Liu-Junhua
/

TwiFF-7B

visual-chain-of-thought

dynamic-visual-reasoning

Model card Files Files and versions

Liu-Junhua commited on Feb 12

Commit

b0aba2b

·

verified ·

1 Parent(s): 653227b

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -38,7 +38,7 @@ pipeline_tag: any-to-any
 ## 🧠 Method
-<p align="center"><img src="https://github.com/LiuJunhua02/TwiFF/tree/main/assets/data_show.png" width="95%"></p>
 We present TwiFF, a unified model fine-tuned on a high-quality dynamic visual Chain-of-Thought (VCoT) dataset comprising 2.7 million samples. In dynamic multimodal question-answering tasks involving instructional, predictive, and camera, TwiFF iteratively generates future event frames alongside textual reasoning, thereby producing temporally coherent visual reasoning trajectories. Experimental results demonstrate that, on dynamic scenario reasoning benchmarks, our dynamic VCoT approach outperforms both static VCoT methods based on tool-calling paradigms and purely textual chain-of-thought baselines.
 ## ✍️ Citation

 ## 🧠 Method
+<p align="center"><img src="https://github.com/LiuJunhua02/TwiFF/raw/main/assets/data_show.png" width="95%"></p>
 We present TwiFF, a unified model fine-tuned on a high-quality dynamic visual Chain-of-Thought (VCoT) dataset comprising 2.7 million samples. In dynamic multimodal question-answering tasks involving instructional, predictive, and camera, TwiFF iteratively generates future event frames alongside textual reasoning, thereby producing temporally coherent visual reasoning trajectories. Experimental results demonstrate that, on dynamic scenario reasoning benchmarks, our dynamic VCoT approach outperforms both static VCoT methods based on tool-calling paradigms and purely textual chain-of-thought baselines.
 ## ✍️ Citation