全部收录论文
论文 · 第 77 页
共 7039 篇 · 按日期新到旧 · 每页 50 篇 · 第 77 / 141 页
查找论文请用站内搜索。
- Geometry-Aware Online Scheduling for LLM Serving: From Theoretical Bound to System Practice
- SwiftCache: Efficient LLM Serving for Multi-turn Conversations with Heterogeneous KV Cache Sharing
- Harnessing Routing Foresight for Micro-step-level MoE load balancing in RL Post-training
- Alignment Collapse Under KV Cache Quantization: Diagnosis and Mitigation
- DriftSched: Adaptive QoS-Aware Scheduling under Runtime Token Drift for Multi-Tenant GPU Inference
- ProWAFT: A ROMA-LPD Instance for Workload-Aware and Dynamic Fault Tolerance in FPGA-Based CNN Accelerators
- Generative Retrieval via Diffusion Transformer with Metric-Ordered Sequence Training and Hybrid-Policy Preference Optimization
- AGRec: Adapting Autoregressive Decoders with Graph Reasoning for LLM-based Sequential Recommendation
- UniFormer: Efficient and Unified Model-Centric Scaling for Industrial Recommendation
- ScaleToT: Generalizing Structured LLM Reasoning for Billion-Scale Low-Activity User Modeling
- SocialPersona: Benchmarking Personalized Profiling and Response with Multimodal Social-Media Context
- From Clicks to Intent: Cross-Platform Session Embeddings with LLM-Distilled Taxonomy for Financial Services Recommendations
- TRUST: Item-Calibrated Interval Evidence for Temporal Session-Based Recommendation
- AgentX: Towards Agent-Driven Self-Iteration of Industrial Recommender Systems
- NOVA: A Verification-Aware Agent Harness for Architecture Evolution in Industrial Recommender Systems
- Attributed, But Not Incremental: Cannibalization-Corrected Attribution for Large-Scale Advertising
- EMA-FS: Accelerating GBDT Training via Gain-Informed Feature Screening
- Scoring Is Not Enough: Addressing Gaps in Utility-fairness Trade-offs for Ranking
- Temporal Validity in Retrieval Memory: Eliminating Stale-Fact Errors for AI Agents over Evolving Knowledge
- KARLA: Knowledge-base Augmented Retrieval for Language Models
- Why Multi-Step Tool-Use Reinforcement Learning Collapses and How Supervisory Signals Fix It
- Beyond Function Calling: Benchmarking Tool-Using Agents under Tool-Environment Unreliability
- Semantic Consistency Policy Optimization for Reinforcement Learning of LLM Agents
- To Isolate or to Score? Model-Adaptive Assessment for Cost-Efficient Multi-Agent RAG
- OPID: On-Policy Skill Distillation for Agentic Reinforcement Learning
- Semantic Early-Stopping for Iterative LLM Agent Loops
- Knowledge-augmented Agentic AI for Mental Health Medication Information Seeking
- Localizing RL-Induced Tool Use to a Single Crosscoder Feature
- Diagnosing and Mitigating Compounding Failures in Agentic Persuasion via Taxonomic Strategy Retrieval
- GPUSparse: GPU-Accelerated Learned Sparse Retrieval with Parallel Inverted Indices
- TileMaxSim: IO-Aware GPU MaxSim Scoring with Dimension Tiling and Fused Product Quantization
- CAT-Q: Cost-efficient and Accurate Ternary Quantization for LLMs
- Information-Aware KV Cache Compression for Long Reasoning
- Epiphany-Aware KV Cache Eviction Without the Attention Matrix
- PersistentKV: Page-Aware Decode Scheduling for Long-Context LLM Serving on Commodity GPUs
- Creating the NVIDIA Nemotron 3 Ultra NVFP4 Checkpoint with NVIDIA Model Optimizer
- Moebius: Serving Mixture-of-Expert Models with Seamless Runtime Parallelism Switch
- Accelerating Gemini Nano models on Pixel with frozen Multi-Token Prediction
- Previewing GPT-5.6 Sol: a next-generation model
- Recommendation as Generation: Unifying Personalized Video Generation and Recommendation at Industrial Scale
- Evaluating Scene-based In-Situ Item Labeling for Immersive Conversational Recommendation
- TokenMinds: Pretrained User Tokens and Embeddings for User Understanding in Large Recommender Systems
- S2-CAR: Segmentation-Supervised Complexity-Adaptive Recommendation
- Causal-Invariant Cross-Domain Out-of-Distribution Recommendation
- AutoRelAnnotator: Calibrated Model Cascades for Cost-Efficient Relevance Evaluation in Sponsored Search
- Adaptive Re-Ranking
- Extreme Meta-Classification for Large-Scale Zero-Shot Retrieval
- Is GraphRAG Needed? From Basic RAG to Graph-/Agentic Solutions with Context Optimization
- Tracing Target Answers in Poisoned Retrieval Corpora via Token Influence Attribution
- Hybrid-IR: Dual-Path Hybrid Retrieval with Iterative Reasoning for Complex Medical Question Answering