XI'an

Huang Haiduo

Advanced
@haiduo

student of Xi'an Jiaotong University, @Xi'an Jiaotong University

PartialNet. This repository is the official implementation of "Partial Channel Network: Compute Fewer, Perform Better". [AAAI 2026 Accepted]

40

Jakiro. This repository is the official implementation of "Jakiro: Boosting Speculative Decoding with Decoupled Multi-Head via MoE" [ACL 2026 Main Accepted]

37

sparse_model. The four major frameworks for 3D point cloud sparse acceleration, which are currently mainstream, are compared. These include MIT-HAN-LAB's torchsparse, NVIDIA's MinkowskiEngine, TuSimple's spconv, and Facebook Research's SparseConvNet.

28

DeepKD. "DeepKD: A Deeply Decoupled and Denoised Knowledge Distillation Trainer" [NeurIPS 2025 Accepted]

19

YoloP_plus. The improved model for multi-object detection and lane line segmentation based on the YoloP model.

15

Double-Rounding. This repository is the official implementation of "Nearly Lossless Adaptive Bit Switching". [AAAI 2026 Accepted]

9

KernelDNA. This repository is the official implementation of "KernelDNA: Dynamic Kernel Sharing via Decoupled Naive Adapters"

8

SpecVLM. SpecVLM: Fast Speculative Decoding in Vision-Language Models

7

VITA_reproduction. A code reproduction of VITA.

2

transformers. 🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

1

hand-on-rl. Python

1
11
Apply