NeoAMT: Neologism-Aware Agentic Machine Translation with Reinforcement Learning

2026-01-07

arXiv.org

AI Summary

This work addresses the long-overlooked challenge of translating neologisms in machine translation by proposing an intelligent agent framework that integrates Wiktionary retrieval with reinforcement learning. The framework introduces a translation-difficulty-aware adaptive rollout generation mechanism and designs a novel reward function grounded in external knowledge. To support systematic evaluation, we construct the first large-scale neologism translation dataset spanning 16 languages and 75 translation directions. Experimental results demonstrate that the proposed approach significantly improves translation accuracy for neologisms across diverse language pairs, effectively validating the framework’s efficacy and generalization capability in handling complex, emerging lexical phenomena.

Technology Category

• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP
• Multiagent Systems: Adversarial Agents
• Machine Learning: Neuro-Symbolic Learning

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Neologism-aware machine translation aims to translate source sentences containing neologisms into target languages. This field remains underexplored compared with general machine translation (MT). In this paper, we propose an agentic framework, NeoAMT, for neologism-aware machine translation using a Wiktionary search tool. Specifically, we first create a new dataset for neologism-aware machine translation and develop a search tool based on Wiktionary. The new dataset covers 16 languages and 75 translation directions and is derived from approximately 10 million records of an English Wiktionary dump. The retrieval corpus of the search tool is also constructed from around 3 million cleaned records of the Wiktionary dump. We then use it for training the translation agent with reinforcement learning (RL) and evaluating the accuracy of neologism-aware machine translation. Based on this, we also propose an RL training framework that contains a novel reward design and an adaptive rollout generation approach by leveraging"translation difficulty"to further improve the translation quality of translation agents using our search tool.

Problem

• neologism-aware machine translation
• machine translation
• neologisms
• Wiktionary
• reinforcement learning

Innovation

• neologism-aware machine translation
• agentic framework
• Wiktionary-based retrieval
• reinforcement learning
• adaptive rollout generation