MLINFO COLLECTION
画像・動画生成、条件付き生成、編集、3D生成に関係する拡散モデルの論文と実装です。
Before acting on an object, a robot needs to anticipate how its hand motion will change the physical world and what the resulting scene will look like. Existing approaches typically entangle…
Masked diffusion models (MDMs) generate sequences by progressively unmasking tokens, but errors introduced during decoding are typically irreversible. Recent work has shown that iterative correction…
Shadows are observable consequences of latent physical causes, arising from the interaction between illumination, object geometry, and scene structure. A model that understands shadows should be able…
Recent advances in text-to-video generation have substantially improved visual fidelity, yet existing models still struggle to understand the physical evolution implicit in user prompts, making it…
Training-free few-step samplers are often described as nearly quality-lossless under conventional FID/CLIP-style evaluation. It remains unclear whether that conclusion transfers to more realistic and…
Designing RNA molecules that bind to target protein structures requires generating both an RNA conformation and its bound complex geometry, without assuming a predefined binding site. Existing…
Time series anomaly detection (TSAD) is a foundational problem in data mining, aiming to identify observations that deviate significantly from nominal temporal patterns. Driven by escalating data…
Motion Diffusion Models (MDMs) can generate diverse and visually appealing human motion; yet, the resulting trajectories frequently violate physical constraints, limiting their downstream use in…
Adding control to video generation often requires training, which demands additional computational resources, data, or architectural changes. We propose a training-free approach that utilizes…
Long-horizon node outages are harder than conventional missing-value imputation because complete sensor trajectories remain unobserved for extended periods, while the usefulness of spatial and…
Existing 3D generative approaches are typically developed for individual imaging modalities, requiring substantial modality-specific data and limiting their ability to exploit shared structure across…
Video world models predict future frames from past frames and actions. They serve as simulators in which agents train, plan or are evaluated, and that humans can control interactively.
Humanoid fall recovery requires selecting a motion that suits both the surrounding scene and the task. The same fallen pose can admit several ways to stand, with different routes, support…
Increasing the attention context length significantly improves streaming video generation quality, while incurring prohibitive memory and computational costs as the context grows. Existing methods…
Retrieval-augmented generation (RAG) frequently integrates knowledge graphs (KGs) for complex multi-hop question answering. However, their performance is increasingly bottlenecked by the knowledge…
Diffusion-based video generation models (VGMs) can produce visually plausible videos, yet their camera ego-motion often drifts over long horizons. Camera ego-motion determines how the viewpoint…
How do diffusion transformers (DiTs) represent image content and learn rules for reusing it to generalize? We develop an aggregator--predictor view of DiTs as input-adaptive local denoisers, in which…
Understanding and modeling light transport in human-centric scenes is a longstanding challenge. While recent progress in diffusion models has boosted research in relighting, precise illumination…
Generative image compression (GIC) delivers strong perceptual quality at low bitrates. However, GIC methods with strong perceptual quality often rely on pretrained diffusion models with high latency,…
Iris recognition systems are vulnerable to presentation attacks (PAs), including textured contact lenses, print replays, and synthetic irises. Existing PAD methods rely on supervised learning with…
Diffusion language models generate by iteratively denoising a masked sequence under bidirectional attention, without the strict left-to-right commitment of autoregressive decoding. Whether this…
Existing paired clean-degraded datasets have advanced learning-based underwater image enhancement, but they remain either limited in scale or difficult to acquire, and still fail to adequately cover…
In this article, we propose ContactFlowFilter, a novel approach for contact estimation that determines the number of active contacts, their locations, and the associated force vectors. We consider a…
In this paper, we develop a unified framework connecting proper scoring rules and Stein's method for the estimation of unnormalized models. We construct a class of proper scoring rules that…