paper-is-all-you-need

论文就是你所需要的。

论文

OpenAI

论文	年份	论文单位	笔记地址
Scaling Laws for Neural Language Models	2020	OpenAI	./papers/00001-scaling-laws.pdf
Let's Verify Step by Step	2023	OpenAI	./papers/00028-Verify-Step-by-Step.pdf
Efficient Training of Language Models to Fill in the Middle	2022	OpenAI	./papers/00035-fim.pdf

DeepSeek-AI

论文	年份	论文单位	笔记地址
DeepSeek-V3 Technical Report	2024	DeepSeek-AI	./papers/00042-DeepSeek-V3.pdf
DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning	2025	DeepSeek-AI	./papers/00045-DeepSeek_R1.pdf
Inference-Time Scaling for Generalist Reward Modeling	2025	DeepSeek-AI	./papers/00056-DeepSeek-GRM.pdf
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models	2024	DeepSeek-AI	./papers/00057-GRPO.pdf

Microsoft

论文	年份	论文单位	笔记地址
ZeRO: Memory Optimizations Toward Training Trillion Parameter Models	2020	Microsoft	./papers/00002-ZeRO.pdf
Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone	2024	Microsoft	./papers/00004-Phi-3.pdf
DeepSpeed-Chat: Easy, Fast and Affordable RLHF Training of ChatGPT-like Models at All Scales	2023	Microsoft	./papers/00013-DeepSpeed-Chat.pdf
ZeroQuant(4+2): Redefining LLMs Quantization with a New FP6-Centric Strategy for Diverse Generative Tasks	2023	Microsoft	./papers/00014-ZeroQuant(4+2).pdf
DeepSpeed Inference: Enabling Efficient Inference of Transformer Models at Unprecedented Scale	2022	Microsoft	./papers/00015-DeepSpeed-Inference.pdf
DeepSpeed Ulysses: System Optimizations for Enabling Training of Extreme Long Sequence Transformer Models	2023	Microsoft	./papers/00016-DeepSpeed-Ulysses.pdf
ZeRO-Offload: Democratizing Billion-Scale Model Training	2021	Microsoft	./papers/00017-ZeRO-Offload.pdf
1-bit Adam: Communication Efficient Large-Scale Training with Adam's Convergence Speed	2021	Microsoft	./papers/00018-1-bit-Adam.pdf
ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep Learning	2021	Microsoft	./papers/00019-ZeRO-Infinity.pdf
1-bit LAMB: Communication Efficient Large-Scale Large-Batch Training with LAMB's Convergence Speed	2021	Microsoft	./papers/00020-1-bit-LAMB.pdf
The Stability-Efficiency Dilemma: Investigating Sequence Length Warmup for Training GPT Models	2022	Microsoft	./papers/00021-Sequence-Length-Warmup.pdf
Maximizing Communication Efficiency for Large-scale Training via 0/1 Adam	2022	Microsoft	./papers/00023-0-1-Adam.pdf
DeepSpeed-MoE: Advancing Mixture-of-Experts Inference and Training to Power Next-Generation AI Scale	2022	Microsoft	./papers/00023-DeepSpeed-MoE.pdf
Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Model	2022	Microsoft	./papers/00024-Megatron-Turing.pdf
Extreme Compression for Pre-trained Transformers Made Simple and Efficient	2022	Microsoft	./papers/00025-XTC.pdf
ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers	2022	Microsoft	./papers/00026-ZeroQuant.pdf
DeepSpeed Data Efficiency: Improving Deep Learning Model Quality and Training Efficiency via Efficient Data Sampling and Routing	2024	Microsoft	./papers/00027-DeepSpeed-Data-Efficiency.pdf
Understanding INT4 Quantization for Transformer Models: Latency Speedup, Composability, and Failure Cases	2023	Microsoft	./papers/00029-INT4-Quantization.pdf
ZeroQuant-FP: A Leap Forward in LLMs Post-Training W4A8 Quantization Using Floating-Point Formats	2023	Microsoft	./papers/00030-ZeroQuant-FP.pdf
ZeRO++: Extremely Efficient Collective Communication for Giant Model Training	2023	Microsoft	./papers/00031-ZeRO++.pdf
ZeroQuant-HERO: Hardware-Enhanced Robust Optimized Post-Training Quantization Framework for W8A8 Transformers	2023	Microsoft	./papers/00032-ZeroQuant-HERO.pdf
rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking	2025	Microsoft	./papers/00044-rStar-Math.pdf
From Local to Global: A Graph RAG Approach to Query-Focused Summarization	2024	Microsoft	./papers/00046-GraphRAG.pdf
Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning	2025	Microsoft	./papers/00050-Logic-RL.pdf
LongRoPE2: Near-Lossless LLM Context Window Scaling	2025	Microsoft	./papers/00063-LongRoPE2.pdf
BitNet b1.58 2B4T Technical Report	2025	Microsoft	./papers/00064-BitNet.pdf
1-bit AI Infra: Part 1.1, Fast and Lossless BitNet b1.58 Inference on CPUs	2024	Microsoft	./papers/00065-1-bit.pdf
Bitnet.cpp: Efficient Edge Inference for Ternary LLMs	2025	Microsoft	./papers/00066-Bitnet.cpp.pdf
MiniLLM: Knowledge Distillation of Large Language Models	2023	Microsoft	./papers/00075-MiniLLM.pdf

