← 返回列表

標籤

共 403 個標籤，其中 88 個橫跨多篇論文。點一下即可篩選。

LLM58 開源模型15 MoE14 Reasoning14 Agent12 Benchmark10 Tool Use8 Long Context7 Knowledge Distillation6 RLVR6 SFT6 VLM6 GRPO5 Linear Attention5 Pre-training5 Reinforcement Learning5 Synthetic Data5 Distillation4 LLM Agent4 Mamba4 Multimodal4 Muon4 RLHF4 Scaling Law4 Self-Evolution4 AI Agent3 Continual Learning3 Evaluation3 Inference3 LLM-as-Judge3 MCP3 Mid-training3 Multi-Agent3 OCR3 Pretraining3 RL3 World Model3 合成資料3 Agentic2 Agentic RL2 AI Safety2 AutoML2 Catastrophic Forgetting2 Code Generation2 Credit Assignment2 Cross-Architecture2 Data Recipe2 Document Parsing2 Fine-tuning2 FP82 Interpretability2 KV Cache2 Llama2 LLM-as-a-Judge2 Memory2 Meta2 Model Merging2 Multi-Agent Systems2 Open Dataset2 OpenAI2 Physical AI2 Post-training2 Post-Training2 Prompt Injection2 Qwen2 Scaling2 Small Language Model2 Speculative Decoding2 SSM2 SWE-bench2 Tokenizer2 Transformer2 可解釋性2 多模態2 多語言2 強化學習2 影片生成2 持續預訓練2 推理2 數學推理2 文件解析2 機器人2 系統優化2 綜述2 訓練效率2 語言模型2 長上下文2 預訓練2

