5arXiv cs.AI (Artificial Intelligence)·
GraphVid: Graph-conditioned image-to-video generation with structured interaction control
GraphVid is a new image-to-video generation model that uses structured interaction graphs as conditioning signals, enabling precise multi-object control without requiring users to manually draw per-object trajectories. The authors also release GraphVid-Bench, a large-scale video dataset with relational annotations for training interaction-aware models. Despite using less training data and fewer parameters than prior methods, GraphVid outperforms Motion-I2V substantially, reducing FID by 39.9% and FVD by 37.6% while improving PSNR and SSIM metrics.
Related guides (2)
Related events (8)
Veo 3.1 Ingredients to Video: More consistency, creativity and control
Google DeepMind has released Veo 3.1, an updated video generation model that improves consistency, creativity, and control in generated clips. The update produces more natural and dynamic video content and adds support for vertical video generation. The announcement comes from DeepMind's official blog as a tier-1 source.
IV-CoT: Implicit Visual Chain-of-Thought for Structure-Aware Text-to-Image Generation
Researchers propose Implicit Visual Chain-of-Thought (IV-CoT), a latent visual reasoning framework that decomposes visual conditioning queries into a structural-to-semantic cascade for text-to-image generation. The method uses training-only sketch supervision to guide structural queries without requiring sketch extraction at inference time, enabling implicit CoT reasoning in a single forward pass. IV-CoT achieves improved results on GenEval and T2I-CompBench benchmarks, targeting persistent weaknesses in multimodal LLMs around object counts, spatial relations, and attribute binding.
Build Awesome Datasets for Video Generation
Hugging Face published a blog post on constructing high-quality datasets for video generation models. The post likely covers data collection, preprocessing, and curation pipelines relevant to training video diffusion or generation systems. This is a practical tooling and methodology guide aimed at practitioners working on video AI.
GaP: Graph-as-Policy multi-agent harness for reliable robot automation via self-learning computation graphs
Researchers introduce Graph-as-Policy (GaP), a multi-agent coding harness that generates directed computation graphs combining perception, planning, and control nodes for robotic 'Variational Automation' tasks — those with high variation in object geometry and pose. GaP uses an internal simulation environment to rehearse and iteratively refine graph structures in parallel, improving success rates without relying solely on model-free policies. Evaluation across 8 new benchmarks (4 simulated, 4 real-world) shows significant outperformance over baselines. The work bridges agentic coding systems with interpretable robot programming and Task and Motion Planning.
TunerDiT: Training-free Progressive Steering of Diffusion Transformers for Multi-Event Video Generation
TunerDiT is a training-free method for steering video diffusion transformers (DiTs) to generate long-horizon videos containing multiple sequential events. The approach identifies intrinsic turning points in the DiT denoising trajectory where text conditioning shifts from global layout to fine-grained detail, then applies two steering mechanisms: Event-Partitioned Masking and Cross-Event Prompt Fusion. The authors also introduce Meve, a benchmark prompt suite for multi-event video generation, and report state-of-the-art results across 8 metrics with improved text alignment scaling with event count.
IVGT: Implicit Visual Geometry Transformer for Neural Scene Representation
IVGT is a new neural architecture that implicitly models continuous 3D geometry from unposed multi-view images without requiring explicit pointmap regression. It learns a continuous neural scene representation in a canonical coordinate system, supporting SDF-based surface queries and color prediction via lightweight decoders. The model is trained with multi-dataset joint optimization using 2D supervision and 3D geometric regularization, achieving strong generalization across mesh reconstruction, novel view synthesis, depth/normal estimation, and camera pose estimation tasks.
PEVA: Whole-Body Conditioned Egocentric Video Prediction for Embodied World Models
Researchers from BAIR introduce PEVA (Predicting Ego-centric Video from human Actions), a model that generates first-person video frames conditioned on 48-dimensional whole-body kinematic pose trajectories. The model uses an autoregressive conditional diffusion transformer trained on the Nymeria dataset, which pairs real-world egocentric video with body pose capture. PEVA can generate atomic action videos, simulate counterfactuals, and support long video generation, representing a step toward world models grounded in physically embodied human agents.
OpenCoF: Framework for Chain-of-Frame reasoning via video generation
Researchers introduce OpenCoF, a framework that treats video generation as a reasoning substrate — where logical inference unfolds across temporally connected frames (Chain-of-Frame, CoF) rather than text tokens. The work includes OpenCoF-17K, a 17K-sample reasoning video dataset spanning 11 task families, and Wan-CoF, a fine-tuned video model built on Wan2.2-I2V-A14B that achieves gains on four video reasoning benchmarks. The paper also explores augmenting the model with visual and textual reasoning tokens to capture low-level visual cues and high-level semantic priors. All artifacts are open-sourced.