Alert button
Picture for Zhiyun Lu

Zhiyun Lu

Alert button

Direct Large Language Model Alignment Through Self-Rewarding Contrastive Prompt Distillation

Feb 19, 2024
Aiwei Liu, Haoping Bai, Zhiyun Lu, Xiang Kong, Simon Wang, Jiulong Shan, Meng Cao, Lijie Wen

Viaarxiv icon

Instruction-Following Speech Recognition

Sep 18, 2023
Cheng-I Jeff Lai, Zhiyun Lu, Liangliang Cao, Ruoming Pang

Figure 1 for Instruction-Following Speech Recognition
Figure 2 for Instruction-Following Speech Recognition
Figure 3 for Instruction-Following Speech Recognition
Figure 4 for Instruction-Following Speech Recognition
Viaarxiv icon

Less is More: Removing Text-regions Improves CLIP Training Efficiency and Robustness

May 08, 2023
Liangliang Cao, Bowen Zhang, Chen Chen, Yinfei Yang, Xianzhi Du, Wencong Zhang, Zhiyun Lu, Yantao Zheng

Figure 1 for Less is More: Removing Text-regions Improves CLIP Training Efficiency and Robustness
Figure 2 for Less is More: Removing Text-regions Improves CLIP Training Efficiency and Robustness
Figure 3 for Less is More: Removing Text-regions Improves CLIP Training Efficiency and Robustness
Figure 4 for Less is More: Removing Text-regions Improves CLIP Training Efficiency and Robustness
Viaarxiv icon

E2E Segmenter: Joint Segmenting and Decoding for Long-Form ASR

Apr 22, 2022
W. Ronny Huang, Shuo-yiin Chang, David Rybach, Rohit Prabhavalkar, Tara N. Sainath, Cyril Allauzen, Cal Peyser, Zhiyun Lu

Figure 1 for E2E Segmenter: Joint Segmenting and Decoding for Long-Form ASR
Figure 2 for E2E Segmenter: Joint Segmenting and Decoding for Long-Form ASR
Figure 3 for E2E Segmenter: Joint Segmenting and Decoding for Long-Form ASR
Figure 4 for E2E Segmenter: Joint Segmenting and Decoding for Long-Form ASR
Viaarxiv icon

Unsupervised Data Selection via Discrete Speech Representation for ASR

Apr 05, 2022
Zhiyun Lu, Yongqiang Wang, Yu Zhang, Wei Han, Zhehuai Chen, Parisa Haghani

Figure 1 for Unsupervised Data Selection via Discrete Speech Representation for ASR
Figure 2 for Unsupervised Data Selection via Discrete Speech Representation for ASR
Figure 3 for Unsupervised Data Selection via Discrete Speech Representation for ASR
Figure 4 for Unsupervised Data Selection via Discrete Speech Representation for ASR
Viaarxiv icon

Improving the fusion of acoustic and text representations in RNN-T

Jan 25, 2022
Chao Zhang, Bo Li, Zhiyun Lu, Tara N. Sainath, Shuo-yiin Chang

Figure 1 for Improving the fusion of acoustic and text representations in RNN-T
Figure 2 for Improving the fusion of acoustic and text representations in RNN-T
Figure 3 for Improving the fusion of acoustic and text representations in RNN-T
Figure 4 for Improving the fusion of acoustic and text representations in RNN-T
Viaarxiv icon

Input Length Matters: An Empirical Study Of RNN-T And MWER Training For Long-form Telephony Speech Recognition

Oct 08, 2021
Zhiyun Lu, Yanwei Pan, Thibault Doutre, Liangliang Cao, Rohit Prabhavalkar, Chao Zhang, Trevor Strohman

Figure 1 for Input Length Matters: An Empirical Study Of RNN-T And MWER Training For Long-form Telephony Speech Recognition
Figure 2 for Input Length Matters: An Empirical Study Of RNN-T And MWER Training For Long-form Telephony Speech Recognition
Figure 3 for Input Length Matters: An Empirical Study Of RNN-T And MWER Training For Long-form Telephony Speech Recognition
Figure 4 for Input Length Matters: An Empirical Study Of RNN-T And MWER Training For Long-form Telephony Speech Recognition
Viaarxiv icon

Exploring Targeted Universal Adversarial Perturbations to End-to-end ASR Models

Apr 06, 2021
Zhiyun Lu, Wei Han, Yu Zhang, Liangliang Cao

Figure 1 for Exploring Targeted Universal Adversarial Perturbations to End-to-end ASR Models
Figure 2 for Exploring Targeted Universal Adversarial Perturbations to End-to-end ASR Models
Figure 3 for Exploring Targeted Universal Adversarial Perturbations to End-to-end ASR Models
Figure 4 for Exploring Targeted Universal Adversarial Perturbations to End-to-end ASR Models
Viaarxiv icon

Improving Streaming Automatic Speech Recognition With Non-Streaming Model Distillation On Unsupervised Data

Oct 22, 2020
Thibault Doutre, Wei Han, Min Ma, Zhiyun Lu, Chung-Cheng Chiu, Ruoming Pang, Arun Narayanan, Ananya Misra, Yu Zhang, Liangliang Cao

Figure 1 for Improving Streaming Automatic Speech Recognition With Non-Streaming Model Distillation On Unsupervised Data
Figure 2 for Improving Streaming Automatic Speech Recognition With Non-Streaming Model Distillation On Unsupervised Data
Figure 3 for Improving Streaming Automatic Speech Recognition With Non-Streaming Model Distillation On Unsupervised Data
Figure 4 for Improving Streaming Automatic Speech Recognition With Non-Streaming Model Distillation On Unsupervised Data
Viaarxiv icon

Uncertainty Estimation with Infinitesimal Jackknife, Its Distribution and Mean-Field Approximation

Jun 13, 2020
Zhiyun Lu, Eugene Ie, Fei Sha

Figure 1 for Uncertainty Estimation with Infinitesimal Jackknife, Its Distribution and Mean-Field Approximation
Figure 2 for Uncertainty Estimation with Infinitesimal Jackknife, Its Distribution and Mean-Field Approximation
Figure 3 for Uncertainty Estimation with Infinitesimal Jackknife, Its Distribution and Mean-Field Approximation
Figure 4 for Uncertainty Estimation with Infinitesimal Jackknife, Its Distribution and Mean-Field Approximation
Viaarxiv icon