---
title: "OpenCoF：通过视频生成实现推理学习"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.08763v1"
permalink: "https://feed.astralor.com/items/itm_b8f60678adbec8143f5ee99a"
published: "2026-07-09T17:58:29.000Z"
score: 78
channel: "ai"
tags: ["Chain-of-Frame","OpenCoF","Wan-CoF","多模态","推理学习","视频推理"]
---

# OpenCoF：通过视频生成实现推理学习

## AI 点评

> 这是近期最让我兴奋的多模态推理论文之一。它跳出了传统文本CoT的框架，利用视频生成的天生时序优势来做推理，启发意义很大。如果你关注具身智能、多模态推理或视频生成，这篇必看。唯一缺点是基准测试目前还比较小，期待后续扩展到更复杂场景。

## 速览

OpenCoF 是一个针对推理导向视频生成的新框架，包含 OpenCoF-17K 数据集（覆盖11个任务族）和微调模型 Wan-CoF。与传统的 Chain-of-Thought 不同，OpenCoF 通过时序帧链（Chain-of-Frame）进行推理，在四个视频推理基准上显著超越基线 Wan2.2-I2V-A14B。论文还探索了视觉与文本推理令牌机制，分别捕捉低级视觉线索和高级语义先验，以增强空间和时间推理。实验表明，更强大的视频推理需要广泛的时间监督和显式的中间推理状态组织机制。全部数据集、模型和代码已开源。

## 关键点

- 提出 Chain-of-Frame (CoF) 推理范式，通过视频帧时序链替代传统文本 Chain-of-Thought。
- 构建 OpenCoF-17K 数据集与 Wan-CoF 模型，覆盖11个推理任务族。
- 引入视觉与文本推理令牌，分别处理低级视觉线索与高级语义先验。
- 在四个视频推理基准上显著优于基线模型，并开源全部资源。

## 全文

Reasoning has become a core capability for large models, especially when reliable decisions require understanding logical consequences. Recent video generation models offer a reasoning path distinct from previous Chain-of-Thought (CoT): reasoning can unfold through temporally connected frames, known as Chain-of-Frame (CoF) reasoning. However, existing video generators are primarily trained on general video corpora, still lacking diverse supervision and dedicated designs for CoF reasoning. To address this gap, we introduce OpenCoF, a framework comprising the OpenCoF-17K dataset, a reasoning video dataset spanning 11 task families, and Wan-CoF, a fine-tuned video model for studying whether diverse temporal supervision improves CoF behavior. Across four video reasoning benchmarks, Wan-CoF achieves considerable gains over the Wan2.2-I2V-A14B baseline. Building on this, we empirically explore more advanced designs for CoF capabilities, i.e., equipping the model with visual and textual reasoning tokens. This mechanism respectively captures low-level visual cues and high-level semantic priors for spatial and temporal reasoning. Through performance comparisons and attention analysis, we examine how these tokens contribute across model depth, denoising steps, space, and time. Our results suggest that stronger video reasoning requires both broad temporal supervision and explicit mechanisms for organizing intermediate reasoning state. We open-source the dataset, model, and code to facilitate future research on reasoning-oriented video generation.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_b8f60678adbec8143f5ee99a