MLINFO COLLECTION
製品画像、顕微鏡画像、表面欠陥を対象に、分類・検出・セグメンテーションを選ぶための導線です。
Autonomous mobile robots commonly pair a global planner with a local model predictive controller (MPC), but nonconvex obstacle avoidance and a limited horizon can make the robot freeze in complex…
Efforts to monitor advanced AI for rare misalignments face a data challenge: abundant aligned examples but only a handful of misaligned ones. We test activation probes in this "few vs.
End-of-turn (EoT) detection has been studied mostly through features, models, and modalities, while the speech genre of the training data has drawn far less attention. We isolate this factor.
Agentic reinforcement learning (RL) often requires repeated tool calls in task-specific environments. A training job may create thousands of these environments in bursts while sharing a cluster with…
Background: Contemporary healthcare institutions generate daily very large amounts of data, which contain highly useful information for the improvement of work organization, planning and overall…
Event cameras offer microsecond temporal resolution, high dynamic range and no motion blur, but referring video object segmentation (RVOS) lacks event-stream datasets and models. We introduce…
In this work, we formalize the concept of intrinsic dimension of objective landscapes introduced in Li et al. (2018), which measures the structural complexity of learning problems by training along…
Thinking vision-language models reason for thousands of tokens, yet still miss small details in high-resolution images. Zooming methods pay for extra inference passes, and outcome rewards never tell…
Super-telephoto imaging on mobile phones suffers from complex, device-dependent degradations that are poorly captured by conventional handcrafted pipelines. Collecting large-scale aligned training…
Unified multimodal models (UMMs) jointly support visual understanding and image generation, yet information flow between these two capabilities remains largely asymmetric: understanding…
Current Vision-Language Models (VLMs) frequently fail to recognize culture-specific concepts, especially from underrepresented regions. Effectively injecting this long-tail knowledge into existing…
Spiking neural networks (SNNs) promise event-driven, low-power inference, but on vision tasks they still acquire rotation invariance only through data augmentation. Group-equivariant CNNs build such…
Reasoning informs decision-making by connecting intended actions with their expected consequences. Recent vision-language-action (VLA) models incorporate embodied chains of thought (ECoTs) to guide…
One-step distillation turns pretrained diffusion models into fast generators, but strong results often require substantial training machinery and compute. Pure adversarial distillation offers a…
Intermediate-fusion cooperative perception lets a vehicle detect objects beyond its own field of view by fusing features shared by nearby agents. When only a collaborator's cameras are degraded by…
Giving embodied agents the freedom to act autonomously, initiate interactions, and understand their surroundings through experience is a step toward bringing them to life in the physical world.…
Medical AI learns from the answers doctors record, but rarely from how they interpret the evidence to reach them. Here we introduce PathCowork, a framework that turns pathologists' narrated…
Label-free test-time reinforcement learning adapts vision--language models using rewards derived from their own predictions. However, misinterpreted visual evidence and sensitivity to textual…
Multimodal large language models can recognize visual content yet struggle to reason reliably over it across multiple steps. In counting, a model may see all objects but lose track of which ones it…
Predicting where to grasp and how to move is fundamentally a geometry problem. Vision-language models treat it as language generation instead.
FP8 formats specify how values are stored, not how products are summed; this gap can break checkpoints. Evo 2, a DNA language model, documents a Hopper FP8 requirement for its 1B and 40B checkpoints.
Accurate lung tumor segmentation in computed tomography (CT) is essential for treatment planning, response assessment, and quantitative imaging analysis. However, 3D lung tumor segmentation remains…
Quantifying degradation between a pristine signal and its degraded observation is fundamental to signal processing, yet existing full-reference measures are inherently modality-specific. We introduce…
World Action Models (WAMs) can condition robot actions on generated visual futures, but this additional context is neither cost-free nor uniformly beneficial. We examine what a predicted future…