时间线
新鲜度是这个库的命门。 别的书架里一本 1995 年的书今天照样成立; 论文不是 —— 半年前的最优做法今天可能已经被推翻。所以这里按时间倒序摆。
共 233 篇,从 2013-01 到 2026-08。 前沿视图 只看最近 12 个月。
看时间要看两个数:
published是它有多老,updated是它的结论被改过没有。 一篇 2021 年的论文若 2026 年还在改,说明它仍是活的。
2026 Q3(12 篇)
2026-08-11Qwen-MusicAVQA-7B: A Multimodal Model for Music Audio-Visual QA
paper=qwen-musicavqa-7b-a-multimodal-model-for-music@arXiv:2608.11329v12026-08-04Qwen-3D: A Generalist 3D Vision-Language Model for Spatial Understanding
paper=qwen-3d-a-generalist-3d-vision-language-model@arXiv:2608.02980v12026-08-03Qwen-CUA: Native Computer Use for (almost) Everything
paper=qwen-cua-native-computer-use-for-almost-everything@arXiv:2608.02352v12026-07-30GLM-RAG: Graph Language Models for Graph-Based Retrieval-Augmented Generation
paper=glm-rag-graph-language-models-for-graph-based@arXiv:2607.28397v12026-07-30Qwen-UI-Agent Technical Report: Toward Next-Generation Real-World Centric Foundation GUI Agents
paper=qwen-ui-agent-technical-report-toward-next-generation@arXiv:2607.28227v12026-07-29Qwen-Audio-3.0-Gen-Preview Technical Report
paper=qwen-audio-3-0-gen-preview-technical-report@arXiv:2607.27011v22026-07-27Kimi K3: Open Frontier Intelligence · 最后修订 2026-08-07
paper=kimi-k3-open-frontier-intelligence@arXiv:2607.24653v22026-07-27Qwen-Audio-3.0-TTS: Freely Controllable and Highly Robust Speech Synthesis with Multi-Stage Training Paradigm
paper=qwen-audio-3-0-tts-freely-controllable-and@arXiv:2607.23938v12026-07-13Interaction Scaling: Grounding the Third Axis of Test-Time Compute
paper=interaction-scaling-grounding-the-third-axis@arXiv:2607.11598v12026-07-13Qwen-Audio-VAE Technical Report
paper=qwen-audio-vae-technical-report@arXiv:2607.11738v12026-07-13Qwen-Music Technical Report
paper=qwen-music-technical-report@arXiv:2607.11699v22026-07-08RLVP: Penalize the Path, Reward the Outcome
paper=rlvp@arXiv:2607.07435v1
2026 Q2(22 篇)
2026-06-29Whose Side Is Your Agent On? Multi-Party Principal Loyalty in LLM Agents
paper=whose-side-is-your-agent-on@arXiv:2606.30383v12026-06-28Agent-Computer Observation Interfaces Enable Dynamic Computer Use
paper=agent-computer-observation-interfaces-enable-dynamic@arXiv:2606.29472v12026-06-25Qwen-Image-2.0-RL Technical Report
paper=qwen-image-2-0-rl-technical-report@arXiv:2606.27608v12026-06-25Qwen-Image-Agent: Bridging the Context Gap in Real-World Image Generation
paper=qwen-image-agent-bridging-the-context-gap-in@arXiv:2606.26907v22026-06-23Qwen-AgentWorld: Language World Models for General Agents
paper=qwen-agentworld-language-world-models-for-general-agents@arXiv:2606.24597v12026-06-23The Latent Bridge: A Continuous Slow-Fast Channel for Real-Time Game Agents
paper=the-latent-bridge-a-continuous-slow@arXiv:2606.24470v12026-06-16Qwen-RobotManip Technical Report: Alignment Unlocks Scale for Robotic Manipulation Foundation Models
paper=qwen-robotmanip-technical-report-alignment-unlocks-scale-for@arXiv:2606.17846v22026-06-16Qwen-RobotNav Technical Report: A Scalable Navigation Model Designed for an Agentic Navigation System
paper=qwen-robotnav-technical-report-a-scalable-navigation-model@arXiv:2606.18112v32026-06-15Qwen-RobotWorld Technical Report: Unifying Embodied World Modeling through Language-Conditioned Video Generation
paper=qwen-robotworld-technical-report-unifying-embodied-world-modeling@arXiv:2606.17030v32026-06-15User as Code: Executable Memory for Personalized Agents
paper=user-as-code-executable-memory-for@arXiv:2606.16707v12026-06-14Models Take Notes at Prefill: KV Cache Can Be Editable and Composable
paper=models-take-notes-at-prefill-kv@arXiv:2606.17107v12026-06-02Qwen-Image-Flash: Beyond Objective Design
paper=qwen-image-flash-beyond-objective-design@arXiv:2606.03746v22026-05-28Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot Embodiments · 最后修订 2026-06-01
paper=qwen-vla-unifying-vision-language-action-modeling-across@arXiv:2605.30280v22026-05-27Qwen-Image-Bench: From Generation to Creation in Text-to-Image Evaluation · 最后修订 2026-06-25
paper=qwen-image-bench-from-generation-to-creation-in@arXiv:2605.28091v22026-05-13Qwen-Image-VAE-2.0 Technical Report
paper=qwen-image-vae-2-0-technical-report@arXiv:2605.13565v12026-05-12Qwen-Scope: Turning Sparse Features into Development Tools for Large Language Models
