paper-with-me

Papers

Skywork-R1V3 Technical Report

2025-07-08 · Wei Shen, Jiangbo Pei, Yi Peng, Xuchen Song, Yang Liu, Jian Peng, Haofeng Sun, Yunzhuo Hao, Peiyu Wang, Yahui Zhou

We introduce Skywork-R1V3, an advanced, open-source vision-language model (VLM) that pioneers a new approach to visual reasoning. Its key innovation lies in effectively transferring reasoning skills from text-only Large Language Models (LLMs) to visual tasks. The strong performance of Skywork-R1V3 primarily stems from our elaborate post-training RL framework, which effectively activates and enhances the model's reasoning ability, without the need for additional continue pre-training. Through this framework, we further uncover the fundamental role of the connector module in achieving robust cross-modal alignment for multimodal reasoning models. In addition, we introduce a unique indicator of reasoning capability, the entropy of critical reasoning tokens, which has proven highly effective for checkpoint selection during RL training. Skywork-R1V3 achieves state-of-the-art results on MMMU, significantly improving from 64.3% to 76.0%. This performance matches entry-level human capabilities. Remarkably, our RL-powered post-training approach enables even the 38B parameter model to rival top closed-source VLMs. The implementation successfully transfers mathematical reasoning to other subject-related reasoning tasks. We also include an analysis of curriculum learning and reinforcement finetuning strategies, along with a broader discussion on multimodal reasoning. Skywork-R1V3 represents a significant leap in multimodal reasoning, showcasing RL as a powerful engine for advancing open-source VLM capabilities.

📄 PDF Abstract BibTeX arXiv:2507.06167

Code (2)

SkyworkAI/Skywork-R1V 공식 구현 pytorch
seephys/seephys-project pytorch

Tasks

cross-modal alignmentMathematical ReasoningMultimodal ReasoningVisual Reasoning

Similar Papers 제목 키워드 기반

Skywork Open Reasoner 1 Technical Report

2025-05-28 · Jujie He, Jiacai Liu, Chris Yuhao Liu, Rui Yan 외

The success of DeepSeek-R1 underscores the significant role of reinforcement learning (RL) in enhancing the reasoning capabilities of large language models (LLMs). In this work, we present Skywork-OR1, an effective and s…

MathReinforcement Learning (RL)

Skywork-MoE: A Deep Dive into Training Techniques for Mixture-of-Experts Language Models

2024-06-03 · Tianwen Wei, Bo Zhu, Liang Zhao, Cheng Cheng 외

In this technical report, we introduce the training methodologies implemented in the development of Skywork-MoE, a high-performance mixture-of-experts (MoE) large language model (LLM) with 146 billion parameters and 16 e…

Language ModelingLanguage ModellingLarge Language ModelMixture-of-Experts

Skywork: A More Open Bilingual Foundation Model

2023-10-30 · Tianwen Wei, Liang Zhao, Lichang Zhang, Bo Zhu 외

In this technical report, we present Skywork-13B, a family of large language models (LLMs) trained on a corpus of over 3.2 trillion tokens drawn from both English and Chinese texts. This bilingual foundation model is the…

Language ModelingLanguage Modellingmodel

SkyMath: Technical Report

2023-10-25 · Liu Yang, Haihua Yang, Wenjun Cheng, Lei Lin 외

Large language models (LLMs) have shown great potential to solve varieties of natural language processing (NLP) tasks, including mathematical reasoning. In this work, we present SkyMath, a large language model for mathem…

GSM8KLanguage ModelingLanguage ModellingLarge Language Model+1

Skywork-Reward: Bag of Tricks for Reward Modeling in LLMs

2024-10-24 · Chris Yuhao Liu, Liang Zeng, Jiacai Liu, Rui Yan 외

In this report, we introduce a collection of methods to enhance reward modeling for LLMs, focusing specifically on data-centric techniques. We propose effective data selection and filtering strategies for curating high-q…

Position