Enhancing Video-LLM Reasoning via Agent-of-Thoughts Distillation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929712530456576 |
|---|---|
| author | Shi, Yudi Di, Shangzhe Chen, Qirui Xie, Weidi |
| author_facet | Shi, Yudi Di, Shangzhe Chen, Qirui Xie, Weidi |
| contents | This paper tackles the problem of video question answering (VideoQA), a task that often requires multi-step reasoning and a profound understanding of spatial-temporal dynamics. While large video-language models perform well on benchmarks, they often lack explainability and spatial-temporal grounding. In this paper, we propose Agent-of-Thoughts Distillation (AoTD), a method that enhances models by incorporating automatically generated Chain-of-Thoughts (CoTs) into the instruction-tuning process. Specifically, we leverage an agent-based system to decompose complex questions into sub-tasks, and address them with specialized vision models, the intermediate results are then treated as reasoning chains. We also introduce a verification mechanism using a large language model (LLM) to ensure the reliability of generated CoTs. Extensive experiments demonstrate that AoTD improves the performance on multiple-choice and open-ended benchmarks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_01694 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Enhancing Video-LLM Reasoning via Agent-of-Thoughts Distillation Shi, Yudi Di, Shangzhe Chen, Qirui Xie, Weidi Computer Vision and Pattern Recognition This paper tackles the problem of video question answering (VideoQA), a task that often requires multi-step reasoning and a profound understanding of spatial-temporal dynamics. While large video-language models perform well on benchmarks, they often lack explainability and spatial-temporal grounding. In this paper, we propose Agent-of-Thoughts Distillation (AoTD), a method that enhances models by incorporating automatically generated Chain-of-Thoughts (CoTs) into the instruction-tuning process. Specifically, we leverage an agent-based system to decompose complex questions into sub-tasks, and address them with specialized vision models, the intermediate results are then treated as reasoning chains. We also introduce a verification mechanism using a large language model (LLM) to ensure the reliability of generated CoTs. Extensive experiments demonstrate that AoTD improves the performance on multiple-choice and open-ended benchmarks. |
| title | Enhancing Video-LLM Reasoning via Agent-of-Thoughts Distillation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2412.01694 |