Hong-Min Chu

7 papers · 2018–2024 · 4 conferences · across top CS/AI conferences

Achievements

🌍 Conference Polyglot (4) 🏃 Academic Marathon (6) 🌉 Interdisciplinary Bridge 🧭 Keyword Pioneer 🐝 Cross-Pollinator (15) 🚀 Conference Pioneer

Conferences

ICLR (4) ACML (1) ECCV (1) NIPS (1)

Top co-authors

Tom Goldstein (5) Jonas Geiping (4) Micah Goldblum (4) Arpit Bansal (2) Avi Schwarzschild (2) Ping-yeh Chiang (2) Jie Li (1) Eitan Borgnia (1) Bhavya Kailkhura (1) Chih-Kuan Yeh (1)

Keywords

image restoration (1) multi-label classification (1) deep learning (1) generative model (1) diffusion model (1) recurrent neural network (1) end-to-end training (1) label correlation (1) memory structure (1) deterministic degradation (1)

Papers

Universal Guidance for Diffusion Models ICLR 2024

NEFTune: Noisy Embeddings Improve Instruction Finetuning ICLR 2024

Panning for Gold in Federated Learning: Targeted Text Extraction under Arbitrarily Large-Scale Aggregation ICLR 2023

Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise NIPS 2023

WrapNet: Neural Net Inference with Ultra-Low-Precision Arithmetic ICLR 2021

Deep Learning with a Rethinking Structure for Multi-label Classification ACML 2019

Deep Generative Models for Weakly-Supervised Multi-Label Classification ECCV 2018