全部收录论文
论文 · 第 27 页
共 7039 篇 · 按日期新到旧 · 每页 50 篇 · 第 27 / 141 页
查找论文请用站内搜索。
- LexiHorizon: Stabilizing Reinforcement Learning for Long-Horizon Deep Search
- Dynamic Harness Search: Building Multi-Agent Systems Per-Query via Prediction
- WHALE: A Simple Recipe for Joint Harness-Weight Optimization
- Control How Your GPU Shares Work with Green Contexts
- StagQ: Constraint-Driven Multi-Precision Weight Quantization for LLMs
- Tailoring the Quantization Space for 1-Bit KV Cache Compression
- Crossflow: Prefill-Decode Elasticity for Agentic LLM Serving
- Closing the Context Gap: Activation Alignment for Tabular In-Context Learning
- How DOCA GPUNetIO Unifies GPU-Initiated Networking Across the NVIDIA Software Stack
- Differentiable Bit-Widths: Co-optimizing Pruning and Quantization via SVD for Ultra-Efficient LLM Compression
- Capability Scaling-Down Laws for LLM Compression
- AICR v1.0: Open, stable, and verifiable GPU cluster configuration
- MAPS: Memory-Aware Predictive Scheduling Framework for Large Language Model Serving
- When History Misleads: Asymmetric Margin Supervision for Instruction-Guided LLM Generative Recommendation
- Reasoning with Evidence, Not Merely Rationales: Verifiable Preference Proofs for LLM-Based Recommendation
- Beyond One Epoch: Uncertainty-Weighted Sensitivity Regularization for Recommendation Models
- MuSeR: Scalable Long-sequence Recommendation with Multi-interest Modeling
- Adaptive Sparsity Optimization with Learnable Soft Top-K and Per-Term Thresholding for Efficient Retrieval
- Learning Query Encoders Can Be Hard Even When Vector Retrieval Is Geometrically Easy
- Benchmarking Literature Retrieval for a Model Organism: A Dictyostelium Case Study
- Query-aware routing for Cross-lingual performance gains in Encoders
- Page-Aware Retrieval-Augmented Generation for EvalLLM 2026: A Five-Variant Study on French PDFs
- The Memory Server in the Lab: 21 Studies, 125 Test Questions, +20.8 Percentage Points
- APDMem: Agent-Controlled Progressive Disclosure for Query-Adaptive Long-Term Memory
- CUEing User Simulators: Calibrated User Embeddings for Multi-Turn Benchmarking
- Beyond Correctness: Resolving Underspecification in Agentic Text-to-SQL
- When History Fails to Become Experience: Action Calibration in Language Agents
- LongCat-DeepResearch Technical Report
- Mr.LHDR: A Benchmark for Multimodal Real-World Long-Horizon Deep Research Agents
- Counterexample Generation via Per-Theorem Symbolic Verifiers: When Imitation Hurts and Reinforcement Repairs
- Asterism: Exploring and Synthesizing Scattered Observations into Literature-Grounded Hypotheses and Theories
- Finding the Move Is Not Winning the Game: XiangqiBench for Closed-Loop Evaluation of LLM Agents
- Open and Emergent Problems in Agentic Privacy and Security: A Contextual Angle
- SOLO: Certified-Recall Metric Similarity Search with Scan-Only Sampled Inverted Lists
- MRVQ: One Resident Index for Dimension- and Rate-Elastic Vector Search
- WakeKV: Reactive, Reversible KV Residency for Heads That Change Their Minds
- How Full-Stack NIM Optimizations Deliver 2.5x More Users on Nemotron 3 Ultra
- vLLM x Novita AI: Chord, Faster INT4 MoE for Kimi K2.x. Up to 1.3x on H200, 2.15x on Untuned B300
- Text-Centric Post-Training for Omni-Modal Reasoning
- Large Language Continuous Diffusion Models
- How we trained the fastest DSpark for Kimi-K3 using GB300 NVL72
- Watermarking in vLLM
- On-Premises Multi-Course RAG Tutoring for Business Education: Hardware-Software Trade-offs in a Campus AI Tutor
- OneTrans-V2: Unifying Retrieval, Pre-rank, and Fine-rank with One Transformer in Industrial Recommender
- SPRINT: Single-Step Generative Recommendation via Average Probability Velocity
- Neither Black nor White: Balancing Semantic and Collaborative Signals with Graph-Informed Semantic IDs (GrIS)
- Bootstrapping Conversational Recommendation Agents At Spotify: Synthetic Data Generation and Self-Improvement Loops
- Do Multilingual Encoders Produce Language-Consistent Semantic IDs?
- Beyond the Beam: Constructive Repair and Candidate Completion for Generative Recommendation
- X-Rec Technical Report