Tag
Introduces DiffARFNO, a two-stage framework combining autoregressive Fourier-MIONet with a conditional DDIM corrector for long-horizon droplet evolution prediction in inkjet printing, achieving state-of-the-art performance on ANSYS Fluent datasets.
HarmoHOI is a unified diffusion framework that jointly generates synchronized multi-view hand-object interaction videos and globally aligned 3D point tracks, achieving state-of-the-art performance in visual quality, motion plausibility, and multi-view geometric consistency.
This paper introduces CoDiffGRN, a co-evolutionary discrete diffusion framework for gene regulatory network inference, along with a new benchmark BEELINE-KGC for inductive evaluation. It achieves state-of-the-art performance in novel regulatory discovery.
WanSong is a pure diffusion-based music generation model that directly produces high-fidelity, multilingual songs up to 5 minutes long with dual stems (vocals and background music) in a single run, addressing challenges in efficient generation, long-form audio, and controllability.
HDR is a unified framework integrating hierarchical latents into causal video generation for multi-step visual reasoning, achieving better reasoning consistency, lower latency, and strong data efficiency compared to baselines.
This paper introduces a novel diffusion-based generative model for structure-based drug design that decouples pocket and ligand representation learning and incorporates multi-scale interaction signals and property-aware optimization to generate developable 3D molecules with improved binding affinity and ADMET properties.
Introduces Generalized Poisson Flow (GPFlow), a variable-length generative framework for protein design that learns an inhomogeneous generalized Poisson process, enabling flexible length exploration and improving designability across structure, sequence, and peptide co-design tasks.
Introduces MobiDiff, an end-to-end discrete diffusion framework for generating human mobility data by denoising multi-channel semantic skeletons, achieving faster inference and competitive fidelity on real-world datasets.
This paper presents CineMobile, a method for efficient on-device image-to-video generation that achieves a 40x speedup over the teacher model through distillation-guided pruning, diffusion distillation, and hybrid quantization, enabling cinematic camera motion effects on mobile devices.
Explores using Google's Gemma diffusion model as a speculative model for efficient large language model inference.
AMD releases Micro-World, an action-controlled interactive world model built on the Wan2.1 family, with open-source weights, code, and a curated dataset for controllable world modeling.
A test of four new frontier AI models (MiMo-V2.5-Pro, MiniMax M3, Mercury 2, LongCat-2.0) using riddles that require genuine reasoning rather than pattern-matching reveals that while most models perform reasonably, LongCat-2.0 repeatedly generates fabricated information with false confidence.
CONFLUX is a 3D latent diffusion model for chest CT synthesis that achieves high-fidelity volumetric generation with controllable clinical attributes, enhanced by a reinforcement learning post-training stage to improve conditioning reliability. The model and a synthetic dataset of ~200k chest CT volumes are released.
This paper introduces CoIn, a novel framework for 3D scene inpainting that bridges 2D diffusion models and 3D Gaussian Splatting via a multi-stage consistency pipeline, enabling both object removal and insertion with flexible masks.
RS-Diffuser 提出了一种风险敏感的离线扩散规划框架，结合扩散轨迹生成和分布值批评器，通过尾部感知目标在推理时灵活调整风险偏好，在安全关键任务中提升了回报和鲁棒性。
This paper introduces A2World, a diffusion-based world model pretrained on large-scale robot manipulation data to learn transferable dynamics priors. The model can be adapted into a real-world simulator (A2World-sim) for policy evaluation or a video-action prediction model (A2World-policy) for action prediction, demonstrating benefits for both simulator-centric and policy-centric robot learning.
PhysisForcing is a training framework that enhances embodied video generation for robotic manipulation by enforcing physical consistency through pixel-level trajectory alignment and semantic-level relational alignment losses in a DiT-based architecture, achieving notable improvements on benchmarks.
This paper proposes a CKM-driven framework for multi-UAV trajectory planning in urban inspection, using diffusion models to reconstruct high-fidelity channel quality maps and a graph attention network with soft actor-critic algorithm for communication-aware path planning.
This technical report presents Qwen-Image-2.0-RL, a post-training pipeline using reinforcement learning from human feedback and on-policy distillation to enhance visual quality and instruction-following in image generation and editing tasks.
This paper introduces Prob-BBDM, a probabilistic Brownian Bridge Diffusion Model for efficient and high-quality MRI sequence synthesis from 2D axial slices, achieving up to 88.46% SSIM and 26.09 dB PSNR with only 4 diffusion steps, and demonstrating clinical utility in tumor segmentation.