中文
相关论文

相关论文: CBSiMT: Mitigating Hallucination in Simultaneous M…

200 篇论文

Simultaneous machine translation (SiMT) outputs translation while reading the source sentence. Unlike conventional sequence-to-sequence (seq2seq) training, existing SiMT methods adopt the prefix-to-prefix (prefix2prefix) training, where the…

计算与语言 · 计算机科学 2024-05-29 Shoutao Guo , Shaolei Zhang , Yang Feng

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…

计算与语言 · 计算机科学 2024-06-12 Meizhi Zhong , Kehai Chen , Zhengshan Xue , Lemao Liu , Mingming Yang , Min Zhang

Simultaneous machine translation (SimulMT) speeds up the translation process by starting to translate before the source sentence is completely available. It is difficult due to limited context and word order difference between languages.…

计算与语言 · 计算机科学 2022-05-05 Chih-Chiang Chang , Shun-Po Chuang , Hung-yi Lee

Simultaneous machine translation (SiMT) starts translating while receiving the streaming source inputs, and hence the source sentence is always incomplete during translating. Different from the full-sentence MT using the conventional…

计算与语言 · 计算机科学 2022-03-24 Shaolei Zhang , Yang Feng

Machine Translation (MT) is undergoing a paradigm shift, with systems based on fine-tuned large language models (LLM) becoming increasingly competitive with traditional encoder-decoder models trained specifically for translation tasks.…

计算与语言 · 计算机科学 2025-01-30 Zilu Tang , Rajen Chatterjee , Sarthak Garg

The Neural Machine Translation (NMT) model is essentially a joint language model conditioned on both the source sentence and partial translation. Therefore, the NMT model naturally involves the mechanism of the Language Model (LM) that…

计算与语言 · 计算机科学 2021-06-01 Mengqi Miao , Fandong Meng , Yijin Liu , Xiao-Hua Zhou , Jie Zhou

Simultaneous machine translation (SimulMT) models start translation before the end of the source sentence, making the translation monotonically aligned with the source sentence. However, the general full-sentence translation test set is…

计算与语言 · 计算机科学 2023-03-14 Mengge Liu , Wen Zhang , Xiang Li , Jian Luan , Bin Wang , Yuhang Guo , Shuoying Chen

The primary objective of simultaneous machine translation (SiMT) is to minimize latency while preserving the quality of the final translation. Drawing inspiration from CPU branch prediction techniques, we propose incorporating branch…

计算与语言 · 计算机科学 2023-12-25 Aoxiong Yin , Tianyun Zhong , Haoyuan Li , Siliang Tang , Zhou Zhao

Neural conditional language generation models achieve the state-of-the-art in Neural Machine Translation (NMT) but are highly dependent on the quality of parallel training dataset. When trained on low-quality datasets, these models are…

计算与语言 · 计算机科学 2023-06-16 Joël Tang , Marina Fomicheva , Lucia Specia

Simultaneous machine translation (SiMT) generates translation while reading the whole source sentence. However, existing SiMT models are typically trained using the same reference disregarding the varying amounts of available source…

计算与语言 · 计算机科学 2023-10-27 Shoutao Guo , Shaolei Zhang , Yang Feng

Despite the success of neural machine translation (NMT), simultaneous neural machine translation (SNMT), the task of translating in real time before a full sentence has been observed, remains challenging due to the syntactic structure…

计算与语言 · 计算机科学 2020-10-06 Yun Chen , Liangyou Li , Xin Jiang , Xiao Chen , Qun Liu

Simultaneous machine translation (SiMT) generates translation before reading the entire source sentence and hence it has to trade off between translation quality and latency. To fulfill the requirements of different translation quality and…

计算与语言 · 计算机科学 2022-03-22 Shaolei Zhang , Yang Feng

Simultaneous Machine Translation (SiMT) aims to yield a real-time partial translation with a monotonically growing the source-side context. However, there is a counterintuitive phenomenon about the context usage between training and…

计算与语言 · 计算机科学 2023-11-14 Meizhi Zhong , Lemao Liu , Kehai Chen , Mingming Yang , Min Zhang

Neural sequence models can generate highly fluent sentences, but recent studies have also shown that they are also prone to hallucinate additional content not supported by the input. These variety of fluent but wrong outputs are…

计算与语言 · 计算机科学 2021-06-04 Chunting Zhou , Graham Neubig , Jiatao Gu , Mona Diab , Paco Guzman , Luke Zettlemoyer , Marjan Ghazvininejad

Neural Machine Translation (NMT) generates target words sequentially in the way of predicting the next word conditioned on the context words. At training time, it predicts with the ground truth words as context while at inference it has to…

计算与语言 · 计算机科学 2019-06-18 Wen Zhang , Yang Feng , Fandong Meng , Di You , Qun Liu

Most dominant neural machine translation (NMT) models are restricted to make predictions only according to the local context of preceding words in a left-to-right manner. Although many previous studies try to incorporate global information…

计算与语言 · 计算机科学 2022-03-18 Chulun Zhou , Fandong Meng , Jie Zhou , Min Zhang , Hongji Wang , Jinsong Su

Token-level adaptive training approaches can alleviate the token imbalance problem and thus improve neural machine translation, through re-weighting the losses of different target tokens based on specific statistical metrics (e.g., token…

计算与语言 · 计算机科学 2022-03-08 Songming Zhang , Yijin Liu , Fandong Meng , Yufeng Chen , Jinan Xu , Jian Liu , Jie Zhou

In simultaneous translation (SimulMT), the most widely used strategy is the wait-k policy thanks to its simplicity and effectiveness in balancing translation quality and latency. However, wait-k suffers from two major limitations: (a) it is…

计算与语言 · 计算机科学 2022-04-28 Guangxu Xun , Mingbo Ma , Yuchen Bian , Xingyu Cai , Jiaji Huang , Renjie Zheng , Junkun Chen , Jiahong Yuan , Kenneth Church , Liang Huang

The problem of hallucination and omission, a long-standing problem in machine translation (MT), is more pronounced when a large language model (LLM) is used in MT because an LLM itself is susceptible to these phenomena. In this work, we…

计算与语言 · 计算机科学 2024-11-22 Qiyu Wu , Masaaki Nagata , Zhongtao Miao , Yoshimasa Tsuruoka

Large Language Models (LLMs) have advanced machine translation but remain vulnerable to hallucinations. Unfortunately, existing MT benchmarks are not capable of exposing failures in multilingual LLMs. To disclose hallucination in…

计算与语言 · 计算机科学 2025-10-29 Xinwei Wu , Heng Liu , Jiang Zhou , Xiaohu Zhao , Linlong Xu , Longyue Wang , Weihua Luo , Kaifu Zhang
‹ 上一页 1 2 3 10 下一页 ›