paper=qwen-scope-turning-sparse-features-into-development-tools@arXiv:2605.11887v12026-05-11Qwen Goes Brrr: Off-the-Shelf RAG for Ukrainian Multi-Domain Document Understanding
paper=qwen-goes-brrr-off-the-shelf-rag-for@arXiv:2605.10296v12026-05-11Qwen-Image-2.0 Technical Report
paper=qwen-image-2-0-technical-report@arXiv:2605.10730v12026-04-29GLM-5V-Turbo: Toward a Native Foundation Model for Multimodal Agents · 最后修订 2026-05-12
paper=glm-5v-turbo-toward-a-native-foundation-model@arXiv:2604.26752v32026-04-27GLM-5 Serving Parameter Tuning for OpenClaw: Single-Deployment MaaS Inference Optimization for Long-Context Agent Workloads
paper=glm-5-serving-parameter-tuning-for-openclaw-single@arXiv:2607.02518v12026-04-26DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
paper=deepseek-v4-towards-highly-efficient-million-token-context@arXiv:2606.19348v12026-04-14DeepSeek Robustness Against Semantic-Character Dual-Space Mutated Prompt Injection
paper=deepseek-robustness-against-semantic-character-dual-space-mutated@arXiv:2604.12548v1
2026 Q1(7 篇)
2026-03-11GLM-OCR Technical Report
paper=glm-ocr-technical-report@arXiv:2603.10910v22026-03-09How Much Do LLMs Hallucinate in Document Q&A Scenarios? A 172-Billion-Token Study Across Temperatures, Context Lengths, and Hardware Platforms
paper=how-much-do-llms-hallucinate-in@arXiv:2603.08274v12026-02-24Qwen-BIM: developing large language model for BIM-based design with domain-specific benchmark and dataset
paper=qwen-bim-developing-large-language-model-for-bim@arXiv:2602.20812v12026-02-17GLM-5: from Vibe Coding to Agentic Engineering
paper=glm-5-from-vibe-coding-to-agentic-engineering@arXiv:2602.15763v22026-02-02Kimi K2.5: Visual Agentic Intelligence · 最后修订 2026-08-07
paper=kimi-k2-5-visual-agentic-intelligence@arXiv:2602.02276v22026-01-28DeepSeek-OCR 2: Visual Causal Flow
paper=deepseek-ocr-2-visual-causal-flow@arXiv:2601.20552v12026-01-08GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization
paper=gdpo@arXiv:2601.05242v1
2025 Q4(8 篇)
2025-12-23Making Large Language Models Efficient Dense Retrievers
paper=making-large-language-models-efficient-dense@arXiv:2512.20612v12025-12-17Qwen-Image-Layered: Towards Inherent Editability via Layer Decomposition
paper=qwen-image-layered-towards-inherent-editability-via-layer@arXiv:2512.15603v12025-12-16GLM-TTS Technical Report
paper=glm-tts-technical-report@arXiv:2512.14291v12025-12-02DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
paper=deepseek-v3-2@arXiv:2512.02556v12025-11-27DeepSeekMath-V2: Towards Self-Verifiable Mathematical Reasoning
paper=deepseekmath-v2@arXiv:2511.22570v12025-10-30Kimi Linear: An Expressive, Efficient Attention Architecture · 最后修订 2025-11-01
paper=kimi-linear-an-expressive-efficient-attention-architecture@arXiv:2510.26692v22025-10-21DeepSeek-OCR: Contexts Optical Compression
paper=deepseek-ocr-contexts-optical-compression@arXiv:2510.18234v12025-10-16Equipping agents for the real world with Agent Skills
article=anthropic-agent-skills@2026-08-30
2025 Q3(8 篇)
2025-09-27Kimi-Dev: Agentless Training as Skill Prior for SWE-Agents · 最后修订 2025-12-08
paper=kimi-dev-agentless-training-as-skill-prior-for@arXiv:2509.23045v32025-09-02DeepSeek performs better than other Large Language Models in Dental Cases
paper=deepseek-performs-better-than-other-large-language-models@arXiv:2509.02036v12025-08-08GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models
paper=glm-4-5-agentic-reasoning-and-coding-arc@arXiv:2508.06471v12025-08-04Qwen-Image Technical Report
paper=qwen-image-technical-report@arXiv:2508.02324v12025-07-28Kimi K2: Open Agentic Intelligence · 最后修订 2026-02-03
paper=kimi-k2-open-agentic-intelligence@arXiv:2507.20534v22025-07-24Group Sequence Policy Optimization
paper=group-sequence-policy-optimization@arXiv:2507.18071v22025-07-14DeepSeek: Paradigm Shifts and Technical Evolution in Large AI Models
paper=deepseek-paradigm-shifts-and-technical-evolution-in-large@arXiv:2507.09955v12025-07-01GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning · 最后修订 2026-01-01
paper=glm-4-5v-and-glm-4-1v-thinking@arXiv:2507.01006v6
2025 Q2(14 篇)
2025-06-16MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention
paper=minimax-m1@arXiv:2506.13585v12025-06-16Qwen vs. Gemma Integration with Whisper: A Comparative Study in Multilingual SpeechLLM Systems · 最后修订 2025-07-07
paper=qwen-vs-gemma-integration-with-whisper-a-comparative@arXiv:2506.13596v22025-06-02DeepSeek in Healthcare: A Survey of Capabilities, Risks, and Clinical Applications of Open-Source Large Language Models
paper=deepseek-in-healthcare-a-survey-of-capabilities-risks@arXiv:2506.01257v12025-06-01MCP-Zero: Active Tool Discovery for Autonomous LLM Agents
paper=mcp-zero@arXiv:2506.01056v42025-05-29Qwen Look Again: Guiding Vision-Language Reasoning Models to Re-attention Visual Information
paper=qwen-look-again-guiding-vision-language-reasoning-models@arXiv:2505.23558v22025-05-14Qwen3 Technical Report
paper=qwen3-technical-report@arXiv:2505.09388v12025-05-01FineScope : SAE-guided Data Selection Enables Domain Specific LLM Pruning and Finetuning · 最后修订 2026-02-27
paper=finescope-sae-guided-data-selection-enables@arXiv:2505.00624v32025-04-30DeepSeek-Prover-V2: Advancing Formal Mathematical Reasoning via Reinforcement Learning for Subgoal Decomposition · 最后修订 2025-07-18
paper=deepseek-prover-v2-advancing-formal-mathematical-reasoning-via@arXiv:2504.21801v22025-04-29The Leaderboard Illusion · 最后修订 2025-05-12
paper=the-leaderboard-illusion@arXiv:2504.20879v22025-04-25Kimi-Audio Technical Report
paper=kimi-audio-technical-report@arXiv:2504.18425v12025-04-18Rethinking Inference-Time Scaling: Efficiency Limits and Linguistic Signals · 最后修订 2026-08-01
paper=rethinking-inference-time-scaling-efficiency-limits@arXiv:2504.14047v22025-04-10DeepSeek-R1 vs. o3-mini: How Well can Reasoning LLMs Evaluate MT and Summarization? · 最后修订 2025-05-30
paper=deepseek-r1-vs-o3-mini-how-well-can@arXiv:2504.08120v32025-04-10Kimi-VL Technical Report · 最后修订 2025-06-23
paper=kimi-vl-technical-report@arXiv:2504.07491v32025-04-02DeepSeek-R1 Thoughtology: Let's think about LLM Reasoning · 最后修订 2026-01-15
paper=deepseek-r1-thoughtology-lets-think-about-llm-reasoning@arXiv:2504.07128v3
2025 Q1(14 篇)
2025-03-27GLM Inference with AI-Generated Synthetic Data Using Misspecified Linear Regression · 最后修订 2025-07-14
paper=glm-inference-with-ai-generated-synthetic-data-using@arXiv:2503.21968v22025-03-26Understanding R1-Zero-Like Training: A Critical Perspective · 最后修订 2025-10-06
paper=understanding-r1-zero-like-training-a@arXiv:2503.20783v22025-03-18DAPO: An Open-Source LLM Reinforcement Learning System at Scale · 最后修订 2025-05-20
paper=dapo@arXiv:2503.14476v22025-03-14DeepSeek Powered Solid Dosage Formulation Design and Development
paper=deepseek-powered-solid-dosage-formulation-design-and-development@arXiv:2503.11068v22025-03-13DeepSeek-Inspired Exploration of RL-based LLMs and Synergy with Wireless Networks: A Survey · 最后修订 2025-10-20
paper=deepseek-inspired-exploration-of-rl-based-llms-and@arXiv:2503.09956v42025-02-25DeepSeek-R1 Outperforms Gemini 2.0 Pro, OpenAI o1, and o3-mini in Bilingual Complex Ophthalmology Reasoning
paper=deepseek-r1-outperforms-gemini-2-0-pro-openai@arXiv:2502.17947v12025-02-25DeepSeek vs. ChatGPT vs. Claude: A Comparative Study for Scientific Computing and Scientific Machine Learning Tasks · 最后修订 2025-03-08
paper=deepseek-vs-chatgpt-vs-claude-a-comparative-study@arXiv:2502.17764v22025-02-23DeepSeek reshaping healthcare in China's tertiary hospitals
paper=deepseek-reshaping-healthcare-in-chinas-tertiary-hospitals@arXiv:2502.16732v22025-02-19DeepSeek-V3, GPT-4, Phi-4, and LLaMA-3.3 generate correct code for LoRaWAN-related engineering tasks · 最后修订 2025-04-07
paper=deepseek-v3-gpt-4-phi-4-and-llama@arXiv:2502.14926v32025-02-11DeepSeek on a Trip: Inducing Targeted Visual Hallucinations via Representation Vulnerabilities
paper=deepseek-on-a-trip-inducing-targeted-visual-hallucinations@arXiv:2502.07905v12025-01-28SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training · 最后修订 2025-05-26
paper=sft-memorizes-rl-generalizes-a-comparative@arXiv:2501.17161v22025-01-22DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning · 最后修订 2026-01-04
paper=deepseek-r1@arXiv:2501.12948v22025-01-22Kimi k1.5: Scaling Reinforcement Learning with LLMs · 最后修订 2025-06-03
paper=kimi-k1-5-scaling-reinforcement-learning-with-llms@arXiv:2501.12599v42025-01-16Qwen it detect machine-generated text?
paper=qwen-it-detect-machine-generated-text@arXiv:2501.09813v1
2024 Q4(4 篇)
2024-12-27DeepSeek-V3 Technical Report · 最后修订 2025-02-18
paper=deepseek-v3-technical-report@arXiv:2412.19437v22024-12-19Qwen2.5 Technical Report · 最后修订 2025-01-03
paper=qwen2-5-technical-report@arXiv:2412.15115v22024-12-03GLM-4-Voice: Towards Intelligent and Human-Like End-to-End Spoken Chatbot
paper=glm-4-voice-towards-intelligent-and-human-like@arXiv:2412.02612v12024-10-28AutoGLM: Autonomous Foundation Agents for GUIs
paper=autoglm-autonomous-foundation-agents-for-guis@arXiv:2411.00820v1
2024 Q3(3 篇)
2024-09-07Late Chunking: Contextual Chunk Embeddings Using Long-Context Embedding Models · 最后修订 2025-07-07
paper=late-chunking-contextual-chunk-embeddings-using@arXiv:2409.04701v32024-08-19Docling Technical Report · 最后修订 2024-12-09
paper=docling-technical-report@arXiv:2408.09869v52024-07-31The Llama 3 Herd of Models · 最后修订 2024-11-23
paper=the-llama-3-herd-of-models@arXiv:2407.21783v3
2024 Q2(7 篇)
2024-06-22What Matters in Transformers? Not All Attention is Needed · 最后修订 2024-10-17
paper=what-matters-in-transformers-not-all@arXiv:2406.15786v62024-06-18ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools · 最后修订 2024-07-30
paper=chatglm-a-family-of-large-language-models-from@arXiv:2406.12793v22024-04-24From Local to Global: A Graph RAG Approach to Query-Focused Summarization · 最后修订 2025-02-19
paper=graphrag@arXiv:2404.16130v22024-04-22Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone · 最后修订 2024-08-30
paper=phi-3-technical-report-a-highly-capable-language@arXiv:2404.14219v42024-04-22Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone · 最后修订 2024-08-30
paper=phi-3-technical-report-a-highly@arXiv:2404.14219v42024-04-03ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline
paper=chatglm-math-improving-math-problem-solving-in-large@arXiv:2404.02893v12024-04-01ChatGLM-RLHF: Practices of Aligning Large Language Models with Human Feedback
paper=chatglm-rlhf-practices-of-aligning-large-language-models@arXiv:2404.00934v2
2024 Q1(8 篇)
2024-03-13Gemma: Open Models Based on Gemini Research and Technology · 最后修订 2024-04-16
paper=gemma-open-models-based-on-gemini-research-and@arXiv:2403.08295v42024-03-12ORPO: Monolithic Preference Optimization without Reference Model
paper=orpo@arXiv:2403.07691v22024-03-08Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context · 最后修订 2024-12-16
paper=gemini-1-5-unlocking-multimodal-understanding@arXiv:2403.05530v52024-03-07Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference
paper=chatbot-arena-an-open-platform-for@arXiv:2403.04132v12024-02-27Evaluating Very Long-Term Conversational Memory of LLM Agents
paper=evaluating-very-long-term-conversational-memory@arXiv:2402.17753v12024-02-05DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models · 最后修订 2024-04-27
paper=deepseekmath-grpo@arXiv:2402.03300v32024-01-10Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
paper=sleeper-agents-training-deceptive-llms-that-persist-through@arXiv:2401.05566v32024-01-08Mixtral of Experts
paper=mixtral-of-experts@arXiv:2401.04088v1
2023 Q4(11 篇)
2023-12-19Gemini: A Family of Highly Capable Multimodal Models · 最后修订 2025-05-09
paper=gemini-a-family-of-highly-capable-multimodal-models@arXiv:2312.11805v52023-12-11Dense X Retrieval: What Retrieval Granularity Should We Use? · 最后修订 2024-10-04
paper=dense-x-retrieval-what-retrieval-granularity@arXiv:2312.06648v32023-12-01Mamba: Linear-Time Sequence Modeling with Selective State Spaces · 最后修订 2024-05-31
paper=mamba@arXiv:2312.00752v22023-11-20GPQA: A Graduate-Level Google-Proof Q&A Benchmark
paper=gpqa@arXiv:2311.12022v12023-11-16When "A Helpful Assistant" Is Not Really Helpful: Personas in System Prompts Do Not Improve Performances of Large Language Models · 最后修订 2024-10-09
paper=when-a-helpful-assistant-is-not@arXiv:2311.10054v32023-11-14Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models · 最后修订 2023-12-21
paper=qwen-audio-advancing-universal-audio-understanding-via-unified@arXiv:2311.07919v22023-11-06CogVLM: Visual Expert for Pretrained Language Models · 最后修订 2024-02-04
paper=cogvlm-visual-expert-for-pretrained-language-models@arXiv:2311.03079v22023-10-10Mistral 7B
paper=mistral-7b@arXiv:2310.06825v12023-10-09Take a Step Back: Evoking Reasoning via Abstraction in Large Language Models · 最后修订 2024-03-12
paper=take-a-step-back-evoking-reasoning@arXiv:2310.06117v22023-10-05DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines
paper=dspy@arXiv:2310.03714v12023-10-03Large Language Models Cannot Self-Correct Reasoning Yet · 最后修订 2024-03-14
paper=large-language-models-cannot-self-correct@arXiv:2310.01798v2
2023 Q3(10 篇)
2023-09-28Qwen Technical Report
paper=qwen-technical-report@arXiv:2309.16609v12023-09-26Ragas: Automated Evaluation of Retrieval Augmented Generation · 最后修订 2025-04-28
paper=ragas@arXiv:2309.15217v22023-09-15EvoPrompt: Connecting LLMs with Evolutionary Algorithms Yields Powerful Prompt Optimizers · 最后修订 2025-05-01
paper=evoprompt@arXiv:2309.08532v32023-09-07Large Language Models as Optimizers · 最后修订 2024-04-15
paper=large-language-models-as-optimizers@arXiv:2309.03409v32023-08-24Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond · 最后修订 2023-10-13
paper=qwen-vl-a-versatile-vision-language-model-for@arXiv:2308.12966v32023-08-01MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework · 最后修订 2024-11-01
paper=metagpt@arXiv:2308.00352v72023-07-31Reinforcement Learning for Generative AI: State of the Art, Opportunities and Open Research Challenges · 最后修订 2024-02-08
paper=reinforcement-learning-for-generative-ai-state@arXiv:2308.00031v42023-07-18Llama 2: Open Foundation and Fine-Tuned Chat Models
paper=llama-2-open-foundation-and-fine-tuned-chat@arXiv:2307.09288v22023-07-18Llama 2: Open Foundation and Fine-Tuned Chat Models
paper=llama-2-open-foundation-and-fine@arXiv:2307.09288v22023-07-06Lost in the Middle: How Language Models Use Long Contexts · 最后修订 2023-11-20
paper=lost-in-the-middle@arXiv:2307.03172v3
2023 Q2(12 篇)
2023-06-20Textbooks Are All You Need · 最后修订 2023-10-02
paper=textbooks-are-all-you-need@arXiv:2306.11644v22023-06-14MiniLLM: On-Policy Distillation of Large Language Models · 最后修订 2026-01-31
paper=minillm@arXiv:2306.08543v62023-06-09Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena · 最后修订 2023-12-24
paper=judging-llm-as-a-judge-with@arXiv:2306.05685v42023-05-31Let's Verify Step by Step
paper=lets-verify-step-by-step@arXiv:2305.20050v12023-05-29Direct Preference Optimization: Your Language Model is Secretly a Reward Model · 最后修订 2024-07-29
paper=dpo@arXiv:2305.18290v32023-05-27The Curse of Recursion: Training on Generated Data Makes Models Forget · 最后修订 2024-04-14
paper=the-curse-of-recursion-training-on@arXiv:2305.17493v32023-05-23QLoRA: Efficient Finetuning of Quantized LLMs
paper=qlora@arXiv:2305.14314v12023-05-22GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints · 最后修订 2023-12-23
paper=gqa@arXiv:2305.13245v32023-05-17Tree of Thoughts: Deliberate Problem Solving with Large Language Models · 最后修订 2023-12-03
paper=tree-of-thoughts@arXiv:2305.10601v22023-04-28Are Emergent Abilities of Large Language Models a Mirage? · 最后修订 2023-05-22
paper=emergent-abilities-a-mirage@arXiv:2304.15004v22023-04-28Segment Anything Model for Medical Images? · 最后修订 2024-01-17
paper=segment-anything-model-for-medical-images@arXiv:2304.14660v72023-04-05Segment Anything
paper=segment-anything@arXiv:2304.02643v1
2023 Q1(8 篇)
2023-03-30CodeGeeX: A Pre-Trained Model for Code Generation with Multilingual Benchmarking on HumanEval-X · 最后修订 2024-07-10
paper=codegeex-a-pre-trained-model-for-code-generation@arXiv:2303.17568v22023-03-15GPT-4 Technical Report · 最后修订 2024-03-04
paper=gpt-4-technical-report@arXiv:2303.08774v62023-02-28GLM-Dialog: Noise-tolerant Pre-training for Knowledge-grounded Dialogue Generation
paper=glm-dialog-noise-tolerant-pre-training-for-knowledge@arXiv:2302.14401v12023-02-27LLaMA: Open and Efficient Foundation Language Models
paper=llama-open-and-efficient-foundation-language-models@arXiv:2302.13971v12023-02-27LLaMA: Open and Efficient Foundation Language Models
paper=llama@arXiv:2302.13971v12023-02-23Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection · 最后修订 2023-05-05
paper=not-what-youve-signed-up-for@arXiv:2302.12173v22023-01-30BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models · 最后修订 2023-06-15
paper=blip-2@arXiv:2301.12597v32023-01-26On the Importance of Noise Scheduling for Diffusion Models · 最后修订 2023-05-21
paper=on-the-importance-of-noise-scheduling@arXiv:2301.10972v4
2022 Q4(13 篇)
2022-12-20Precise Zero-Shot Dense Retrieval without Relevance Labels
paper=hyde@arXiv:2212.10496v12022-12-15Constitutional AI: Harmlessness from AI Feedback
paper=constitutional-ai-harmlessness-from-ai-feedback@arXiv:2212.08073v12022-12-06Robust Speech Recognition via Large-Scale Weak Supervision
paper=robust-speech-recognition-via-large-scale-weak-supervision@arXiv:2212.04356v12022-12-06Robust Speech Recognition via Large-Scale Weak Supervision
paper=robust-speech-recognition-via-large-scale@arXiv:2212.04356v12022-11-25GLM for partially pooled categorical predictors with a case study in biosecurity
paper=glm-for-partially-pooled-categorical-predictors-with-a@arXiv:2211.13848v12022-11-03Large Language Models Are Human-Level Prompt Engineers · 最后修订 2023-03-10
paper=large-language-models-are-human-level@arXiv:2211.01910v22022-10-26Will we run out of data? Limits of LLM scaling based on human-generated data · 最后修订 2024-06-04
paper=will-we-run-out-of-data@arXiv:2211.04325v22022-10-20Scaling Instruction-Finetuned Language Models · 最后修订 2022-12-06
paper=scaling-instruction-finetuned-language-models@arXiv:2210.11416v52022-10-19TabLLM: Few-shot Classification of Tabular Data with Large Language Models · 最后修订 2023-03-17
paper=tabllm@arXiv:2210.10723v22022-10-13MTEB: Massive Text Embedding Benchmark · 最后修订 2023-03-19
paper=mteb@arXiv:2210.07316v32022-10-07Automatic Chain of Thought Prompting in Large Language Models
paper=automatic-chain-of-thought-prompting-in@arXiv:2210.03493v12022-10-06ReAct: Synergizing Reasoning and Acting in Language Models · 最后修订 2023-03-10
paper=react@arXiv:2210.03629v32022-10-05GLM-130B: An Open Bilingual Pre-trained Model · 最后修订 2023-10-25
paper=glm-130b-an-open-bilingual-pre-trained-model@arXiv:2210.02414v2
2022 Q3(1 篇)
2022-09-22Efficient Few-Shot Learning Without Prompts
paper=efficient-few-shot-learning-without-prompts@arXiv:2209.11055v1
2022 Q2(7 篇)
2022-06-15Emergent Abilities of Large Language Models · 最后修订 2022-10-26
paper=emergent-abilities-of-large-language-models@arXiv:2206.07682v22022-05-27FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness · 最后修订 2022-06-23
paper=flashattention@arXiv:2205.14135v22022-05-25RLPrompt: Optimizing Discrete Text Prompts with Reinforcement Learning · 最后修订 2022-10-22
paper=rlprompt@arXiv:2205.12548v32022-05-24Large Language Models are Zero-Shot Reasoners · 最后修订 2023-01-29
paper=zero-shot-cot@arXiv:2205.11916v42022-04-13Hierarchical Text-Conditional Image Generation with CLIP Latents
paper=hierarchical-text-conditional-image-generation-with-clip-latents@arXiv:2204.06125v12022-04-12Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
paper=training-a-helpful-and-harmless-assistant-with-reinforcement@arXiv:2204.05862v12022-04-05PaLM: Scaling Language Modeling with Pathways · 最后修订 2022-10-05
paper=palm-scaling-language-modeling-with-pathways@arXiv:2204.02311v5
2022 Q1(7 篇)
2022-03-29Training Compute-Optimal Large Language Models
paper=chinchilla@arXiv:2203.15556v12022-03-21Self-Consistency Improves Chain of Thought Reasoning in Language Models · 最后修订 2023-03-07
paper=self-consistency@arXiv:2203.11171v42022-03-11BERTopic: Neural topic modeling with a class-based TF-IDF procedure
paper=bertopic@arXiv:2203.05794v12022-03-10Conditional Prompt Learning for Vision-Language Models · 最后修订 2022-10-06
paper=conditional-prompt-learning-for-vision-language@arXiv:2203.05557v22022-03-04Training language models to follow instructions with human feedback
