CLJun 11, 2024

On the Hallucination in Simultaneous Machine Translation

arXiv:2406.07239v127 citations
Originality Synthesis-oriented
AI Analysis

This addresses a critical issue for SiMT systems, but the approach is incremental as it focuses on analysis rather than a new method.

The paper tackles the problem of hallucination in Simultaneous Machine Translation (SiMT) by analyzing its distribution and target-side context usage, finding that reducing over-reliance on target-side information can alleviate it.

It is widely known that hallucination is a critical issue in Simultaneous Machine Translation (SiMT) due to the absence of source-side information. While many efforts have been made to enhance performance for SiMT, few of them attempt to understand and analyze hallucination in SiMT. Therefore, we conduct a comprehensive analysis of hallucination in SiMT from two perspectives: understanding the distribution of hallucination words and the target-side context usage of them. Intensive experiments demonstrate some valuable findings and particularly show that it is possible to alleviate hallucination by decreasing the over usage of target-side information for SiMT.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes