Learning from Next-Frame Prediction: Autoregressive Video Modeling Encodes Effective Representations
Read the original on arXiv Computer Vision →The Flow has not summarised this story yet — read it at arXiv Computer Vision.
The Flow has not summarised this story yet — read it at arXiv Computer Vision.
arXiv:2610.30724v2 Announce Type: replace-cross Abstract: Video provides a natural source of supervision: what happens next serves as a learning signal for visual world modeling. Yet masked video pre...
The paper introduces VANE, a self‑supervised visual pretraining framework that learns from next‑frame prediction in embedding space. It uses a causal encoder to summarize past observations, a state predictor to infer the representation of the entire next frame, and a lightweight generative decoder to reconstruct fine‑grained spatial detail. By applying the same objective to both videos and still images through a local‑to‑global crop sequence, VANE achieves state‑of‑the‑art performance on anticipation, recognition, and dense perception benchmarks, notably outperforming V‑JEPA on EPIC‑KITCHENS‑100 action anticipation.
The paper introduces Future Chunk Prediction (FCP), an auxiliary objective for autoregressive video diffusion models that predicts temporally distant chunks without seeing the intermediate frames. By jointly optimizing FCP with teacher‑forced denoising, the model learns to capture visual evolution rather than relying on local continuity, achieving better physical plausibility while using less training compute. Visualizations indicate that FCP encourages a more structured representation space, potentially benefiting future autoregressive world‑model development.
arXiv:2606. 14765v1 Announce Type: cross Abstract: Self-supervised video representation learning has recently advanced through contrastive learning, masked reconstruction, and predictive representation learning.
arXiv:2610.01942v1 Announce Type: new Abstract: Predicting the future evolution of a scene is a fundamental capability for world modeling. Recent work has shown that operating in the feature space of...
arXiv:2610.33857v1 Announce Type: new Abstract: We introduce TrackTok, an efficient video tokenization framework that produces object-centric tokens, i.e., tokens that consistently encode the same se...