Google

论文	年份	论文单位	笔记地址
Chain-of-Thought Prompting Elicits Reasoning in Large Language Models	2022	Google	./papers/00008-CoT.pdf
Distilling the Knowledge in a Neural Network	2015	Google	./papers/00076-distill-model.pdf
STaR: Bootstrapping Reasoning With Reasoning	2022	Google	./papers/00077-STaR.pdf
ReAct: Synergizing Reasoning and Acting in Language Models	2022	Google	./papers/00078-ReAct.pdf
Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities	2025	Google	./papers/00080-Gemini-2.5.pdf
Gemma 3 Technical Report	2025	Google	./papers/00081-Gemma-3.pdf

Meta

论文	年份	论文单位	笔记地址
The Llama 3 Herd of Models	2024	Meta	./papers/00006-Llama3.pdf

DeepMind

论文	年份	论文单位	笔记地址
Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters	2024	DeepMind	./papers/00036-Scaling-LLM-Test-Time-Compute-Optimally.pdf
Improve Mathematical Reasoning in Language Models by Automated Process Supervision	2024	DeepMind	./papers/00039-Automated-Process-Supervision.pdf

NVIDIA

论文	年份	论文单位	笔记地址
Reducing Activation Recomputation in Large Transformer Models	2022	NVIDIA	./papers/00012-selective-activation-recomputation.pdf

Alibaba Group

论文	年份	论文单位	笔记地址
AlphaMath Almost Zero: process Supervision without process	2024	Alibaba Group	./papers/00007-AlphaMath.pdf
Qwen2.5-Coder Technical Report	2024	Alibaba Group	./papers/00033-Qwen2.5-Coder.pdf
Qwen3 Technical Report	2025	Alibaba Group	./papers/00069-Qwen3_Technical_Report.pdf
Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models	2025	Alibaba Group	./papers/00073-Qwen3-Embedding.pdf
WebSailor: Navigating Super-human Reasoning for Web Agent	2025	Alibaba Group	./papers/00082-WebSailor.pdf

ByteDance

论文	年份	论文单位	笔记地址
HybridFlow: A Flexible and Efficient RLHF Framework	2024	ByteDance	./papers/00054-HybridFlow.pdf
DAPO: An Open-Source LLM Reinforcement Learning System at Scale	2025	ByteDance	./papers/00055-DAPO.pdf
VAPO: Efficient and Reliable Reinforcement Learning for Advanced Reasoning Tasks	2025	ByteDance	./papers/00067-VAPO.pdf
Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learning	2025	ByteDance	./papers/00068-Seed1.5-Thinking.pdf
Seed-Coder: Let the Code Model Curate Data for Itself	2025	ByteDance	./papers/00070-Seed-Coder.pdf

