開発者ツール
B
448 件中 61〜90 件目を表示
-
On-Chain Behavioral Signatures for Classifying Centralized Exchange Addresses on Ethereum Using Proof-of-Reserves
-
Learning Nonlocal Interactions in Large-Scale Heterogeneous Metamaterials through Scientific Machine Learning
-
Text as an Index for Zero-Shot Cross-Domain Knowledge Tracing
-
Fuzzy selective state model for forecasting the dynamics of consumer preferences
-
AI evaluators favor job applications written by certain AI models, independent of candidate quality
-
Learning to Interact and Direct: Co-Evolving Video Director Agents
-
How to Check and Configure NVIDIA GPUDirect Storage for Your PlatformNVIDIA、GPUDirect Storageの確認・設定手順を解説NVIDIAは開発者ブログで、GPUDirect Storage(GDS)を自社環境で使えるか確認し、設定する方法を解説した。GDSはGPUとストレージ間のデータ経路を効率化する技術で、オンプレミスとクラウドの双方でGPUのストレージアクセスを高速化するとしている。
-
ReuScope: Auditing Data Reuse and Evaluation Sensitivity in Capsule Vision 2024
-
DIFFNAS: Predictor-Free Neural Architecture Search via Diffusion Model and Zero-Cost Evaluation
-
RECAST: REconstruction of Chain-of-Thought via Auditable Structured Traces
-
Continual Reinforcement Learning Is an Integration Problem
-
Backpropagation-free Deep Learning for Time-Series Classification
-
What Standard Evaluations Hide About Detectors of Hijacked LLM Agents: Rare Attacks, Unseen Tools and Benchmark Transfer
-
Towards Out-of-Distribution Uncertainty Quantification for Machine-Learned Interatomic Potentials via Pre-Training
-
Accurate and Auditable ECG Classification via Visibility Graph Neural Networks
-
Whose Values, Which Contexts, and What Signals? Grounding Cultural Value Alignment with Re3Culture
-
Implicit Assumptions About Tasks in Continual Learning
-
Different Models, Shared Failures: Auditing Frozen-Encoder Pools for Medical Segmentation
-
Clinical-Protocol-Aware Cyber Resilience for Connected Clinical Technologies Under AI-Enabled Adversarial Conditions
-
Task Vectors Can Be Linearly Decomposed into Interpretable and Controllable Components
-
Simple Features Beat Most AI Control Monitors
-
Advancing Clinical Trial Outcome Prediction with Domain-Specific Large Language Models
-
From synthetic enterprise data evaluations to real Microsoft Copilot AI confidenceMS Research、実データ不使用の合成企業データでCopilotを評価Microsoft Researchは、Copilotを企業規模で評価するには組織の複雑さを反映したテストデータが要る一方、実際の企業データは機密情報を含み評価に使えないと説明する。そこで合成した企業データによる評価の取り組みを紹介している。
-
TSBPO: Tree-Structured Branching for Policy Optimization in Interactive Language Agents
-
When the Past Shapes Action: Understanding Embodied Memory
-
OT-CAL: Optimal-Transport-Guided Adaptive Sample Selection for Efficient Visual Learning
-
GRIT: Gradient Masking for Visually Misgrounded Tokens in Text-Rich Reasoning
-
Saidzoda-Tajik-LFM-2.6B: Adapting a 2.6B Hybrid Model to Tajik Tool Calling, and What Continued Pretraining Takes Away
-
Backdoors in official Claude Code extensionsClaude Code拡張に不正コードか、研究者がRAT混入を指摘セキュリティ研究者Kevin Beaumont氏が、AnthropicのVisual Studio向けClaude Code拡張にマルウェアが含まれ、デスクトップにリモートアクセス型トロイの木馬を作ると投稿した。詳細は外部記事への誘導のみで、現時点では投稿者の主張であり検証はされていない。
-
SCRBench: Beyond Equivalence Checking for Semantic Change Reasoning in Code