---
title: "OpenCoF: Learning to Reason Through Video Generation"
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","Multimodal","Reasoning Learning","Video Reasoning"]
---

# OpenCoF: Learning to Reason Through Video Generation

## AI commentary

> This is one of the most exciting multimodal reasoning papers I've seen recently. It breaks the traditional text CoT framework and leverages the inherent temporal advantage of video generation for reasoning, which is highly inspiring. If you are interested in embodied intelligence, multimodal reasoning, or video generation, this is a must-read. The only drawback is that the benchmarks are still small-scale; I look forward to future extensions to more complex scenarios.

## At a glance

OpenCoF is a new framework for reasoning-oriented video generation, comprising the OpenCoF-17K dataset covering 11 task families and the fine-tuned Wan-CoF model. Unlike traditional Chain-of-Thought, OpenCoF performs reasoning through temporally connected frames (Chain-of-Frame), achieving significant gains over the Wan2.2-I2V-A14B baseline on four video reasoning benchmarks. The paper also explores visual and textual reasoning tokens to capture low-level visual cues and high-level semantic priors for spatial and temporal reasoning. Results show that stronger video reasoning requires broad temporal supervision and explicit mechanisms for organizing intermediate reasoning states. The dataset, model, and code are open-sourced.

## Key points

- Introduces Chain-of-Frame (CoF) reasoning paradigm using temporally connected video frames instead of text Chain-of-Thought.
- Builds OpenCoF-17K dataset and Wan-CoF model covering 11 reasoning task families.
- Adds visual and textual reasoning tokens to handle low-level visual cues and high-level semantic priors.
- Significantly outperforms baseline on four video reasoning benchmarks and open-sources all resources.

## Full text

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.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_b8f60678adbec8143f5ee99a