3	+5.1	36.9	47.3	+10.4	38.3	49.6	+11.3	47.9	58.3	+10.4	45.4	54.7	+9.3

Appendix

Table 9: Best-of-8 reranking under five aggregation strategies, InternVL2.5-38B policy.

Split	#Units	#Steps	#ShouldVis	#ModelVis	#Omit	OmissionRate	Completeness
Overall
TOTAL	415	1780	1305	1314	77	0.0590	0.9410
By policy file
InternVL2.5-8B	153	689	453	419	48	0.1060	0.8940
InternVL2.5-26B	159	666	507	518	25	0.0493	0.9507
InternVL2.5-38B	103	425	345	377	4	0.0116	0.9884

Appendix

Table 10: Checklist completeness audit. Under-reporting (omissions) and over-reporting (false positives) of visualdependency compared to an independent lenient visual-necessity annotation.

Type	TP	FP	FN	Prec	Rec	F1
LogicVista
numeric	19	6	3	0.7600	0.8636	0.8085
relation	23	11	12	0.6765	0.6571	0.6667
structure	40	4	4	0.9091	0.9091	0.9091
OVERALL	82	21	19	0.7961	0.8119	0.8039
MMMU

Appendix

Table 11: Human-annotated fidelity of extracted constraints. Precision/Recall/F1 of constraints predicted by

E_{\phi}

, evaluated on 600 sampled instances (534 valid), reported by constraint type and dataset.

Method	Calls per question	Expected cost formula	vs. Baseline
Call breakdown (Extractor / Judge / Tool)
EVPV (ours)	$1$ extractor + $T$ judge + $0$ tool	$\text{Lat}(E_{\phi})+T\!\cdot\!\text{Lat}(J)$	Baseline
TIM-PRM Kuang et al. (2025)	$0$ + $T$ + $p_{\text{tool}}T$	$T\!\cdot\!\text{Lat}(J)+p_{\text{tool}}T\!\cdot\!\text{Lat}(\text{tool})$	+ $p_{\text{tool}}T$ tool calls
EVPV $-$ TIM-PRM ( $\Delta$ )	$+1$ extractor, $-\,p_{\text{tool}}T$ tools	$\text{Lat}(E_{\phi})-p_{\text{tool}}T\!\cdot\!\text{Lat}(\text{tool})$	Saves $p_{\text{tool}}T$ tools

Appendix

Table 12: Inference-time cost comparison (per question).

Group	Variant	DynaMath	MMMU	MathVerse	MathVision	WeMath	Overall	$\bm{\Delta}$
Full Method (Reference)
Full Method	Full (EVPV + gating)	69.57	68.86	67.09	65.27	69.11	67.46	+0.00
Evidence type
Evidence type	w/o structured facts (caption-only)	67.75	58.09	63.48	60.68	67.10	63.38	$-$ 4.08
Evidence type	w/o constraints (facts = $\emptyset$ )	66.66	55.80	62.61	59.13	65.81	62.11	$-$ 5.35
Evidence type	w/ shuffled facts (structure corrupted)	62.86	52.57	59.81	58.52	64.77	59.82	$-$ 7.64

Appendix

Table 13: Complete ablation results on VisualProcessBench (Macro-F1, %).

\Delta

= variant

-

Full Method. Best per group is highlighted by the best score in that group (see group-wise shading).

Process Reward Models

Process reward models assess intermediate reasoning steps and actions to guide learning, search, and evaluation.

See all 140 papers

Improving Vision-language Models with Perception-centric Process Reward Models
Apr 27, 2026Yingqian Min, Kun Zhou, Yifan Li +6Process Reward Models Object Hallucination in VLMs
Gaoling School of Artificial Intelligence, Renmin University of China. · Bytedance. · University of California, San Diego. +1
V-Rubrics: Visual Faithfulness via Rubric-Based Reinforcement Learning
Aug 26, 2026Shulin Tian, Minglun Li, Yuhao Dong +6Faithfulness in VLMs RL for VLMs
S-Lab, Nanyang Technological University · A*STAR · Independent Researcher +1
Seeing is not Enough: Vision-Language Models Perceive Evidence but Fail to Act
Sep 18, 2026Yuyang Dai, Bofei Huang, Hongbo Zhang +1VLM Reasoning Visually Grounded Reasoning
Florida State University

Apr 27, 2026J/K move · Enter open · S save

Improving Vision-language Models with Perception-centric Process Reward Models

Yingqian Min, Kun Zhou, Yifan Li, Yuhuan Wu+5

Gaoling School of Artificial Intelligence, Renmin University of China. · Bytedance. · University of California, San Diego.+1

Process Reward Models Object Hallucination in VLMs VLM Reasoning Visually Grounded Reasoning RL for VLMs

Open paper AskPDF arXiv

Recent advancements in reinforcement learning with verifiable rewards (RLVR) have significantly improved the complex reasoning ability of vision-language models (VLMs). However, its outcome-level supervision is too coarse to diagnose and correct errors within the reasoning chain. To this end, we propose Perceval, a process reward model (PRM) that enables token-level error grounding, which can extract image-related claims from the response and compare them one by one with the visual evidence in the image, ultimately returning claims that contain perceptual errors. Perceval is trained with perception-intensive supervised training data. We then integrate Perceval into the RL training process to train the policy models. Specifically, compared to traditional GRPO, which applies sequence-level advantages, we apply token-level advantages by targeting penalties on hallucinated spans identified by Perceval, thus enabling fine-grained supervision signals. In addition to augmenting the training process, Perceval can also assist VLMs during the inference stage. Using Perceval, we can truncate the erroneous portions of the model's response, and then either have the model regenerate the response directly or induce the model to reflect on its previous output. This process can be repeated multiple times to achieve test-time scaling. Experiments show significant improvements on benchmarks from various domains across multiple reasoning VLMs trained with RL, highlighting the promise of perception-centric supervision as a general-purpose strategy. For test-time scaling, it also demonstrates consistent performance gains over other strategies, such as major voting. Our code and data will be publicly released at https://github.com/RUCAIBox/Perceval.

Abstract

Process Reward Models

Explore similar work

Improving Vision-language Models with Perception-centric Process Reward Models

V-Rubrics: Visual Faithfulness via Rubric-Based Reinforcement Learning

Seeing is not Enough: Vision-Language Models Perceive Evidence but Fail to Act