全部收录论文
论文 · 第 130 页
共 7039 篇 · 按日期新到旧 · 每页 50 篇 · 第 130 / 141 页
查找论文请用站内搜索。
- Aligning Language Models from User Interactions
- Asynchronous Verified Semantic Caching for Tiered LLM Architectures
- NanoVDR: Distilling a 2B Vision-Language Retriever into a 70M Text-Only Encoder for Visual Document Retrieval
- Task-Specific Knowledge Distillation via Intermediate Probes
- Test-Time Strategies for More Efficient and Accurate Agentic RAG
- Towards AI Search Paradigm
- LMEB: Long-horizon Memory Embedding Benchmark
- InterDeepResearch: Enabling Human-Agent Collaborative Information Seeking through Interactive Deep Research
- Diagnosing Retrieval Bias Under Multiple In-Context Knowledge Updates in Large Language Models
- ToolTree: Efficient LLM Agent Tool Planning via Dual-Feedback Monte Carlo Tree Search and Bidirectional Pruning
- Structured Distillation for Personalized Agent Memory: 11x Token Reduction with Retrieval Preservation
- Spend Less, Reason Better: Budget-Aware Value Tree Search for LLM Agents
- Efficient and Interpretable Multi-Agent LLM Routing via Ant Colony Optimization
- AI Planning Framework for LLM-Based Web Agents
- Prompt Injection as Role Confusion
- XSkill: Continual Learning from Experience and Skills in Multimodal Agents
- OpenSage: Self-programming Agent Generation Engine
- Building Effective AI Coding Agents for the Terminal: Scaffolding, Harness, Context Engineering, and Lessons Learned
- ARL-Tangram: Unleash the Resource Efficiency in Agentic Reinforcement Learning
- 98× Faster LLM Routing Without a Dedicated GPU: Flash Attention, Prompt Compression, and Near-Streaming for the vLLM Semantic Router
- Serving Hybrid LLM Loads with SLO Guarantees Using CPU-GPU Attention Piggybacking
- Architecture-Aware LLM Inference Optimization on AMD Instinct GPUs: A Comprehensive Benchmark and Deployment Study
- TaxBreak: Unmasking the Hidden Costs of LLM Inference Through Overhead Decomposition
- CMoE: Converting Mixture-of-Experts from Dense to Accelerate LLM Inference
- Cost-Efficient Multimodal LLM Inference via Cross-Tier GPU Heterogeneity
- EWSJF: An Adaptive Scheduler with Hybrid Partitioning for Mixed-Workload LLM Inference
- LLM Query Scheduling with Prefix Reuse and Latency Constraints
- KernelFoundry: Hardware-Aware Evolutionary GPU Kernel Optimization
- Google Cloud AI infrastructure at NVIDIA GTC 2026
- Beyond Interleaving: Causal Attention Reformulations for Generative Recommender Systems
- Breaking User-Centric Agency: A Tri-Party Framework for Agent-Based Recommendation
- UtilityMax Prompting: A Formal Framework for Multi-Objective Large Language Model Optimization
- LASER: An Efficient Target-Aware Segmented Attention Framework for End-to-End Long Sequence Modeling
- CAMMSR: Category-Guided Attentive Mixture of Experts for Multimodal Sequential Recommendation
- FlexRec: Adapting LLM-based Recommenders for Flexible Needs via Reinforcement Learning
- Refine-POI: Reinforcement Fine-Tuned Large Language Models for Next Point-of-Interest Recommendation
- Social Knowledge for Cross-Domain User Preference Modeling
- Sharpness-Aware Minimization for Generalized Embedding Learning in Federated Recommendation
- BTZSC: A Benchmark for Zero-Shot Text Classification Across Cross-Encoders, Embedding Models, Rerankers and LLMs
- LACONIC: Dense-Level Effectiveness for Scalable Sparse Retrieval via a Two-Phase Training Curriculum
- Rank1: Test-Time Compute for Reranking in Information Retrieval
- A Systematic Study of Pseudo-Relevance Feedback with LLMs
- Constraint-Aware Generative Re-ranking for Multi-Objective Optimization in Advertising Feeds
- Differentiable Geometric Indexing for End-to-End Generative Retrieval
- TaSR-RAG: Taxonomy-guided Structured Reasoning for Retrieval-Augmented Generation
- Seq vs Seq: An Open Suite of Paired Encoders and Decoders
- Structured Linked Data as a Memory Layer for Agent-Orchestrated Retrieval
- Does Reasoning Make Search More Fair? Comparing Fairness in Reasoning and Non-Reasoning Rerankers
- Meta-Reinforcement Learning with Self-Reflection for Agentic Search
- Ares: Adaptive Reasoning Effort Selection for Efficient LLM Agents