只出現在單篇論文的標籤（315）

3D Stacking1 Activation1 Adaptive Computation1 Agent Skills1 Agentic AI1 AGI1 AI Scientist1 Allen AI1 Apple1 ARC-AGI1 Attention1 Attention Mask1 Attention Mechanism1 Autoregressive1 Bias1 ByteDance1 Chain-of-Thought1 Chinchilla1 Chinese NLP1 CISPO1 Claude Code1 CLI1 Code Agent1 Code Synthesis1 Common Crawl1 Compression1 Computer System Validation1 Context Extension1 Continued Pre-training1 CoT1 Cross-Entropy1 Cross-Tokenizer1 Cryptographic Protocol1 CUDA Kernel1 DAPO1 Data1 Data Augmentation1 Data Collection1 Data Curation1 Data Mixing1 Data Parallelism1 Data-Centric AI1 Database1 Datacenter AI1 Decoding1 DeepSeek1 Delegated Work1 Delegation1 Delta Rule1 DeltaNet1 Democratic Discourse1 Diffusion1 Diffusion LLM1 Diffusion Model1 Diffusion Transformer1 Disaggregation1 Distributionally Robust Optimization1 Document Editing1 Document VQA1 Domain Reweighting1 DPPO1 Ed255191 Emergent Behavior1 Few-Shot Learning1 Financial Agent1 Financial LLM1 Financial Security1 Fisher-Rao1 Flash Memory1 FrogNano1 Gating1 Generalization1 GPT-31 GPT-51 Gradient Noise Scale1 GUI Agent1 Hallucination1 Hardware1 Harness1 Harness Engineering1 Harness Optimization1 HBF1 HBM1 HDPO1 Hedgehog1 Hybrid Architecture1 Hybrid Attention1 Hybrid Model1 Hypergraph1 ICLR20231 ICML20241 IETF1 Image Captioning1 In-Context Learning1 Information Bottleneck1 Information Extraction1 Information Theory1 Inverse RL1 Karcher Mean1 Kimi1 KL Divergence1 Knowledge Integration1 Large Vocabulary1 Large-Batch Training1 Latent Reasoning1 Layout Detection1 Lightning Attention1 Linear RNN1 LLaMA1 LLM Agents1 LLM Alignment1 LLM Inference1 LLM Routing1 LLM Training1 LLM推理1 LLM規劃1 Load Balancing1 Long CoT1 Long-Context1 Long-Horizon Evaluation1 LoopLM1 Low-Cost Training1 Mamba-21 Mamba21 MCTS1 Memory Efficiency1 Memory Wall1 MergeKit1 Meta-cognition1 Meta-Cognition1 Meta-Optimization1 microVM1 Mind Virus1 Mixed-Task Agentic RL1 Mixture-of-Experts1 MLP1 Model Ensemble1 Model Routing1 Monte Carlo Graph Search1 MTP1 Multi-Agent System1 Multi-Capability1 Multi-Head Attention1 Multi-objective Optimization1 Multi-token Prediction1 Multilingual1 Neural Computer1 NeurIPS1 NLP1 OLMo1 Olympiad1 OmniDocBench1 On-device1 On-policy1 On-Policy1 On-Policy Distillation1 On-Policy Learning1 Open Recipe1 Optimization1 Optimization Dynamics1 Optimizer1 Parallelism1 Parameter Efficiency1 Pareto Frontier1 Peer-Preservation1 Performance-Cost1 Policy Distillation1 Position Interpolation1 Position Paper1 Preference Optimization1 Pretraining Data1 Probing1 Process Reward1 Processing-Near-Memory1 Production1 Prompt Engineering1 Prompting1 Protocol1 Protocols1 Provenance1 Pruning1 Quantization1 Qwen2.51 Qwen2.5-VL1 Qwen31 Qwen3.51 RAG1 Reasoning Model1 Repeated Data1 Replay1 Representation Learning1 Reward Design1 Reward Hacking Defense1 Riemannian Geometry1 Ring Attention1 RoPE1 Routing1 RTX 50901 Rubric Reward1 RULER1 Safety1 Schmidhuber1 Security1 Self-Distillation1 Self-Improvement1 Self-Instruct1 Sequence Modeling1 SGD1 SIGMOD1 Skill2Env1 Skills1 SLERP1 Software Architecture1 Software Engineering1 Sparse1 Sparse Attention1 Sparse Autoencoder1 Sparse Update1 SQL1 State Space Duality1 Summarization1 Survey1 System Design1 T51 TCO1 Test-time Scaling1 Test-Time Training1 Text-to-Image1 Tülu1 UDF1 Universal Transformer1 Video Model1 Vision Transformer1 Vision-Language1 Visual Perception1 VQA1 Wan2.11 Web Agent1 Web Scraping1 WSD Scheduler1 Xiaohongshu1 世界知識1 中文評測1 中期訓練1 事實性1 互動式環境1 代理式推理1 代理式搜尋1 分散式檔案系統1 初始化1 判決預測1 剪枝1 加速1 動態剪枝1 台灣法律1 合成任務1 合成環境1 向量量化1 圖搜尋1 基礎模型1 基礎設施1 多代理系統1 安全性1 容器1 小型模型1 小模型1 工具使用1 工具創建1 思維鏈監控1 成本效益1 技術報告1 指令遵循1 推測式解碼1 推測解碼1 推理對齊1 推理模型1 推論加速1 擴散模型1 數據增強1 數據生成1 數據集生成1 架構設計1 模型再造1 沙箱1 法律 NLP1 混合注意力1 潛在推理1 災難性遺忘1 知行落差1 知識整合1 稀疏模型1 程式碼代理1 程式碼智能體1 端側部署1 策略發現1 紅隊測試1 終端代理1 結構化剪枝1 繁體中文1 自動駕駛1 自我進化1 規劃1 訓練策略1 評測1 詞彙擴充1 資料合成1 資料多樣性1 資料生成1 資料選擇1 資訊理論1 近鄰搜尋1 零樣本1 預訓練動態1 高效訓練1 黑盒模型1

本站翻譯內容基於 arXiv 公開論文，並標註原始授權。CC BY 授權論文可自由閱覽，其他授權論文需登入後閱讀。若原作者有疑慮，請來信，我們將立即處理。

有任何功能建議，歡迎來信 [email protected]