Trains end-to-end driving without human trajectory supervision by decoupling perception and action: DriveVFM distills multiple frozen vision foundation models into a single camera backbone, and DriveRL trains a privileged closed-loop RL teacher whose rollouts supervise a camera-only planner, yielding state-of-the-art closed-loop benchmark results.
Trains LVLMs to improve 3D spatial reasoning using a synthetic block-stacking dataset of 15,000 problems that cover 3D→2D projection, viewpoint transforms, and structural combination; includes color-anchored cues and provides code and data.
AI Agent Papers2026
Lets vision-language models control robots via a compact semantic action interface that maps intent to discrete action units; supports zero-shot use of closed-source VLMs, low-cost fine-tuning of open VLMs, and GUI-based demonstration collection.
AI Video Papers2026
Translates natural-language instructions into executable programs that maintain an explicit, persistent global world state and compiles state-augmented 3D oriented bounding boxes into pixel-aligned conditioning signals for pretrained video generators. The approach decouples state evolution from rendering, enabling programmable entity control, off-screen state, and long-horizon interactive scenarios.
Jiayin Chen, Yicheng Xu +1Miyang Technology (Shanghai) Co., Ltd., Shanghai, China, Key Laboratory of System Software (Chinese Academy of Sciences), Beijing, China +4
Suppresses grid-like and granular “digital ripple” artifacts introduced by iterative reference-conditioned image editing using selective spectral notching, structure-aware smoothing, and cleaned-reference regeneration. Prioritizes low-distortion filtering when artifacts are spectrally separable and visual reconstruction when artifacts overlap content.
An 8B native unified multimodal model that understands, reasons about, and generates high-fidelity visual content without a separate visual encoder or VAE—focused on spatially coherent patch reconstruction, native 4K generation, multi-expert distillation, and improved text rendering and image editing.
AI Video Papers2026
Haoyu Zhao, Zihao Zhao +11Affiliation: National University of Singapore, Affiliation: Fudan University +1
Evaluates whether MiniMax-H3 can infer latent physical events from incomplete multimodal evidence and produce videos consistent with those inferences. Builds a 517-instance benchmark across four reasoning scenarios (multi-view, audio disambiguation, video continuation, audiovisual integration) and quantifies modality-specific failure modes.
Introduces physical rank consistency (PRC) and ActionPiece, a learned action tokenizer that preserves local physical relationships via joint supervision of representation learning and quantization, improving autoregressive vision-language-action policy performance across multiple benchmarks.
Enables specifying exact 24-bit hex colors for object-level image generation and editing via a unified prompt interface, using a Paint-500K dataset and timestep-gated pure-color anchors to improve color fidelity.
AI Video Papers2026
Zhenchen Tang, Yang Li +7New Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, School of Artificial Intelligence, University of Chinese Academy of Sciences +2
Introduces a rubric-based approach for video reward modeling that generates explicit, query-adaptive evaluation criteria before scoring to reduce scalar drift; proposes RGPO, a two-stage training (seed warm-up + joint optimization) that achieves strong pointwise and pairwise evaluation with high data efficiency.
Teaches vision-language models to predict and integrate physical-world state transitions from interaction trajectories (observation → action → next observation). Introduces a three-level curriculum and the LSI-108K dataset, and applies supervised fine-tuning plus on-policy distillation to improve local transition modeling and long-horizon spatial integration.
Hangyu Ran, Yuhao Zheng +3Ant Group, The Hong Kong University of Science and Technology (Guangzhou) +2
Converts natural-language SVG instructions into instance-aware six-item rubrics and uses a vision–language judge to score rendered SVGs item-by-item as dense RL rewards. Produces stronger human alignment and lifts benchmark rubric scores substantially on MMSVG datasets.