Shanghai, China

Hm Xiong

Advanced
@hmxiong

My research interests lie in the field of MLLM, Agents, LLM Reasoning and Inference Optimization.

StreamChat. Official repo for "Streaming Video Understanding and Multi-round Interaction with Memory-enhanced Knowledge" ICLR2025

111

3UR-LLM. Official repo for "3UR-LLM: An End-to-End Multimodal Large Language Model for 3D Scene Understanding" TMM2025

13

Transformer-Series. Python

7

OpenMMLabCamp. Python

3

PathWeave. Code for paper "LLMs Can Evolve Continually on Modality for X-Modal Reasoning" NeurIPS2024

3

CUDA-Learn-Note. 🎉CUDA 笔记 / 大模型手撕CUDA / C++笔记，更新随缘: flash_attn、sgemm、sgemv、warp reduce、block reduce、dot product、elementwise、softmax、layernorm、rmsnorm、hist etc.

2

paper-reading. 深度学习经典、新论文逐段精读

2

pytorch-distributed-training. Simple tutorials on Pytorch DDP training

1

llama2.c. Inference Llama 2 in one file of pure C

1

ScanNet_Vis. Python

1

CRATE. Code for CRATE (Coding RAte reduction TransformEr).

1

RWKV-LM. RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.

1

GaLore. Python

1

VILA. VILA - a multi-image visual language model with training, inference and evaluation recipe, deployable from cloud to edge (Jetson Orin and laptops)

1

Awesome-LLM-Inference. 📖A curated list of Awesome LLM/VLM Inference Papers with codes: WINT8/4, Flash-Attention, Paged-Attention, Parallelism, etc. 🎉🎉

1

github-slideshow. A robot powered training repository :robot:

1

Tarurs. competition files

1
17
Apply