推論・効率化
A
118 件中 1〜30 件目を表示
-
Build a Validated Kubernetes Runtime for NVIDIA DynamoNVIDIA、Dynamo向け検証済みKubernetes実行環境の構築手順を公開AIアプリを作るチームには、モデル設定から動作する推論エンドポイントまでを再現できる手順が必要になる。NVIDIAは、分散推論基盤「Dynamo」をKubernetes上で動かすための検証済みランタイムの構築方法を解説した。モデルの設定を決め、それを実際に使える推論サービスへ落とし込む流れを標準化し、運用の手間を減らすことを狙う内容である。
-
Run AI Models for Edge Application Development with Just a Few CommandsNVIDIA、数コマンドでエッジ向けAIモデルを実行する開発手法を紹介NVIDIAの開発者ブログが、エッジ向けアプリ開発でAIモデルを少ないコマンドで動かす方法を紹介。AIがデータセンターを越え、運転者の注意や車内環境の監視など、エッジで動くエージェントの開発が広がっている背景を示す。
-
Latent Sparse Attention: Towards Subquadratic Ultra-Long Context Attention
-
A Better Critic Matters for Distribution Matching Distillation
-
ACCT: Asymmetric Cross-LLM Context Transfer
-
Subquadratic Attention with a Hierarchical Tree-Structured Indexer
-
RAMA: Retrieval-Augmented Multi-Agent Routing with Episodic Memory for LEO Satellite Networks
-
The Knowledge Layer Model (KLM): Evidence-Visible AI Output, Routing Diagnostics, and Negative Results
-
There Is No Free Loop: When Does Looping a Trained Model Pay for Its Inference Compute?
-
When Does Test-Time Search Help World Action Models? An Empirical Audit of Candidate Selection and Simulator Integrity
-
Prune Gradually, Adapt Sparsely: Efficient Pruning of Large Language Models at Extreme Sparsity
-
Text as an Index for Zero-Shot Cross-Domain Knowledge Tracing
-
Dynamical Memory Transformer: Selective Recurrent Memories with Constant Cost Attention
-
Trainable Vector Quantization for Large Language Models
-
ElTab: Elastically Looped Tabular Transformers
-
Meta Ray-Banで向かいの席から撮影→指の動きからスマホのパスワードを解読する盗み見攻撃Meta Ray-Banで撮影した指の動きからスマホのパスワードを解読する攻撃を実証テルアビブ大学の研究者らが論文「Watch and Crack」で、スマートグラスで撮影した指の動きからパスワードを推測する手法を実証した。向かいの席からMeta Ray-Banで撮影し、スマホのパスワード入力を盗み見る攻撃が成立しうることを示した研究報告だ。
-
Rethinking Neural Thickets Without Weight Perturbations
-
An ODE-based Sampler for Discrete Diffusion Models
-
Difficulty Calibration Across Time for Membership Inference in Diffusion Models
-
TextSeal2: Distortion-Free LLM Watermarking with Copula Mixing and Memoryless Races
-
Every Token Needs Its Own Teacher: Multi-teacher Adaptive On-policy Self-distillation
-
Comparing Human Role-Players and LLM-Simulated Clients in Online Counselling Training: An Analysis of Counselling Patterns
-
On the Fundamental Limits of Prefill-Based Decoding Length Prediction
-
NEOPIC: A Neural Operator Framework for Particle-based Kinetic Plasma Simulations
-
IndicIPR-QA: A Benchmark for Extractive Question Answering on Indian IPR Laws
-
AI in the EGI e-infrastructure -- Past experiences and future strategies
-
Fast and accurate neural triage decoding of ultra-high-rate quantum LDPC codes
-
Soil Moisture Estimation in Precision Agriculture: A Knowledge Distillation Approach
-
Uncertainty-aware Feedback Loop Diagnosis: integrating reinforcement learning and conformal prediction for reliable wafer defect classification
-
Don't Lose the Thread: Empowering Long-Horizon LLM Agents with Cognitive Resource Self-Allocation