全部收录论文
论文 · 第 5 页
共 7039 篇 · 按日期新到旧 · 每页 50 篇 · 第 5 / 141 页
查找论文请用站内搜索。
- What Matters for LLM Persona Simulation: Signal Form in Cold-Start E-Commerce Advertising
- PAGSL: Path-Augmenting Graph Structure Learning for explainable scholar recommendation
- Learning from How Users Interpret: Personalizing Recommendation Explanation Order
- SPARSEUP: Towards Better Recipes for Training Modern Learned Sparse Retrievers
- FondueQA: A Benchmark for Evaluating Parametric Compensation in Multi-Hop QA
- MARS: Training Multilingual Search Agents for Retriever-Free Evidence Acquisition
- EAGR: Evidence-Guided Adaptive Graph Retrieval
- Orthogonal Norm Growth in Two-Tower Retrieval: A Popularity Mechanism for Unified Discovery
- Whispers from the Neighbors: Rethinking Caption Consensus as Inference-Time Evidence for Vision-Language Retrieval
- Differentially Private GraphRAG: Protecting Relation Connectivity and Semantics
- Less Is More: Graph-free Multimodal RAG via Multi-signal Late Fusion
- Improving Contextual Retrieval for Long Documents using Semantic Chunking and Hybrid Retrieval Strategies
- SkinMind: Equipping Agentic RAG with Human-Inspired Graph Probing for Reliable Dermatology AI
- Structure and Lineage Evidence for Vietnamese Statutory Article Retrieval on DRiLL
- A Taxonomy-Driven Synthetic Dataset for Retrieval Effectiveness and Fairness Evaluation in Hiring
- Inside the Trap: A Renewal Law for Prefix-Cache Collapse in LLM Agent Serving
- e-means: outperforming k-means (in one-tenth the epochs)
- ResServe: Accurate and Efficient FP4 LLM Inference with Residue Compensation
- FOSCAR: Accelerating Decode Attention over 2- and 4-Bit KV Caches through Affine Folding
- A Controlled Experimental Evaluation of Configuration and Cache Architecture in LLM Serving Engines
- Accelerating Offloading-Based MoE Inference by Interleaving Concurrent Requests
- ResQAdam: Residual-Preserving Adaptive Optimization for Memory-Efficient Quantization Aware Training
- DARIC: A Position-Independent KV Cache Reuse System with Distribution-Aware Recomputation
- Extending LLDB to Debug Applications That Span CPUs and GPUs
- CodeQuant2: Enhanced Outlier Mitigation and Adaptive Clustering for Low-Precision LRMs
- SpecPilot: Autonomous Algorithm Discovery and Systems Optimization for Speculative Decoding
- PagedMoE: Cross-Precision Expert Caching for Token-Level Dynamic Precision in SSD-Resident MoE Inference
- Supermicro Launches KV Cache Solution with IBM and NVIDIA Demonstrating Significant Inference Performance Benefits
- Should You Split Decode? Disaggregation Across Future Models and Accelerators
- Meituan-Robotics-0: A Vision-Language-Action Foundation Model for Desktop Manipulation
- On Agentic Manipulation Policies
- VICTR: Value-Informed In-Context Learning for Vision-Language-Action Models
- SURE:Spatiotemporal Uncertainty Representation for Efficient Vision-Language-Action Learning
- ForPC: Force-Aware Whole-Body Control for Precise Humanoid Cart Pushing
- Horizon-Coupled Obstacle Guidance for Relative-Action Diffusion Policies
- STEP: Spatiotemporal Training with Explicit Prediction for Perceptive Humanoid Locomotion
- ODEAction: From Predicting Chunks to Evolving Actions
- DKWBC: Decoupling Motion Tracking and Balancing for Humanoid Whole-Body Control
- Spatial-UMI-Dream: Scaling World Action Models with Metric Geometry for UMI Manipulation
- Should the Action Expert Train the VLM? Action-Token Supervision and Gradient Routing in Mixture-of-Transformer VLAs
- From Pixels to Actions -- Towards Native Vision-Language-Action Policies
- WorldContact: A Contact-Centric World Model for Scalable Robot Learning
- SLA: Scaling Spatial-Language-Action Model via Continual Pre-training
- Boston Dynamics adds wheels to Atlas humanoid to test whole-body control
- Make a Robot Arm Feel Weightless with NVIDIA Holoscan 5
- AI WAVE收官丨便利店24小时直播战报出炉,智元推动Physical AI走向真实场景
- RecGroupMemory: Feedback-Aware Cohort Memory for E-Commerce Recommendation
- Preprocessing Choices in Fine-Tuning Data Matter: Utility and Demographic Sensitivity in LLM-Based Recommendation
- Generative Chain of Behavior for User Trajectory Prediction
- SIDRec: End-to-End Semantic ID Representation and Modeling for Recommendation