Research of DeepSeek Engram Architecture based on Qwen-3 and Stable Diffusion series.

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

[ICLR2026] SparseEval: Efficient Evaluation of Large Language Models by Sparse Optimization

OpenCompass is an LLM evaluation platform, supporting a wide range of models (Llama3, Mistral, InternLM2,GPT-4,LLaMa2, Qwen,GLM, Claude, etc) over 100+ datasets.

Advanced AI Explainability for computer vision. Support for CNNs, Vision Transformers, Classification, Object detection, Segmentation, Image similarity and more.

Official code of Attention-MoA: Enhancing Mixture-of-Agents via Inter-Agent Semantic Attention and Deep Residual Synthesis

Code of our AAAI 2025 oral paper titled "Filling Memory Gaps: Enhancing Continual Semantic Parsing via SQL Syntax Variance-Guided LLMs Without Real Data Replay"

Build Real-Time Knowledge Graphs for AI Agents

An interactive NVIDIA-GPU process viewer and beyond, the one-stop solution for GPU process management.

Self-evolving memory OS for LLM & AI Agents: ultra-persistent memory, hybrid-retrieval, and cross-task skill reuse, with 35.24% token savings

A modern GUI client based on Tauri, designed to run in Windows, macOS and Linux for tailored proxy experience

Visual testing tool for MCP servers

Code for the paper "Evaluating Large Language Models Trained on Code"

[ICML2024 (Oral)] Official PyTorch implementation of DoRA: Weight-Decomposed Low-Rank Adaptation

Fine-Tuning Dataset Auto-Generation for Graph Query Languages.

Retrieval and Retrieval-augmented LLMs

使深信服（Sangfor）开发的非自由的 VPN 软件 EasyConnect 和 aTrust 运行在 docker 或 podman 中，并作为网关和/或提供 socks5、http 代理服务

TrustRAG：The RAG Framework within Reliable input,Trusted output

提供多款 Shadowrocket 规则，拥有强劲的广告过滤功能。每日 8 时重新构建规则。

A MULTI-GENERATOR ENSEMBLE FRAMEWORK FOR NATURAL LANGUAGE TO SQL

This code implements a model for schema linking for text-to-SQL. Schema linking identifies the subset of the database schema that is needed to construct the SQL query to answer the user's informati…

在没有sudo权限的情况下，在linux上使用clash

Curated tutorials and resources for Large Language Models, Text2SQL, Text2DSL、Text2API、Text2Vis and more.

Official implementation of our NeurIPS 2023 paper "Augmenting Language Models with Long-Term Memory".

Adapt an LLM model to a Mixture-of-Experts model using Parameter Efficient finetuning (LoRA), injecting the LoRAs in the FFN.

A project page template for academic papers. Demo at https://eliahuhorwitz.github.io/Academic-project-page-template/

Code and data for "StructLM: Towards Building Generalist Models for Structured Knowledge Grounding" (COLM 2024)

东南大学校徽矢量图，转自微信公众号：金木屋