EvoVid: Temporal-Centric Self-Evolution for Video Large Language Models

TL;DR

EvoVid introduces a temporal-centric self-evolving framework for Video-LLMs, enabling autonomous improvement from raw videos by leveraging temporal rewards, thus reducing reliance on human annotations.

Contribution

The paper presents a novel temporal-centric self-evolution approach for Video-LLMs, incorporating new rewards for temporal reasoning and supervision from unannotated videos.

Findings

Consistent performance improvements across multiple models and benchmarks.

Achieves competitive results with supervised methods.

Demonstrates scalability and effectiveness of temporal-centric self-evolution.

Abstract

Recent Video Large Language Models (Video-LLMs) have demonstrated strong capabilities in video reasoning through reinforcement learning (RL). However, existing RL pipelines rely heavily on human-annotated tasks and solutions, making them costly to scale and fundamentally constrained by human expertise. Self-evolving frameworks have recently emerged as a promising alternative through autonomous Questioner-Solver self-play. Unfortunately, these approaches are primarily designed for static modalities such as text and images, fundamentally failing to capture the temporal dynamics that are central to video reasoning. In this work, we propose $\textbf{EvoVid}$, a temporal-centric self-evolving framework that enables Video-LLMs to improve directly from raw, unannotated videos. Specifically, we introduce two complementary temporal-centric rewards: a temporal-aware Questioner reward that encourages temporally dependent question generation through temporal perturbation sensitivity, and a temporal-grounded Solver reward that provides automatic temporal supervision via inherent video segment localization. Extensive experiments across four base models and six benchmarks demonstrate consistent improvements over both base models and existing self-evolving baselines, achieving competitive performance with supervised methods. These results highlight temporal-centric self-evolution as an effective and scalable paradigm for video understanding and reasoning.