A sliding layer merging method that dynamically selects and fuses consecutive layers from top to bottom according to a pre-defined similarity threshold, thereby simplifying the model structure whil…

Official PyTorch implementation of the paper "Dataset Distillation with Neural Characteristic Function" (NCFM, Rating: 555) in CVPR 2025.

Chat with your database or your datalake (SQL, CSV, parquet). PandasAI makes data analysis conversational using LLMs and RAG.

Official implementation for KV-Edit: Training-Free Image Editing for Precise Background Preservation

[CVPR 2025] Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Models

A Conversational Speech Generation Model

SpargeAttention: A training-free sparse attention that can accelerate any model inference.

Production-tested AI infrastructure tools for efficient AGI development and community-driven innovation

Learning Flow Fields in Attention for Controllable Person Image Generation

Code Implementation of "PhotoDoodle: Learning Artistic Image Editing from Few-Shot Pairwise Data"

Official codebase for "SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution"

PyTorch implementation of FractalGen https://arxiv.org/abs/2502.17437

The python library for real-time communication

Official repo for ART:Anonymous Region Transformer for Variable Multi-Layer Transparent Image Generation

An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System

SIFT: Grounding LLM Reasoning in Contexts via Stickers

Toolkit for linearizing PDFs for LLM datasets/training

Towards Economical Inference: Enabling DeepSeek's Multi-Head Latent Attention in Any Transformer-based LLMs

FlashMLA: Efficient MLA decoding kernels

🔥 Rankify: A Comprehensive Python Toolkit for Retrieval, Re-Ranking, and Retrieval-Augmented Generation 🔥

OSUM: Open Speech Understanding Model, open-sourced by ASLP@NPU.

The official code of "Weak-to-Strong Diffusion with Reflection".

A curated collection of Python examples for optimization-based solid simulation, emphasizing algorithmic convergence, penetration-free, and inversion-free conditions, designed for readability and u…

Profile-Based Long-Term Memory for AI Applications

Knowledge Unlearning for Large Language Models