Stars
This repository is for the "LLM-Aligned Geographic Item Tokenization for Local-Life Recommendation".
Algorithm powering the For You feed on X
Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models
[Pytorch] The repo contains the code for "FORGE: Forming Semantic Identifiers for Generative Retrieval in Industrial Datasets"
Memory layers use a trainable key-value lookup mechanism to add extra parameters to a model without increasing FLOPs. Conceptually, sparsely activated memory layers complement compute-heavy dense f…
Awesome AI Memory | LLM Memory | A curated knowledge base on AI memory for LLMs and agents, covering long-term memory, reasoning, retrieval, and memory-native system design. Awesome-AI-Memory 是一个 集…
[ICLR 2026] LightMem: Lightweight and Efficient Memory-Augmented Generation
Unofficial implementation of Titans, SOTA memory for transformers, in Pytorch
😊 TPTT: Transforming Pretrained Transformers into Titans
Reproduced the DFT method without using Verl. https://arxiv.org/abs/2508.05629
[ICLR 2026] The official implementation of the paper “Anchored Supervised Fine-Tuning”
A Reproduction of GDM's Nested Learning Paper
[ICLR 2025] MiniPLM: Knowledge Distillation for Pre-Training Language Models
[NeurIPS 2025 Spotlight] TPA: Tensor ProducT ATTenTion Transformer (T6) (https://arxiv.org/abs/2501.06425)
🚀🚀 「大模型」2小时完全从0训练64M的小参数GPT！🌏 Train a 64M-parameter GPT from scratch in just 2h!
Phi2-Chinese-0.2B 从0开始训练自己的Phi2中文小模型，支持接入langchain加载本地知识库做检索增强生成RAG。Training your own Phi2 small chat model from scratch.
Text-to-Video generation model using a Hierarchical Reasoning Model (HRM) optimized for T4 GPUs.
Hierarchical Reasoning Model Official Release
Python Implementation of MUVERA (Multi-Vector Retrieval via Fixed Dimensional Encodings)
BERT-based intent and slots detector for chatbots.