Pith. sign in

REVIEW 7 cited by

Skywork R1V2: Multimodal Hybrid Reinforcement Learning for Reasoning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2504.16656 v4 pith:4DMP4G5L submitted 2025-04-23 cs.CV

classification cs.CV
keywords r1v2skyworkoptimizationreasoningreinforcementgrpohybridlearning
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We present Skywork R1V2, a next-generation multimodal reasoning model and a major leap forward from its predecessor, Skywork R1V. At its core, R1V2 introduces a hybrid reinforcement learning paradigm that jointly leverages the Mixed Preference Optimization (MPO) and the Group Relative Policy Optimization (GRPO), which harmonizes reward-model guidance with rule-based strategies, thereby addressing the long-standing challenge of balancing sophisticated reasoning capabilities with broad generalization. To further enhance training efficiency, we propose the Selective Sample Buffer (SSB) mechanism, which effectively addresses the vanishing advantages dilemma inherent in GRPO by prioritizing high-value samples throughout the optimization process. Notably, we observe that excessive reinforcement signals can induce visual hallucinations--a phenomenon we systematically monitor and mitigate through calibrated reward thresholds throughout the training process. Empirical results affirm the exceptional capability of R1V2, with benchmark-leading performances such as 62.6 on OlympiadBench, 78.9 on AIME2024, 63.6 on LiveCodeBench, and 73.6 on MMMU. These results underscore R1V2's superiority over existing open-source models and demonstrate significant progress in closing the performance gap with premier proprietary systems, including Gemini 2.5 and OpenAI-o4-mini. The Skywork R1V2 model weights have been publicly released to promote openness and reproducibility https://huggingface.co/Skywork/Skywork-R1V2-38B.

Discussion (0). Sign in to comment.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. VL-Cogito: Progressive Curriculum Reinforcement Learning for Advanced Multimodal Reasoning

    cs.CV 2025-07 conditional novelty 6.0 of 10

    VL-Cogito, trained with progressive curriculum RL, online difficulty weighting, and dynamic length rewards, matches or beats prior reasoning MLLMs on ten multimodal benchmarks.

  2. The Synergy Dilemma of Long-CoT SFT and RL: Investigating Post-Training Techniques for Reasoning VLMs

    cs.CL 2025-07 conditional novelty 6.0 of 10

    Long-CoT SFT and RL trade off strengths in reasoning VLMs, and five combination strategies fail to produce additive gains.

  3. Rex-Thinker: Grounded Object Referring via Chain-of-Thought Reasoning

    cs.CV 2025-06 conditional novelty 6.0 of 10

    Training a multimodal LLM on GPT-4o-generated chain-of-thought referring traces, then optimizing with GRPO, improves referring accuracy and abstention on HumanRef.

  4. MedBookVQA: A Systematic and Comprehensive Medical Benchmark Derived from Open-Access Book

    cs.AI 2025-06 conditional novelty 6.0 of 10

    MedBookVQA is a new 5,000-question, textbook-derived multimodal benchmark for testing medical AI systems, with labels for imaging modality, body anatomy, and clinical specialty.

  5. Skywork-R1V3 Technical Report

    cs.CL 2025-07 conditional novelty 5.0 of 10

    A 38B open-source VLM reaches 76.0% on MMMU using RL post-training and connector-only tuning, with a critical-token entropy metric for checkpoint selection.

  6. FinLMM-R1: Enhancing Financial Reasoning in LMM through Scalable Data and Reward Design

    cs.CL 2025-06 conditional novelty 5.0 of 10

    A two-stage RL framework with length, image-selection, and adversarial rewards, trained on 89,378 ASP-built financial image-question pairs, improves multimodal reasoning over LMM-R1.

  7. WeThink: Toward General-purpose Vision-Language Reasoning via Reinforcement Learning

    cs.CV 2025-06 conditional novelty 5.0 of 10

    WeThink, a 120K-image QA dataset with AI-generated reasoning paths, combined with hybrid-reward reinforcement learning, improves a 7B vision-language model across 14 benchmarks.

Pith tools