NeurIPS 2026 accepted papers on Video
280 of the 6,080 papers accepted at NeurIPS 2026 (4.6%) are on Video, compared with 4.0% at NeurIPS 2025. They fall into 4 topics.
See NeurIPS 2026 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Video generation · 141 papers
- $h$-control: Training-Free Camera Control via Block-Conditional Gibbs Refinement
- 3DSPA: A 3D Semantic Point Autoencoder for Evaluating Video Realism
- Accelerating Video Inverse Problem Solvers with Autoregressive Diffusion Models
- Activation Steering of Video Generation Models via Reduced-Order Linear Optimal Control
- AdaState: Self-Evolving Anchors for Streaming Video Generation
- Advancing Narrative Long Video Generation via Training-Free Identity-Aware Memory
- ARGUS: Stacked Multi-View Identity Mosaic Injection for Subject-Preserving Video Generation
- Attend Locally, Remember Linearly: Linear Attention as Cross-Frame Memory for Autoregressive Video Diffusion
- Audible World Models: Spatially Aware Sound Generation for 3D Worlds
- Auteur: Language-Driven Cinematographic Framing for Human-Centric Video Generation
- AVENUE: Audio-Video Editing Understanding and Evaluation
- Bidirectional Sparse Attention for Faster Video Diffusion Training
and 129 more in the interactive map.
Video-language understanding · 86 papers
- A Benchmark for Omni-Modal Reasoning in Long Videos
- AdaCodec: A Predictive Visual Code for Video MLLMs
- Addressable Memory for Video World Models
- Animation2Code: Evaluating Temporal Visual Reasoning in Video-to-Code Generation
- AVOC: Enhancing Hour-Level Audio-Video Understanding in Omni-Modal LLMs via Retrieval-Inspired Token Compression
- Before It Fades: Reinforcing Temporal Representations at Inference Time in VideoLLMs
- Behavioral Geometric Supervision Aligns Video Foundation Models with Human Social Perception
- BusterX: MLLM-Powered AI-Generated Video Forgery Detection and Explanation
- CollabVR: Collaborative Video Reasoning with Vision-Language and Video Generation Models
- Controllable Multi-label Video Safety Detection via Adaptive Tversky Policy Optimization
- CoRDS: Coreset-Based Representative and Diverse Selection for Streaming Video Understanding
- DELTAVID: Enhancing Fine-Grained Spatiotemporal Perception with Cross-Video Differences
and 74 more in the interactive map.
Video segmentation and tracking · 40 papers
- Can 4D Foundation Models Remember?
- CoANeRV: Coordinate-Aware Token-Space Neural Video Representation
- DeformMaster: An Interactive Physics-Neural World Model for Deformable Objects from Videos
- EgoTac: In-the-wild Tactile Prediction from Egocentric Vision
- Exposing and Mitigating Temporal Attack in Deepfake Video Detection
- FacePhys: State of the Heart Learning
- GazeFlow: From Human Gaze Behavior to Generative Egocentric Gaze Prediction
- GMOS: Grounding Moving Object Segmentation in 3D Space and Time
- LiBrA-Net: Lie-Algebraic Bilateral Affine Fields for Real-Time 4K Video Dehazing
- NEvo: Neural-Guided Evolutionary Video Synthesis for Dynamic Visual Selectivity
- Principia: Relational Physics Tests for Video Models
- Rethinking the Readout: Unlocking Video Backbones for AI-Generated Video Detection
and 28 more in the interactive map.
Action recognition · 13 papers
- Action Images: End-to-End Policy Learning via Multiview Video Generation
- LookWhen? Fast Video Recognition by Learning When, Where, and What to Compute
- TrAction: Action Recognition with Sparse Trajectories
- World Models as Group Actions
- Adaptive Residual Quantization for Memory-Efficient Temporal Action Segmentation
- Beyond Score: A Dataset for Joint Action and Score Predictions in 2v2 Sports
- EpiStream: Utility-Aware Temporal Abstraction for Dense-Action Streams
- Generalizing Action-Conditioned Latent World Models with Video Model Rewards
- HyrCap: Hybrid Rank-Calibration of Action Proposals for Temporal Event Understanding
- Interaction-Aligned Robot Learning from Human Videos with Structured Graph Modeling
- MotorSense: A Video-EMG Dataset of Motoric Representations for Action Understanding
- Not All Layers Are Equal in Image-to-Video Transfer
and 1 more in the interactive map.