全部收录论文
论文 · 第 73 页
共 7039 篇 · 按日期新到旧 · 每页 50 篇 · 第 73 / 141 页
查找论文请用站内搜索。
- Infinite Worlds with Versatile Interactions
- Final Checkpoints Are Not Enough: Analyzing Latent Reasoning Faithfulness Along Training Trajectories
- DSpark: Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation
- GIFT: Geometry-Informed Low-precision Gradient Communication for LLM Pretraining
- SCAPE: Accurate and Efficient LLM Training with Extreme Sparse Communication
- HyperParallel-Mpipe: A Composable Algebra System for Optimizing MLLM Training over Supernode Clusters
- Trees from Marginals: Autoregressive drafting with factorized priors
- Memory-Orchestrated Semantic System (MOSS): An Auditable Agentic Memory Architecture
- BaseRT: Best-in-Class LLM Inference on Apple Silicon via Native Metal
- DeLS-Spec: Decoupled Long-Short Contexts for Parallel Speculative Drafting
- PALS: Percentile-Aware Layerwise Sparsity for LLM Pruning
- A Practical Guide to GPU-Initiated Communication for Molecular Dynamics at Scale
- LLMs Meet Isolation Kernel: Lightweight, Learning-free Binary Embeddings for Fast Retrieval
- Prompting Beats Fine-Tuning: Generative Expected Value Scoring for Statutory Term Retrieval
- Modality Relevance is not Modality Utility: Post-hoc Selective Modality Escalation for Cost-Aware Multimodal RAG
- Inject or Navigate? Token-Efficient Retrieval for LLM Analysis of Transactional Legal Documents
- Nemotron-Labs-Diffusion: A Tri-Mode Language Model Unifying Autoregressive, Diffusion, and Self-Speculation Decoding
- SpanUQ: Span-Level Uncertainty Quantification for Large Language Model Generation
- MemDefrag: Latent Memory Defragmentation for Large Language Models
- Vision as Unified Multimodal Generation
- LLMs Drive Recommendation Systems Advancements
- Create a LangChain Deep Agents Harness Profile for NVIDIA Nemotron 3 Ultra to Improve Performance
- SCOReD: Student-Aware CoT Optimization for Recommendation Distillation
- InfluMatch: Frontier-Quality KOL Search at 4B-Model Cost
- When Should LLMs Search? Counterfactual Supervision for Search Routing
- PORTS: Preference-Optimized Retrievers for Tool Selection with Large Language Models
- 推荐系统走向新范式快手技术沙龙聚焦生成式推荐技术落地
- Learn to Pool: Lightweight Fine-Tuning for Flexible Multi-Vector Compression
- Retrieving a Set, Not Independent Passages: Set-Level Compatibility Learning for Efficient Set Exploration
- Quantifying and Expanding the Theoretical Capacity of Late-Interaction Retrieval Models
- DynaKRAG: A Unified Framework for Learnable Evidence Control in Multi-Hop Retrieval-Augmented Generation
- Faithful or Findable? Evaluating LLM-Generated Metadata for RDF Dataset Search
- Revisiting Text Ranking in Deep Research
- Scientific Code Search at Scale: A Multi-Domain Dataset and Benchmark
- OrchANN: Hierarchical Orchestration for Skewed Out-of-Core Vector Search
- CMDR: Contextual Multimodal Document Retrieval
- Hybrid RAG: Dense and Sparse Retrieval for Better AI Answers
- Query rewriting for RAG: how to improve retrieval accuracy
- Uncertainty-Aware Cross-Modal Remote Sensing Image-Text Retrieval via Evidential Learning
- Linking Hadith Narrator Identities Across Heterogeneous Arabic Biographical Databases: A Multi-Signal Entity Resolution Pipeline
- Text Distance from Nested and Hierarchical Repetitions: A Compression-Based Perspective
- BaFCo: A Document Understanding Benchmark for Complex Bangla Form Comprehension
- SearchEyes: Towards Frontier Multimodal Deep Search Intelligence via Search World Simulation
- CurateEvo: Data-Curation Evolving for Agentic Post-Training
- TurnOPD: Making On-Policy Distillation Turn-Aware for Efficient Long-Horizon Agent Training
- Memory in the Loop: In-Process Retrieval as ExtendedWorking Memory for Language Agents
- SWE-Review: Closing the Loop on Issue Resolution with Agentic Code Review
- From Passive Retrieval to Active Memory Navigation: Learning to Use Memory as a Structured Action Space
- Narrative World Model: Narratology-Grounded Writer Memory for Long-Form Fiction
- Flint: A visualization language for the AI era