arXiv 2609.34309 · cs.CV · 2026-09-28
MaLiang-Harness: A Programmable Path to Image and Video Generation
Haoyu Zhao, Zihao Zhang, Xudong Wang, Jiaxi Gu, Zuxuan Wu, Yu-Gang Jiang, Shuicheng Yan
National University of Singapore
在最新榜 #28 2026-W40 第 7 名 本站收录于 2026-10-21
一句话要点
MaLiang-Harness 让多模态模型写程序生成图像和视频，并在“构建-检查-修改”循环中缩小程序与画面的差距。
MaLiang-Harness has MLLMs generate images and videos via programs in a build-inspect-revise loop that closes the program-to-visual gap.
阅读 arXiv 原文 PDF Hugging Face 讨论 代码仓库 项目页
主题与学科
同主题：视觉生成
- Any Time, Any Space: A World-State-Centric Framework for Agentic Video Generation ▲364
- DiffusionART: Efficient, Stable and Scalable Reinforcement of Diffusion Models ▲325
- UltraText Bench: A Comprehensive Bilingual Benchmark for Evaluating Visual Text Rendering in Image Generation ▲197
- Matrix-Game 4.0: Harnessing Video Models for Real-Time Playable Game Worlds ▲182
- Kandinsky 6.0 Video: Foundation Models for Synchronized Video and Audio Generation ▲174
数据说明
元数据来自 arXiv API（已核实）；点赞来自 Hugging Face 每日论文；引用来自 Semantic Scholar。最近刷新 2026-10-22。
要点由本站根据摘要自写，不转载摘要原文，请阅读原文。