KimiTeam

论文	年份	论文单位	笔记地址
Kimi K2: Open Agentic Intelligence	2025	KimiTeam	./papers/00083-Kimi-K2.pdf

EleutherAI

论文	年份	论文单位	笔记地址
YaRN: Efficient Context Window Extension of Large Language Models	2023	EleutherAI	./papers/00034-YaRN.pdf

Tsinghua University

论文	年份	论文单位	笔记地址
MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies	2024	Tsinghua University	./papers/00003-MiniCPM.pdf
ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools	2024	Tsinghua University	./papers/00005-ChatGLM.pdf
LongWriter: Unleashing 10,000+ Word Generation from Long Context LLMs	2024	Tsinghua University	./papers/00038-LongWriter.pdf
ReST-MCTS*: LLM Self-Training via Process Reward Guided Tree Search	2024	Tsinghua University	./papers/00043-ReST-MCTS.pdf

Peking University

论文	年份	论文单位	笔记地址
Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations	2024	Peking University	./papers/00040-Math-Shepherd.pdf
ChartMoE: Mixture of Diversely Aligned Expert Connector for Chart Understanding	2024	Peking University	./papers/00049-ChartMoE.pdf

Carnegie Mellon University

论文	年份	论文单位	笔记地址
Optimizing Test-Time Compute via Meta Reinforcement Fine-Tuning	2025	Carnegie Mellon University	./papers/00053-MRT.pdf

National University of Singapore

论文	年份	论文单位	笔记地址
Understanding R1-Zero-Like Training: A Critical Perspective	2025	National University of Singapore	./papers/00048-understand-r1-zero.pdf

Shanghai Jiao Tong University

论文	年份	论文单位	笔记地址
LIMO: Less is More for Reasoning	2025	Shanghai Jiao Tong University	./papers/00051-LIMO.pdf

Nanjing University

论文	年份	论文单位	笔记地址
A Comprehensive Survey on Long Context Language Modeling	2025	Nanjing University	./papers/00058-LCLM.pdf

Renmin University of China

论文	年份	论文单位	笔记地址
Imitate, Explore, and Self-Improve: A Reproduction Report on Slow-thinking Reasoning Systems	2024	Renmin University of China	./papers/00037-o1-like.pdf
Search-o1: Agentic Search-Enhanced Large Reasoning Models	2025	Renmin University of China	./papers/00059-Search-o1.pdf
WebThinker: Empowering Large Reasoning Models with Deep Research Capability	2025	Renmin University of China	./papers/00074-WebThinker.pdf
R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning	2025	Renmin University of China	./papers/00062-R1-Searcher.pdf

University of Illinois Urbana-Champaign

论文	年份	论文单位	笔记地址
DeepRetrieval: Hacking Real Search Engines and Retrievers with Large Language Models via Reinforcement Learning	2025	University of Illinois Urbana-Champaign	./papers/00060-DeepRetrieval.pdf
Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning	2025	University of Illinois Urbana-Champaign	./papers/00061-Search-R1.pdf

其他

论文	年份	论文单位	笔记地址
LightRAG: Simple and Fast Retrieval-Augmented Generation	2024	Beijing University of Posts and Telecommunications	./papers/00047-LightRAG.pdf
HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models	2024	The Ohio State University	./papers/00071-HippoRAG.pdf
From RAG to Memory: Non-Parametric Continual Learning for Large Language Models	2025	The Ohio State University	./papers/00072-HippoRAG2.pdf
Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model	2024	University of Waterloo	./papers/00011-Chinese-Tiny-LLM.pdf
OpenR: An Open Source Framework for Advanced Reasoning with Large Language Models	2024	University College London	./papers/00041-OpenR.pdf
Retrieval-Augmented Generation with Graphs (GraphRAG)	2025	Michigan State University	./papers/00052-GraphRAG-review.pdf
RoFormer: Enhanced Transformer with Rotary Position Embedding	2021	Zhuiyi Technology	./papers/00009-RoPE.pdf
Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions	2022	Allen Institute for AI	./papers/00079-IRCoT.pdf
	2025

工具

1. Overleaf: https://www.overleaf.com/

论文集

1. huggingface daily papers
2. arxiv