Some papers that have been of great help in my work, especially in the fields of ML and DL.

CL-bench: A Benchmark for Context Learning

Next-Token Prediction is All You Need

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

AlignX-Family is an open-source research suite for advancing personalization in large language models-spanning data, code, models, and beyond!

REFRAG-style RAG (compress → sense/select → expand) — Single-file reference implementation

Official implementation of GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization

搜索、推荐、广告、用增等工业界实践文章收集（来源：知乎、Datafuntalk、技术公众号）

Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models

Visual Causal Flow

An elegant PyTorch deep reinforcement learning library.

The absolute trainer to light up AI agents.

CC Workflow Studio

Use Claude Code as the foundation for coding infrastructure, allowing you to decide how to interact with the model while enjoying updates from Anthropic.

MiMo-V2-Flash: Efficient Reasoning, Coding, and Agentic Foundation Model

Kimi Code CLI is your next CLI agent.

Run Claude Code on OpenAI models

DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Code for the paper "Simple Context Compression: Mean-Pooling and Multi-Ratio Training"

DART-GUI: Efficient Multi-turn RL for GUI Agents via Decoupled Training and Adaptive Data Curation

The paper list of "Memory in the Age of AI Agents: A Survey"

This repository introduce a comprehensive paper list, datasets, methods and tools for memory research.

What's In My Big Data (WIMBD) - a toolkit for analyzing large text datasets

An Open Phone Agent Model & Framework. Unlocking the AI Phone for Everyone

Reproducible, flexible LLM evaluations

AllenAI's post-training codebase

Organize the Web: Constructing Domains Enhances Pre-Training Data Curation

Official repository of "CoMP: Continual Multimodal Pre-training for Vision Foundation Models"

RLinf: Reinforcement Learning Infrastructure for Embodied and Agentic AI

Ip2region is an offline IP address manager framework and locator with both IPv4 and IPv6 supported, supporting billions of data segments, ten microsecond searching performance, xdb search client fo…