paper=instructgpt@arXiv:2203.02155v12022-02-25Rethinking the Role of Demonstrations: What Makes In-Context Learning Work? · 最后修订 2022-10-20
paper=rethinking-the-role-of-demonstrations-what@arXiv:2202.12837v22022-01-28Chain-of-Thought Prompting Elicits Reasoning in Large Language Models · 最后修订 2023-01-10
paper=chain-of-thought@arXiv:2201.11903v6
2021 Q4(3 篇)
2021-11-30Show Your Work: Scratchpads for Intermediate Computation with Language Models
paper=show-your-work-scratchpads-for-intermediate@arXiv:2112.00114v12021-10-27Training Verifiers to Solve Math Word Problems · 最后修订 2021-11-18
paper=training-verifiers-to-solve-math-word@arXiv:2110.14168v22021-10-15Generated Knowledge Prompting for Commonsense Reasoning · 最后修订 2022-09-28
paper=generated-knowledge-prompting-for-commonsense-reasoning@arXiv:2110.08387v3
2021 Q3(5 篇)
2021-09-11Physics-based Deep Learning · 最后修订 2025-03-26
paper=physics-based-deep-learning@arXiv:2109.05237v42021-09-03Finetuned Language Models Are Zero-Shot Learners · 最后修订 2022-02-08
paper=finetuned-language-models-are-zero-shot-learners@arXiv:2109.01652v52021-09-02Learning to Prompt for Vision-Language Models · 最后修订 2022-10-06
paper=learning-to-prompt-for-vision-language@arXiv:2109.01134v62021-08-16On the Opportunities and Risks of Foundation Models · 最后修订 2022-07-12
paper=on-the-opportunities-and-risks-of@arXiv:2108.07258v32021-07-07Evaluating Large Language Models Trained on Code
paper=evaluating-large-language-models-trained-on-code@arXiv:2107.03374v2
2021 Q2(5 篇)
2021-06-23On a new statistical technique for the real-time recognition of ultra-low multiplicity astrophysical neutrino burst
paper=on-a-new-statistical-technique-for@arXiv:2106.12345v12021-06-20NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction · 最后修订 2023-02-01
paper=neus@arXiv:2106.10689v32021-06-17LoRA: Low-Rank Adaptation of Large Language Models · 最后修订 2021-10-16
paper=lora@arXiv:2106.09685v22021-06-02Decision Transformer: Reinforcement Learning via Sequence Modeling
paper=decision-transformer-reinforcement-learning-via-sequence@arXiv:2106.01345v22021-04-20RoFormer: Enhanced Transformer with Rotary Position Embedding · 最后修订 2023-11-08
paper=rope@arXiv:2104.09864v5
2021 Q1(5 篇)
2021-03-18GLM: General Language Model Pretraining with Autoregressive Blank Infilling · 最后修订 2022-03-17
paper=glm@arXiv:2103.10360v22021-03-05Measuring Mathematical Problem Solving With the MATH Dataset · 最后修订 2021-11-08
paper=measuring-mathematical-problem-solving-with-the@arXiv:2103.03874v22021-02-26Learning Transferable Visual Models From Natural Language Supervision
paper=clip@arXiv:2103.00020v12021-02-24Zero-Shot Text-to-Image Generation
paper=zero-shot-text-to-image-generation@arXiv:2102.12092v22021-01-01Prefix-Tuning: Optimizing Continuous Prompts for Generation
paper=prefix-tuning@arXiv:2101.00190v1
2020 Q4(2 篇)
2020-12-22Intrinsic Dimensionality Explains the Effectiveness of Language Model Fine-Tuning
paper=intrinsic-dimensionality-explains-the-effectiveness-of@arXiv:2012.13255v12020-10-22An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale · 最后修订 2021-06-03
paper=vit@arXiv:2010.11929v2
2020 Q3(1 篇)
2020-09-07Measuring Massive Multitask Language Understanding · 最后修订 2021-01-12
paper=mmlu@arXiv:2009.03300v3
2020 Q2(4 篇)
2020-06-19Denoising Diffusion Probabilistic Models · 最后修订 2020-12-16
paper=denoising-diffusion-probabilistic-models@arXiv:2006.11239v22020-05-28Language Models are Few-Shot Learners · 最后修订 2020-07-22
paper=gpt-3@arXiv:2005.14165v42020-05-22Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks · 最后修订 2021-04-12
paper=rag@arXiv:2005.11401v42020-04-27First return, then explore · 最后修订 2021-09-16
paper=first-return-then-explore@arXiv:2004.12919v6
2020 Q1(3 篇)
2020-02-12GLU Variants Improve Transformer
paper=glu-variants-improve-transformer@arXiv:2002.05202v12020-02-12On Layer Normalization in the Transformer Architecture · 最后修订 2020-06-29
paper=on-layer-normalization-in-the-transformer@arXiv:2002.04745v22020-01-23Scaling Laws for Neural Language Models
