Chain-of-Thought Reasoning Improves Context-Aware Translation with Large Language Models

Shabnam Ataee, Hugo Huart, Andrei Popescu-Belis


Abstract
This paper assesses the ability of large language models (LLMs) to translate texts that include inter-sentential dependencies. We use the English-French DiscEvalMT benchmark (Bawden et al., 2018) with pairs of sentences containing translation challenges for pronominal anaphora and lexical cohesion. We evaluate 12 LLMs from the DeepSeek-R1, GPT, Llama, Mistral and Phi families on two tasks: (1) distinguish a correct translation from a wrong but plausible one; and (2) generate a correct translation. We compare prompts that encourage chain-of-thought reasoning with those that do not. The best models take advantage of reasoning and reach about 90% accuracy on the first task and COMET scores of about 92% on the second task, with GPT-4, GPT-4o and Phi standing out. Moreover, we observe a “wise get wiser” effect: the improvements through reasoning are larger for models that already perform well without reasoning.
Anthology ID:
2026.lrec-1.298
Volume:
Proceedings of the Fifteenth Language Resources and Evaluation Conference
Month:
May
Year:
2026
Address:
Palma de Mallorca, Spain
Editors:
Stelios Piperidis, Núria Bel, Henk van den Heuvel, Nancy Ide, Simon Krek, Antonio Toral
Venue:
LREC
SIG:
Publisher:
ELRA Language Resource Association
Note:
Pages:
3725–3741
Language:
External URL:
https://lrec.elra.info/lrec2026-main-298
DOI:
10.63317/37kz9rawf9fn
Bibkey:
Cite (ACL):
Shabnam Ataee, Hugo Huart, and Andrei Popescu-Belis. 2026. Chain-of-Thought Reasoning Improves Context-Aware Translation with Large Language Models. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 3725–3741, Palma de Mallorca, Spain. ELRA Language Resource Association.
Cite (Informal):
Chain-of-Thought Reasoning Improves Context-Aware Translation with Large Language Models (Ataee et al., LREC 2026)
Copy Citation: