DPaI: Differentiable Pruning at Initialization with Node-Path Balance Principle

Li Xiang, Quan Nguyen-Tri, Lan-Cuong Nguyen, Hoang Pham, Khoat Than +2 more
2/12/2026

No official code found. Best implementation: QuanNguyen-Tri/DPaI (1 star, last commit 2025-05-13).

Semantic Scholar

Code Implementations(1)

QuanNguyen-Tri/DPaICommunity68%

Source code for "DPaI: Differentiable Pruning at Initialization with Node-Path Balance Principle"

10Shell, PythonFeb 28, 20251 years ago

Similar Papers

SafeDICE: Offline Safe Imitation Learning with Non-Preferred Demonstrations

90% match2026

Weaker MVI Condition: Extragradient Methods with Multi-Step Exploration

90% match2026

TrajPrompt: Aligning Color Trajectory with Vision-Language Representations

90% match2026

HT-Step: Aligning Instructional Articles with How-To Videos

90% match2026

BenTo: Benchmark Reduction with In-Context Transferability

90% match2026

FedVAD: Enhancing Federated Video Anomaly Detection with GPT-Driven Semantic Distillation

90% match1 repo162026

Cite this paper

@article{xiang2026dpai,
  title  = {DPaI: Differentiable Pruning at Initialization with Node-Path Balance Principle},
  author = {Li Xiang and Quan Nguyen-Tri and Lan-Cuong Nguyen and Hoang Pham and Khoat Than and Long Tran-Thanh and Hongkai Wen},
  year   = {2026},
  url    = {https://api.semanticscholar.org/CorpusID:4d485ce354a57197733f584b038193afa6a473f1},
  journal = {ICLR 2025 2025}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/4d485ce354a57197733f584b038193afa6a473f1.svg)](https://codeofpaper.com/paper/4d485ce354a57197733f584b038193afa6a473f1)

Discussion