paper=scaling-laws-for-neural-language-models@arXiv:2001.08361v1
2019 Q4(6 篇)
2019-11-19Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model · 最后修订 2020-02-21
paper=mastering-atari-go-chess-and-shogi@arXiv:1911.08265v22019-11-06Fast Transformer Decoding: One Write-Head is All You Need
paper=fast-transformer-decoding-one-write-head@arXiv:1911.02150v12019-10-23Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer · 最后修订 2023-09-19
paper=exploring-the-limits-of-transfer-learning-with-a@arXiv:1910.10683v42019-10-23Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer · 最后修订 2023-09-19
paper=exploring-the-limits-of-transfer-learning@arXiv:1910.10683v42019-10-16Root Mean Square Layer Normalization
paper=root-mean-square-layer-normalization@arXiv:1910.07467v12019-10-04ZeRO: Memory Optimizations Toward Training Trillion Parameter Models · 最后修订 2020-05-13
paper=zero@arXiv:1910.02054v3
2019 Q3(3 篇)
2019-09-26ALBERT: A Lite BERT for Self-supervised Learning of Language Representations · 最后修订 2020-02-09
paper=albert@arXiv:1909.11942v62019-08-27Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks
paper=sentence-bert@arXiv:1908.10084v12019-07-26RoBERTa: A Robustly Optimized BERT Pretraining Approach
paper=roberta@arXiv:1907.11692v1
2019 Q2(1 篇)
2019-05-19HellaSwag: Can a Machine Really Finish Your Sentence?
paper=hellaswag@arXiv:1905.07830v1
2019 Q1(1 篇)
2019-02-02Parameter-Efficient Transfer Learning for NLP · 最后修订 2019-06-13
paper=parameter-efficient-transfer-learning-for-nlp@arXiv:1902.00751v2
2018 Q4(1 篇)
2018-10-11BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding · 最后修订 2019-05-24
paper=bert@arXiv:1810.04805v2
2018 Q3(1 篇)
2018-08-19SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing
paper=sentencepiece@arXiv:1808.06226v1
2018 Q2(1 篇)
2018-04-29Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidates
paper=subword-regularization-improving-neural-network-translation@arXiv:1804.10959v1
2018 Q1(2 篇)
2018-03-14Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
paper=think-you-have-solved-question-answering@arXiv:1803.05457v12018-03-09The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks · 最后修订 2019-03-04
paper=the-lottery-ticket-hypothesis-finding-sparse@arXiv:1803.03635v5
2017 Q3(2 篇)
2017-09-03Einstein's Patents and Inventions
paper=einsteins-patents-and-inventions@arXiv:1709.00666v22017-07-20Proximal Policy Optimization Algorithms · 最后修订 2017-08-28
paper=proximal-policy-optimization-algorithms@arXiv:1707.06347v2
2017 Q2(1 篇)
2017-06-12Attention Is All You Need · 最后修订 2023-08-02
paper=attention-is-all-you-need@arXiv:1706.03762v7
2016 Q3(2 篇)
2016-09-26Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation · 最后修订 2016-10-08
paper=googles-neural-machine-translation-system-bridging@arXiv:1609.08144v22016-07-21Layer Normalization
paper=layer-normalization@arXiv:1607.06450v1
2016 Q2(1 篇)
2016-06-27Gaussian Error Linear Units (GELUs) · 最后修订 2023-06-06
paper=gaussian-error-linear-units-gelus@arXiv:1606.08415v5
2015 Q4(2 篇)
2015-12-10Deep Residual Learning for Image Recognition
paper=deep-residual-learning-for-image-recognition@arXiv:1512.03385v12015-11-19Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks · 最后修订 2016-01-07
paper=unsupervised-representation-learning-with-deep-convolutional@arXiv:1511.06434v2
2015 Q3(1 篇)
2015-08-31Neural Machine Translation of Rare Words with Subword Units · 最后修订 2016-06-10
paper=neural-machine-translation-of-rare-words@arXiv:1508.07909v5
2015 Q1(2 篇)
2015-03-09Distilling the Knowledge in a Neural Network
paper=distilling-the-knowledge-in-a-neural@arXiv:1503.02531v12015-02-11Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift · 最后修订 2015-03-02
paper=batch-normalization-accelerating-deep-network-training@arXiv:1502.03167v3
2013 Q4(1 篇)
2013-12-20Auto-Encoding Variational Bayes · 最后修订 2022-12-10
paper=auto-encoding-variational-bayes@arXiv:1312.6114v11
2013 Q1(1 篇)
2013-01-16Efficient Estimation of Word Representations in Vector Space · 最后修订 2013-09-07
paper=word2vec@arXiv:1301.3781v3