Intrinsic Distribution Matching
One-step and few-step text-to-image generation must accurately render the details specified in the prompt despite limited opportunities for refinement. Distribution matching distillation (DMD) uses…
deep-learning
MLINFO COLLECTION
深層学習
記事一覧
7 / 45Emergence of the transformer architecture as the score of entropic laws of random measures
We show that the transformer architecture emerges as the score of last-token laws in a natural statistical model of language, where sentences are empirical measures of tokens, a language is a law on…
deep-learning
GC-LIF: Spiking Group-Equivariant Neural Network With Group-Coupled Membrane Dynamics
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…
deep-learning
TactoHand: Learning Geometry-Based Dense Hand-Surface Interaction Fields from Video
Understanding how humans interact with the world requires a representation of physical interactions at the hand surface. Tactile sensing captures these interactions, but collecting tactile data at…
deep-learning
Generative artificial intelligence prompting strategies among Chinese university students across academic tasks
Generative artificial intelligence (GenAI) is increasingly embedded in university learning, yet effective human-GenAI interaction depends not only on access to these tools but also on students'…
deep-learning
Raincheck: Deferred Multimodal Prefill for Efficient Multi-Turn MLLM Inference
Multi-turn conversations with multimodal large language models (MLLMs) reuse input context, but long multimodal inputs increase time to first token (TTFT). Query-aware hard pruning reduces prefill…
deep-learning
Non-Zero Radius: The Missing Component of Neural Networks
Inspired by the analogy between the all-or-none activation of biological neurons and the true-false of logic circuits, McCulloch and Pitts proposed a mathematical model of the biological neuron in…
deep-learning
LatentWeave: Test-Time Latent Optimization for Multimodal Reasoning
Recent advances in test-time optimization have shown that the multimodal reasoning performance of large multimodal language models (MLLMs) can be improved during inference without modifying backbone…
deep-learning
BEACON: Distilled Begin/End-Aware Structured Decoding for Continuous Sign Language Segmentation
In segmentation-based continuous sign language recognition, an untrimmed video is first partitioned into lexical units; segmentation is therefore a means to an end---downstream recognition. However,…
deep-learning
Comprehensive empirical evaluation of feature extractors in computer vision
Feature detection and matching are fundamental components in computer vision, underpinning a broad spectrum of applications. This study offers a comprehensive evaluation of traditional feature…
deep-learning
From Reasoning to Action: Consequence-Grounded Alignment for Vision-Language-Action Models
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…
deep-learning
ROAD: Modern Adversarial Training for Simple One-Step Distillation at Scale
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…
deep-learning
Judging Before Propagation: Task-Graph Contextual Integrity for Multi-Agent Privacy
Multi-agent LLM systems increasingly rely on persistent memory and inter-agent communication. These capabilities create privacy risks that extend beyond individual interactions: a release appropriate…
deep-learning
Conic-Lifting for Learning-rate-free Convex Optimization
A central topic in online and stochastic optimization is to design algorithms that do not require tuning of hyperparameters. Successful strategies in online learning include AdaGrad-style learning…
deep-learning
Last-Iterate Convergence for Constrained MDPs: From Dual Descent to Primal-Dual
We study policy optimization for infinite-horizon, discounted constrained Markov decision processes. While primal-dual ($\texttt{PD}$) methods are widely used in practice, their theoretical…
deep-learning
Distilling Large-Scale Machine Learning Interatomic Potentials to Predict Long-Timescale Electrolyte Transport Properties
Accurately predicting the transport properties of liquid electrolytes is critical for next-generation battery design and requires force fields that are both quantum-mechanically accurate and fast…
deep-learning
Interpretable by Design Tabular Prediction via In-Context Local Linear Learning
In-context learning (ICL) lets tabular foundation models make predictions without dataset-specific training or fine-tuning. These models achieve strong predictive accuracy, but interpreting how…
deep-learning
Does Scaling Toward AGI Depend on What System-1 Lets System-2 Execute?
Progress toward artificial general intelligence (AGI) has largely relied on scaling a resource-intensive executive, System-2, through larger models, more data and compute, longer reasoning, and more…
deep-learning
Does Federated Training Degrade Explainability? A Comparative Analysis of Grad-CAM Faithfulness in Centralized vs. Federated Deepfake Detection
Automated detection has become the essential safeguard against realistic synthetic media, but deploying it at scale requires training on sensitive data from multiple organizations, putting privacy…
deep-learning
Regional Anchor Routing for Multi-Object Motion Transfer
Compositional motion transfer requires a model to extract motions from multiple reference videos, and apply each motion to a user-specified object in one target scene. Existing methods normally…
deep-learning
Topology Survives Shuffling: Robust Long-Context Memory Access Prediction for Graph Analytics
Graph analytics generates irregular, topology-dependent memory-access streams via indirect traversals, limiting data prefetching. Accurate prediction enables timely prefetches that hide memory…
deep-learning
Calibrating Attention-Based Token Pruning for Efficient Multi-Image Synthesis in UMMs
Unified Multimodal Models (UMMs) enable powerful multi-image synthesis but incur substantial computation from long visual contexts. While token pruning offers a direct route to accelerating…
deep-learning
Achieving Minimax-Optimal First-Order and Best-of-Both-Worlds Regret Simultaneously in Multi-Armed Bandits
We study first-order and best-of-both-worlds regret guarantees in multi-armed bandits. Pogodin and Lattimore (2019) proposed an algorithm that combines minimax-optimal worst-case regret with a…
deep-learning
Sparse Fine-Tuning for Parameter-Efficient Adversarial Training
Deep neural networks achieve strong performance across many tasks but remain vulnerable to adversarial perturbations. Adversarial training (AT) is one of the most effective defenses, yet it suffers…
deep-learning