@HsingHuan Profile picture

Hsing-Huan Chung

@HsingHuan

PhD Student @ UT Austin

Similar User
Theresia Veronika Rampisela photo

@theresia_v_r

Xidulu photo

@xidulu

Giandomenico Cornacchia photo

@GiandomenicoC17

Wenjin Yan photo

@WenjinOR

Chung-Wei Lee photo

@Chung_Wei_

Bharath Sharma photo

@BharathSharma93

Laurent Hoss photo

@laurentho77

SaritaOlovsky photo

@OlovskySarita

Ale𝕏 Joy  photo

@iAlexJoy

SepeKa photo

@Sepehr_Kamahi

Multimodal RecSys and Search photo

@MM_InfoSeeking

Alessandro B. Melchiorre photo

@ale_melchiorre

Major Tom photo

@MajorTomX1

Peter Müllner photo

@mullner_peter

Hsing-Huan Chung Reposted

google Japan is cooking in a way the West couldnt dream of


Hsing-Huan Chung Reposted

🎉At #ICASSP2024, I published a paper showing that synthetic data can effectively improve ASR models (arxiv.org/abs/2309.10707). At #EMNLP2024, we're presenting SYN2REAL to address synthetic-to-real gap in ASR! 🔗Paper: arxiv.org/abs/2406.02925 💡Project: farnhua.github.io/syn2real.githu…

Exploring task vectors: Not just for text LLMs learning new languages (arxiv.org/abs/2310.04799), but also helpful for speech models. Train with domain-specific synthetic data, then adapt using a task vector for real speech (arxiv.org/abs/2406.02925).

Tweet Image 1


Hsing-Huan Chung Reposted

Diffusion based image editing and personalization methods are expensive💰due to training, latent optimization or prompt-tuning🤷‍♂️. Introducing RF-Inversion🎯,the first efficient zero-shot inversion and editing framework for Flux🚀without training,optimization or prompt-tuning🧵⬇️

Tweet Image 1

Hsing-Huan Chung Reposted

no one did it like she does

Tweet Image 1
Tweet Image 2
Tweet Image 3
Tweet Image 4

Hsing-Huan Chung Reposted

10. Optimus Party Mode


Hsing-Huan Chung Reposted
Tweet Image 1

Hsing-Huan Chung Reposted

A monk climbing with bare feet on the cliff that tourists climb on with the help of a rope


Hsing-Huan Chung Reposted

(1/n) Training LLMs can be hindered by out-of-memory, scaling batch size, and seq length. Add one line to boost multi-GPU training throughput by 20% and reduce memory usage by 60%. Introducing Liger-Kernel: Efficient Triton Kernels for LLM Training. github.com/linkedin/Liger…

Tweet Image 1

Hsing-Huan Chung Reposted

The ocean is shrouded in mystery. It covers about 71% of the Earth’s surface, yet we know very little about it when you consider how vast it is - A Thread 1. This fisherman with a deep sea Wolf Eel

Tweet Image 1

Hsing-Huan Chung Reposted

*Long Thread on my Academic Job Market Experience* In 10 days I start my new tenure-track position, a position that took me 5 years on the academic job market to get I wanna talk about my (miserable) experience on the market, and a few takeaways for those on it now


Loading...

Something went wrong.


Something went wrong.