Redefining Machine Simultaneous Interpretation: From Incremental Translation to Human-Like Strategies

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Qianen, Nakamura, Satoshi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915515872575488
author Zhang, Qianen
Nakamura, Satoshi
author_facet Zhang, Qianen
Nakamura, Satoshi
contents Simultaneous Machine Translation (SiMT) requires high-quality translations under strict real-time constraints, which traditional encoder-decoder policies with only READ/WRITE actions cannot fully address. We extend the action space of SiMT with four adaptive actions: SENTENCE_CUT, DROP, PARTIAL_SUMMARIZATION and PRONOMINALIZATION, which enable real-time restructuring, omission, and simplification while preserving semantic fidelity. We implement these actions in a decoder-only large language model (LLM) framework and construct training references through action-aware prompting. To evaluate both quality and latency, we further develop a latency-aware TTS pipeline that maps textual outputs to speech with realistic timing. Experiments on the ACL60/60 English-Chinese and English-German benchmarks show that our framework consistently improves semantic metrics (e.g., COMET-KIWI) and achieves lower delay (measured by Average Lagging) compared to reference translations and salami-based baselines. Notably, combining DROP and SENTENCE_CUT yields the best overall balance between fluency and latency. These results demonstrate that enriching the action space of LLM-based SiMT provides a promising direction for bridging the gap between human and machine interpretation.
format Preprint
id arxiv_https___arxiv_org_abs_2509_21801
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Redefining Machine Simultaneous Interpretation: From Incremental Translation to Human-Like Strategies
Zhang, Qianen
Nakamura, Satoshi
Computation and Language
Simultaneous Machine Translation (SiMT) requires high-quality translations under strict real-time constraints, which traditional encoder-decoder policies with only READ/WRITE actions cannot fully address. We extend the action space of SiMT with four adaptive actions: SENTENCE_CUT, DROP, PARTIAL_SUMMARIZATION and PRONOMINALIZATION, which enable real-time restructuring, omission, and simplification while preserving semantic fidelity. We implement these actions in a decoder-only large language model (LLM) framework and construct training references through action-aware prompting. To evaluate both quality and latency, we further develop a latency-aware TTS pipeline that maps textual outputs to speech with realistic timing. Experiments on the ACL60/60 English-Chinese and English-German benchmarks show that our framework consistently improves semantic metrics (e.g., COMET-KIWI) and achieves lower delay (measured by Average Lagging) compared to reference translations and salami-based baselines. Notably, combining DROP and SENTENCE_CUT yields the best overall balance between fluency and latency. These results demonstrate that enriching the action space of LLM-based SiMT provides a promising direction for bridging the gap between human and machine interpretation.
title Redefining Machine Simultaneous Interpretation: From Incremental Translation to Human-Like Strategies
topic Computation and Language
url https://arxiv.org/abs/2509.21801