ICML 2026 accepted papers on Video
178 of the 6,341 papers accepted at ICML 2026 (2.8%) are on Video, compared with 1.4% at ICML 2025. They fall into 4 topics.
See ICML 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 · 89 papers
- Bridging Your Imagination with Audio-Video Generation via a Unified Director (1 citation)
- Mitigating Surgical Data Imbalance with Dual-Prediction Video Diffusion Model (1 citation)
- AAD-1: Asymmetric Adversarial Distillation for One-Step Autoregressive Video Generation
- CameraNoise: Enabling Faithful Camera Control in Video Diffusion through Geometry-Flow-Guided Noise Warping
- Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation
- CoF-T2I: Video Models as Pure Visual Reasoners for Text-to-Image Generation
- Compression as Adaptation: Implicit Visual Representation with Diffusion Foundation Models
- Context Forcing: Consistent Autoregressive Video Generation with Long Context
- Deep Forcing: Training-Free Long Video Generation with Deep Sink and Participative Compression
- DiLA: Disentangled Latent Action World Models
- DiscoForcing: A Unified Framework for Real-Time Audio-Driven Character Control with Diffusion Forcing
- DreamID-Omni: Unified Framework for Controllable Human-Centric Audio-Video Generation
and 77 more in the interactive map.
Video-language understanding · 56 papers
- A Very Big Video Reasoning Suite
- AD-MIR: Bridging the Gap from Perception to Persuasion in Advertising Video Understanding via Structured Reasoning
- CAST: Modeling Visual State Transitions for Consistent Video Retrieval
- Chain-of-Glimpse: Search-Guided Progressive Object-Grounded Reasoning for Video Understanding
- Decouple and Cache: KV Cache Construction for Streaming Video Understanding
- DynaVieW: Schema-Guided World Modeling for Understanding Hierarchical Visual Dynamics
- E-VAds: An E-commerce Short Videos Understanding Benchmark for MLLMs
- Explainable Forensics of Manipulated Segments in Untrimmed Long Videos
- FlowNar: Scalable Streaming Narration for Long-Form Videos
- Foresee-to-Ground: From Predictive Temporal Perception to Evidence-Driven Reasoning for Video Temporal Grounding
- FrameOracle: Learning What to See and How Much to See in Videos
- From Content to Knowledge: Lightning Fast Long-Video Understanding with Neural Knowledge Representations
and 44 more in the interactive map.
Video segmentation and tracking · 27 papers
- Attend to Anything: Foundation Model for Unified Human Attention Modeling
- CLEAR: Context-Aware Learning with End-to-End Mask-Free Inference for Adaptive Video Subtitle Removal
- EgoTactile: Learning Grasp Pressure for Everyday Objects from Egocentric Video
- Identifiable Token Correspondence for World Models
- Interpreting Physics in Video World Models
- Learning Flexible Generalization in Video Quality Assessment by Bringing Device and Viewing Condition Distributions
- Privacy-Aware Video Anomaly Detection through Orthogonal Subspace Projection
- Rethinking Temporal Consistency in Video Object-Centric Learning: From Prediction to Correspondence
- Return of Frustratingly Easy Unsupervised Video Domain Adaptation
- TD-VAD: Breaking Visual Dependence in Video Anomaly Detection with Text-Driven Learning
- Temporal-Emerged Prompting for Segment Anything in Multiframe Infrared Small Target Detection
- VSCD: Video-based Scene Change Detection in Unaligned Scenes
and 15 more in the interactive map.
Action recognition · 6 papers
- DreamDojo: A Generalist Robot World Model from Large-Scale Human Videos (1 citation)
- Factored Latent Action World Models
- General Covariant Action Modeling: Constructing Generalized Manifolds via Spatio-Temporal Decoupling
- Learning Latent Action World Models in the Wild
- Olaf-World: Orienting Latent Actions for Video World Modeling
- Multi-view Consistent Latent Action Learning for World Modeling and Control