Huck Yang
@huckiyangSr. Research Scientist @NVIDIAAI Generative Voice Correction | Ph.D. MSc @GeorgiaTech | Past: @GoogleAI @AmazonScience | 🗣️ education
Similar User
@mhnt1580
@YungSungChuang
@AcouIntel
@RoshanSSharma2
@jefflai108
@yhtu_
@alex_h_liu
@Sid_Arora_18
@jiatongshi
@GTL094144
@vinsdotton
@RuiLiu60711141
@iamyuanchung
@fishball_Lin
@jack978397
Excited to share our new preprint, "CLaSP: Learning Concepts for Time-Series Signals from Natural Language Supervision"! Our model enables query-by-text/signal retrieval and zero-shot classification for time series data. Check it out: arxiv.org/abs/2411.08397
🚨 Adaptive Decoding via Latent Preference Optimization 🚨 - New layer added to Transformer, selects decoding params automatically *per token* - Learnt via new method, Latent Preference Optimization - Outperforms any fixed temperature decoding method, choosing creativity or…
60% of the pull requests (and probably more than 60% of the code) for our github issue resolver were written by coding agents in November, and the number keeps going up. Amazing times.
📈So far, in November, OpenHands authored or co-authored ~60% of commits to the openhands-resolver github.com/All-Hands-AI/o… repo and ~20% of the main repo github.com/All-Hands-AI/O…
🚀 FastAdaSP on the speech-text token merging at SoTA SLMs will be presented at the oral session today at @emnlpmeeting by Eason and Jiaqi from @WavLab @shinjiw_at_cmu @LTIatCMU @NVIDIAAI ⏰Time: 10:50 AM - 11:00 AM, Wed, Nov 13, 2024 📍Location: Tuttle, Lower Terrace Level,…
☄️Excited to share our #SLT2024 paper "Mamba-based Decoder-Only Approach with Bidirectional Speech Modeling for Speech Recognition"! Code: github.com/YoshikiMas/mad… arXiv: arxiv.org/abs/2411.06968 Mamba and Mamba-2 perform pretty well even completely without explicit attention😎
"LLM agents for Acoustics and Continuous signals" @emnlpmeeting 2024 gather session co-hosted with @PiotrZelasko and @_dmchan will be held in Tuesday Nov 12th Room Foster, Level 2. Feel free to join with us and open to all! 🗣️ - From Continuous Signals Modeling to Audio…
What makes an agent a "conversational" agent? Introducing ReSpAct, reason, speak, act for building conversational agents. LLM based agent decides on when to interact with the user (or user agent) when it is needed, such as disambiguation, clarification, or simply stuck at an…
Ever felt that agents that “just follow orders” often fall short, especially when critical details are on the line? In our latest work, we introduce ReSpAct—an approach that transforms agents into true conversational partners🌐 vardhandongre.github.io/respact-llm/ 🧵 [1/n]
It's a beautiful demonstration of the powers of GPT-4o, and also it's limitations. GPT-4o gives you a salad of claims from the RL literature, some written by over-hyped authors, at the end of which you are not sure if RL can really reason counterfactually (ie, at level 3), or…
Why does RL lead to causal understanding? 🧵🪡 GPT-4o: Reinforcement learning (RL) can lead to causal understanding because, by interacting with an environment, an agent learns not just correlations between actions and outcomes, but also the underlying cause-effect…
👀 Exciting advancements in Automated Audio Captioning.✨ The CMU-NVIDIA team has unveiled a groundbreaking approach that leverages multi-agent collaboration and GPU technology to enhance audio-to-text systems. 🔍 Key innovations include: ✅ Multi-encoder fusion: Combining…
Time series with descriptive Texts! Very nice works by @yohekawag and @Hitachi
Happy to announce our latest preprint: "Domain-Independent Text Generation for Time-Series Data"! Our method can generate descriptive texts from sensor-captured time-series signals, regardless of the domain. Check it out here: arxiv.org/abs/2409.16647
👏👏Check out the 2nd best-performing system in our Generative Error Correction Challenge for LLM-based emotion recognition (yet unreasonably rejected by SLT2024). They achieved 75.1% accuracy on ASR transcriptions of IEMOCAP using GPT-4o.
``Context and System Fusion in Post-ASR Emotion Recognition with Large Language Models,'' Pavel Stepachev, Pinzhen Chen, Barry Haddow, ift.tt/tCG7YmT
Now HEAR this (not just watch) - We've got audio covered for generated videos 🔊 Introducing Movie Gen Audio, which adds 48kHz synced SFX and aligned music to amazing videos from Movie Gen Video (and other sources!) Super honored to work with this amazing team! More to come🔥🔥
🎥 Today we’re premiering Meta Movie Gen: the most advanced media foundation models to-date. Developed by AI research teams at Meta, Movie Gen delivers state-of-the-art results across a range of capabilities. We’re excited for the potential of this line of research to usher in…
僕はよくアメリカの大学システムを長良川の鵜飼に例えるんですが、いらすとやに鵜飼の絵があってビックリ。ちなみに教員（僕）は鵜飼じゃなくて鵜の方です。
My answer: Alec Radford. There were good suggestions below, but to my mind @AlecRad is clearly the person with the largest influence, yet the least recognition. He's been the driver so many amazing developments and should be in the history books as a (the?) father of modern…
Who's the most important, yet most under-recognized AI scientist in the world? In my view, there's only one clear, right answer. One person behind most of the major advances, yet rarely mentioned or celebrated. Who do you think I mean? I'll reveal my answer after hearing your…
United States Trends
- 1. #Arcane 206 B posts
- 2. Jake Paul 1,02 Mn posts
- 3. Jayce 43,8 B posts
- 4. #SaturdayVibes 2.591 posts
- 5. Serrano 246 B posts
- 6. Vander 13,9 B posts
- 7. Good Saturday 23 B posts
- 8. #HappySpecialStage 83,3 B posts
- 9. maddie 18,5 B posts
- 10. #saturdaymorning 1.845 posts
- 11. Jinx 101 B posts
- 12. #SaturdayMood 1.222 posts
- 13. Isha 33,2 B posts
- 14. Canelo 17,6 B posts
- 15. Father Time 10,8 B posts
- 16. Rizwan 7.931 posts
- 17. Woop Woop 1.242 posts
- 18. The Astronaut 29,5 B posts
- 19. Super Tuna 21,8 B posts
- 20. Babar 11,3 B posts
Who to follow
-
Wei-Ning Hsu
@mhnt1580 -
Yung-Sung Chuang @ EMNLP2024
@YungSungChuang -
Anurag Kumar
@AcouIntel -
Roshan Sharma
@RoshanSSharma2 -
Cheng-I Lai
@jefflai108 -
YH
@yhtu_ -
Alexander H. Liu
@alex_h_liu -
Siddhant Arora
@Sid_Arora_18 -
jiatongshi
@jiatongshi -
Guan-Ting (Daniel) Lin
@GTL094144 -
vins.ton
@vinsdotton -
Rui Liu
@RuiLiu60711141 -
Yu-An (Andy) Chung
@iamyuanchung -
Fishball Lin🔮
@fishball_Lin -
RF yao
@jack978397
Something went wrong.
Something went wrong.