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

Hugging Face Daily Papers Papers
visual-faithfulness rubric-based-reinforcement-learning vision-language-models grounding reinforcement-learning academic-research

Summary

Introduces a rubric-based reinforcement learning method to improve visual faithfulness in vision-language models by decomposing responses into atomic propositions and scoring them on visual faithfulness, reasoning consistency, and instruction following.

Vision-language models can produce fluent answers that are insufficiently grounded in the visual evidence: a single unsupported object, chart value, or intermediate inference can undermine an otherwise plausible response. We argue that this is a credit-assignment failure in multimodal post-training. Scalar outcome rewards indicate whether an answer is acceptable, but do not identify which visual facts are grounded, which reasoning steps are valid, or which instruction constraints are missed. We introduce Visual Rubrics-Based Reinforcement Learning, which decomposes reference responses into atomic propositions and scores generated answers along Visual Faithfulness (VF), Reasoning Consistency (RC), and Instruction Following (IF). The resulting rubric items provide structured partial credit and localize rubric credit when supporting evidence spans are available. We first obtain an SFT checkpoint by fine-tuning Qwen3-VL-8B-Instruct on the public OpenMMReasoner-SFT-874K corpus, adapting OpenMMReasoner's cold-start data recipe. We construct V-Rubrics 50K, a 50,248-example training set from 17 visually grounded sources, by applying rule-based filters before deriving example difficulty from rejection-sampling scores and then annotating every example with Gemini-3-Pro under the same structured prompt and protocol. We train our model based on the same SFT checkpoint using component-wise, prefix-localized rubric credit. Experiments show that our rubricbased GRPO improves over both the shared SFT baseline and answer-only GRPO, with the largest gains on knowledge-oriented and visually grounded reasoning benchmarks. The results show rubrics as a useful reward abstraction for visual post-training.
Original Article
View Cached Full Text

Cached at: 08/27/26, 07:21 AM

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

Source: https://huggingface.co/papers/2608.25580

Abstract

Visual Rubrics-Based Reinforcement Learning improves vision-language model grounding by scoring answers on visual faithfulness, reasoning consistency, and instruction following using structured partial credit.

Vision-language models can produce fluent answers that are insufficiently grounded in the visual evidence: a single unsupported object, chart value, or intermediate inference can undermine an otherwise plausible response. We argue that this is a credit-assignment failure in multimodal post-training. Scalar outcome rewards indicate whether an answer is acceptable, but do not identify which visual facts are grounded, which reasoning steps are valid, or which instruction constraints are missed. We introduceVisual Rubrics-Based Reinforcement Learning, which decomposes reference responses into atomic propositions and scores generated answers alongVisual Faithfulness(VF),Reasoning Consistency(RC), andInstruction Following(IF). The resulting rubric items provide structured partial credit and localizerubric creditwhen supporting evidence spans are available. We first obtain an SFT checkpoint by fine-tuningQwen3-VL-8B-Instructon the publicOpenMMReasoner-SFT-874K corpus, adaptingOpenMMReasoner’s cold-start data recipe. We constructV-Rubrics 50K, a 50,248-example training set from 17 visually grounded sources, by applying rule-based filters before deriving example difficulty from rejection-sampling scores and then annotating every example with Gemini-3-Pro under the same structured prompt and protocol. We train our model based on the same SFT checkpoint using component-wise, prefix-localizedrubric credit. Experiments show that our rubricbasedGRPOimproves over both the shared SFT baseline and answer-onlyGRPO, with the largest gains on knowledge-oriented and visually grounded reasoning benchmarks. The results show rubrics as a useful reward abstraction for visual post-training.

View arXiv pageView PDFProject pageGitHub4Add to collection

Get this paper in your agent:

hf papers read 2608\.25580

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2608.25580 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.25580 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.25580 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

RUBAS: Rubric-Based Reinforcement Learning for Agent Safety

arXiv cs.LG

RUBAS is a rubric-based reinforcement learning framework for agent safety that decomposes LLM agent behavior into four dimensions—tool-use safety, argument safety, response safety, and helpfulness—providing fine-grained rewards over complete trajectories. Experiments show RUBAS improves safety over standard alignment baselines while reducing tool-grounded hallucinations and maintaining competitive utility.

RULER: Instance-aware Rubric Rewards for SVG Generation

Hugging Face Daily Papers

RULER introduces instance-aware rubric rewards for SVG generation, using a vision-language judge to optimize reinforcement learning and significantly improve performance over previous methods.