# AVC-DPO: Aligned Video Captioning via Direct Preference Optimization

arXiv: https://arxiv.org/abs/2507.01492 · HTML: https://mlsanity.com/paper/2507.01492/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in MLLM training and visual instruction tuning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 19 of its 20 references that are in the map.
- First posted on arXiv 2 Jul 2025.

## Placed in

- Multimodal LLMs / vision-language models / [MLLM training and visual instruction tuning](https://mlsanity.com/mllm/mllm-training-and-visual-instruction-tuning/)
- LLM alignment / RLHF / preference optimisation / [Diffusion and multimodal model alignment](https://mlsanity.com/alignment/diffusion-and-multimodal-model-alignment/)

## Builds on (most cited first, 19 of 19)

- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [LLaVA-Video: Video Instruction Tuning With Synthetic Data](https://mlsanity.com/paper/2410.02713/) (2024)
- [Inference-Time Scaling for Generalist Reward Modeling](https://mlsanity.com/paper/2504.02495/) (2025)
- [VideoChat-Flash: Hierarchical Compression for Long-Context Video Modeling](https://mlsanity.com/paper/2501.00574/) (2024)
- [AuroraCap: Efficient, Performant Video Detailed Captioning and a New Benchmark](https://mlsanity.com/paper/2410.03051/) (2024)
- [Aria: An Open Multimodal Native Mixture-of-Experts Model](https://mlsanity.com/paper/2410.05993/) (2024)
- [Describe Anything: Detailed Localized Image and Video Captioning](https://mlsanity.com/paper/2504.16072/) (2025)
- [Chain-of-Frames: Advancing Video Understanding in Multimodal LLMs via Frame-Aware Reasoning](https://mlsanity.com/paper/2506.00318/) (2025)
- [VideoCap-R1: Enhancing MLLMs for Video Captioning via Structured Thinking](https://mlsanity.com/paper/2506.01725/) (2025)
- [VidBridge-R1: Bridging QA and Captioning for RL-based Video Understanding Models with Intermediate Proxy Tasks](https://mlsanity.com/paper/2506.09079/) (2025)
- [Progress-Aware Video Frame Captioning](https://mlsanity.com/paper/2412.02071/) (2024)
- [Caption Anything in Video: Fine-grained Object-centric Captioning via Spatiotemporal Multimodal Prompting](https://mlsanity.com/paper/2504.05541/) (2025)
- [VCapsBench: A Large-scale Fine-grained Benchmark for Video Caption Quality Evaluation](https://mlsanity.com/paper/2505.23484/) (2025)
- [ActionArt: Advancing Multimodal Large Models for Fine-Grained Human-Centric Video Understanding](https://mlsanity.com/paper/2504.18152/) (2025)
- [Personalized Video Summarization by Multimodal Video Understanding](https://mlsanity.com/paper/2411.03531/) (2024)
- [Cockatiel: Ensembling Synthetic and Human Preferenced Training for Detailed Video Caption](https://mlsanity.com/paper/2503.09279/) (2025)

## Built on by (most cited first, 2 of 2)

- [VDC-Agent: When Video Detailed Captioners Evolve Themselves via Agentic Self-Reflection](https://mlsanity.com/paper/2511.19436/) (2025)
- [Claim-Level Rubric Rewards for Video Caption Reinforcement Learning](https://mlsanity.com/paper/2607.05150/) (2026)

## Abstract (arXiv)

Although video multimodal large language models (video MLLMs) have achieved substantial progress in video captioning tasks, it remains challenging to adjust the focal emphasis of video captions according to human preferences. To address this limitation, we propose Aligned Video Captioning via Direct Preference Optimization (AVC-DPO), a post-training framework designed to enhance captioning capabilities in video MLLMs through preference alignment. Our approach designs enhanced prompts that specifically target temporal dynamics and spatial information-two key factors that humans care about when watching a video-thereby incorporating human-centric preferences. AVC-DPO leverages the same foundation model's caption generation responses under varied prompt conditions to conduct preference-aware training and caption alignment. Using this framework, we have achieved exceptional performance in the LOVE@CVPR'25 Workshop Track 1A: Video Detailed Captioning Challenge, achieving first place on the Video Detailed Captioning (VDC) benchmark according to the VDCSCORE evaluation metric.