| 09 | Chain-of-Thought Prompting Elicits Reasoning in Large Language Models | 2022 | link |
| 10 | LoRA: Low-Rank Adaptation of Large Language Models | 2021 | link |
| 12 | Scaling Laws for Neural Language Models | 2020 | link |
| 13 | Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks (RAG) | 2020 | link |
| 16 | FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness | 2022 | link |
| 18 | Training Compute-Optimal Large Language Models (Chinchilla) | 2022 | link |
| 21 | ReAct: Synergizing Reasoning and Acting in Language Models | 2022 | link |
| 22 | QLoRA: Efficient Finetuning of Quantized LLMs | 2023 | link |
| 24 | Toolformer: Language Models Can Teach Themselves to Use Tools | 2023 | link |
| 25 | Tree of Thoughts: Deliberate Problem Solving with Large Language Models | 2023 | link |
| 34 | Meta Chain-of-Thought: Towards System 2 Reasoning in LLMs | 2025 | link |
| 35 | rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking | 2025 | link |
| 38 | GRPO: Group Relative Policy Optimization | 2024 | link |
| 39 | RLVR: Reinforcement Learning from Verifiable Rewards | 2024 | link |
| 45 | Speculative Decoding: Fast Inference from Transformers | 2022 | link |
| 50 | Scaling LLM Test-Time Compute: The Theoretical Foundation for Reasoning Models | 2024 | link |
| 51 | Let's Verify Step by Step: Process Reward Models | 2023 | link |
| 52 | PagedAttention: Efficient LLM Serving with vLLM | 2023 | link |
| 53 | Efficient Estimation of Word Representations in Vector Space (Word2Vec) | 2013 | link |
| 54 | RoFormer: Enhanced Transformer with Rotary Position Embedding (RoPE) | 2021 | link |
| 58 | Generative Agents: Interactive Simulacra of Human Behavior | 2023 | link |
| 59 | Model Context Protocol (MCP): An Open Standard for AI Tool Integration | 2024 | link |
| 60 | GraphRAG: From Local to Global - A Graph RAG Approach to Query-Focused Summarization | 2024 | link |
| 61 | AlphaGeometry: Solving Olympiad Geometry Without Human Demonstrations | 2024 | link |
| 62 | AlphaEvolve: A Gemini-Powered Coding Agent for Designing Advanced Algorithms | 2025 | link |
| 63 | Proximal Policy Optimization Algorithms (PPO) | 2017 | link |
| 68 | Highly Accurate Protein Structure Prediction with AlphaFold (AlphaFold 2) | 2021 | link |
| 76 | ZeRO and Megatron-LM: How Trillion-Parameter Models Are Actually Trained | 2019 | link |
| 77 | Self-Consistency Improves Chain of Thought Reasoning in Language Models | 2022 | link |
| 78 | Reflexion: Language Agents with Verbal Reinforcement Learning | 2023 | link |
| 79 | Self-Instruct: Aligning Language Models with Self-Generated Instructions | 2022 | link |
| 80 | FLAN: Finetuned Language Models Are Zero-Shot Learners (Instruction Tuning) | 2021 | link |
| 81 | Emergent Abilities of Large Language Models (and the Mirage Rebuttal) | 2022 | link |
| 82 | Sparse Autoencoders and Monosemanticity: Reading the Features Inside a Model | 2024 | link |
| 83 | Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training | 2024 | link |
| 84 | SWE-bench: Can Language Models Resolve Real-World GitHub Issues? | 2023 | link |
| 85 | Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena | 2023 | link |
| 86 | GPTQ and AWQ: Post-Training Quantization for Large Language Models | 2022 | link |
| 87 | Dense Passage Retrieval, ColBERT, and Sentence-BERT: The Retrieval Half of RAG | 2020 | link |
| 97 | STaR: Bootstrapping Reasoning With Reasoning (Self-Taught Reasoner) | 2022 | link |
| 98 | Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking | 2024 | link |
| 99 | Self-Refine: Iterative Refinement with Self-Feedback | 2023 | link |
| 100 | Voyager: An Open-Ended Embodied Agent with Large Language Models | 2023 | link |
| 101 | Accurate Structure Prediction of Biomolecular Interactions with AlphaFold 3 | 2024 | link |
| 102 | Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm | 2017 | link |
| 103 | KTO: Model Alignment as Prospect Theoretic Optimization | 2024 | link |
| 104 | Genie: Generative Interactive Environments | 2024 | link |
| 105 | Mastering Diverse Domains through World Models (DreamerV3) | 2023 | link |
| 106 | Evolutionary-Scale Prediction of Atomic-Level Protein Structure with a Language Model (ESM-2 / ESMFold) | 2023 | link |
| 107 | Human-Level Play in the Game of Diplomacy by Combining Language Models with Strategic Reasoning (CICERO) | 2022 | link |
| 126 | In-context Learning and Induction Heads (Induction Heads) | 2022 | link |
| 127 | Universal and Transferable Adversarial Attacks on Aligned Language Models (GCG) | 2023 | link |
| 128 | Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision (Weak-to-Strong) | 2023 | link |
| 129 | Red Teaming Language Models with Language Models (LM Red Teaming) | 2022 | link |
| 130 | YaRN: Efficient Context Window Extension of Large Language Models (YaRN) | 2023 | link |
| 132 | RULER: What's the Real Context Size of Your Long-Context Language Models? (RULER) | 2024 | link |
| 133 | The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale (FineWeb) | 2024 | link |
| 134 | Distilling the Knowledge in a Neural Network (Knowledge Distillation) | 2015 | link |
| 136 | Neural Machine Translation of Rare Words with Subword Units (BPE) | 2015 | link |
| 137 | Measuring Massive Multitask Language Understanding (MMLU) | 2020 | link |
| 138 | On the Measure of Intelligence (ARC) | 2019 | link |
| 139 | OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments (OSWorld) | 2024 | link |
| 140 | AI Models Collapse When Trained on Recursively Generated Data (Model Collapse) | 2024 | link |
| 142 | Adam: A Method for Stochastic Optimization (Adam) | 2014 | link |
| 143 | Mixed Precision Training | 2017 | link |
| 144 | Muon: An Optimizer for Hidden Layers, and Muon is Scalable for LLM Training (Muon) | 2024 | link |
| 145 | Measuring AI Ability to Complete Long Tasks (METR Time Horizons) | 2025 | link |
| 146 | Deep Reinforcement Learning from Human Preferences | 2017 | link |
| 147 | Learning to Summarize from Human Feedback | 2020 | link |
| 149 | Why Language Models Hallucinate | 2025 | link |
| 151 | Circuit Tracing and On the Biology of a Large Language Model (Attribution Graphs) | 2025 | link |
| 152 | Reasoning Models Don't Always Say What They Think (CoT Faithfulness) | 2025 | link |
| 153 | Emergent Misalignment: Narrow Finetuning Can Produce Broadly Misaligned LLMs | 2025 | link |
| 154 | s1: Simple Test-Time Scaling (s1) | 2025 | link |
| 155 | DAPO: An Open-Source LLM Reinforcement Learning System at Scale (DAPO) | 2025 | link |