LongVPO: From Anchored Cues to Self-Reasoning for Long-Form Video Preference Optimization

TL;DR

LongVPO introduces a two-stage framework that enables short-video models to understand long videos effectively without extensive annotations, using synthetic data and recursive reasoning.

Contribution

It proposes a novel synthetic preference data generation method and a recursive reasoning pipeline for long-video understanding without human labels.

Findings

Outperforms state-of-the-art models on long-video benchmarks.

Maintains strong performance on short-video tasks.

Uses only 16K synthetic examples for training.

Abstract

We present LongVPO, a novel two-stage Direct Preference Optimization framework that enables short-context vision-language models to robustly understand ultra-long videos without any long-video annotations. In Stage 1, we synthesize preference triples by anchoring questions to individual short clips, interleaving them with distractors, and applying visual-similarity and question-specificity filtering to mitigate positional bias and ensure unambiguous supervision. We also approximate the reference model's scoring over long contexts by evaluating only the anchor clip, reducing computational overhead. In Stage 2, we employ a recursive captioning pipeline on long videos to generate scene-level metadata, then use a large language model to craft multi-segment reasoning queries and dispreferred responses, aligning the model's preferences through multi-segment reasoning tasks. With only 16K synthetic examples and no costly human labels, LongVPO outperforms the state-of-the-art open-source models on multiple long-video benchmarks, while maintaining strong short-video performance (e.g., on MVBench), offering a scalable paradigm for efficient long-form video understanding.