May 7, 2024, 4:50 a.m. | Zhihao Wang, Longyue Wang, Jinsong Su, Junfeng Yao, Zhaopeng Tu

cs.CL updates on arXiv.org arxiv.org

arXiv:2405.02673v1 Announce Type: new
Abstract: Token repetition is a typical form of multi-modal problem in fully non-autoregressive translation (NAT). In this work, we revisit the multi-modal problem in recently proposed NAT models. Our study reveals that these advanced models have introduced other types of information redundancy errors, which cannot be measured by the conventional metric - the continuous repetition ratio. By manually annotating the NAT outputs, we identify two types of information redundancy errors that correspond well to lexical and …

abstract advanced arxiv autoregressive cs.cl errors form information modal multi-modal nat redundancy study the information token translation type types work

Software Engineer for AI Training Data (School Specific)

@ G2i Inc | Remote

Software Engineer for AI Training Data (Python)

@ G2i Inc | Remote

Software Engineer for AI Training Data (Tier 2)

@ G2i Inc | Remote

Data Engineer

@ Lemon.io | Remote: Europe, LATAM, Canada, UK, Asia, Oceania

Artificial Intelligence – Bioinformatic Expert

@ University of Texas Medical Branch | Galveston, TX

Lead Developer (AI)

@ Cere Network | San Francisco, US