Skip to content

Multilingual Reasoning Cascades Need More Context

Jun 2026 · arXiv.org · Vol abs/2606.27306 · 0 citations · 27 references
Computer Science

Abstract

Translation cascades for reasoning translate the query from another language to English, reason in English, and translate the answer back to the original language. This is a competitive approach to multilingual reasoning, but structurally lossy, since each stage discards information later stages may need, including cues for cultural grounding, register, and disambiguation. We examine the benefits of a simple and training-free intervention: a context-aware translation cascade, which additionally provides the original question, the English translated question, and the reasoning trace to the context of the final translation module. We evaluate gains across nine multilingual benchmarks including various task types, three backbone models, and 285 high-, mid-, and low-resource languages, and demonstrate strong gains for open-ended generation across models and resource regimes. We show that the original language question carries most of the beneficial context. Our study emphasizes the need to better design information flow in machine translation cascades for mitigating error propagation, and provides a simple and actionable default strategy: preserve the original user question until the end of the pipeline.

View source

Similar papers

Preprint Jul 2026

Reasoning Before Translation: Enhancing Legal Machine Translation with Structured Reasoning

The results show that the quality of small ``base'' models can be greatly enhanced, and that reinforcement learning with verifiable rewards can be applied to NMT in the legal domain and surpasses the translation quality of supervised fine-tuning.

Aixiu An, Michael Jungo, Eloi Eynard et al. · 0 citations
Preprint Aug 2026

Reasoning about In-Context Samples for Machine-Translation

A novel fragment-based reasoning framework is introduced in which the model first extracts parallel source-target fragments from retrieved similar exemplars, and uses these fragments as intermediate reasoning traces to produce the final translation.

Maxime Bouthors, J. Crego, François Yvon · 0 citations
Preprint Aug 2026

Native Multilingual Chain-of-Thought Reasoning in Low-Resource Southeast Asian Languages

The Onramp-Sequence Cross-Distillation (OSCD) is introduced, a post-training algorithm that projects high-resource reasoning trajectories into low-resource vocabulary subspaces during generative training rollouts via an integrated translator agentic loop, ensuring the stable and efficient translation of dynamically generated reference samples for fine-tuning.

Sean Gip Lim, William-Chandra Tjhi, Hai Leong Chieu · 0 citations
Book Open access Aug 2026

Semantic-Symbolic Knowledge Consensus for Multilingual Question Answering

This paper proposes SeSyCo, a Semantic-Symbolic Knowledge Consensus framework, which leverages the semantic space to diverge monolingual queries into broad multilingual evidence, and subsequently utilize the symbolic space to eliminate language discrepancies, converging the gathered information into a robust consensus for precise SPARQL generation.

Yu Zhang, Ran Song, Xiaofei Gao et al. · 0 citations
Preprint Jul 2026

PluraMath: Extending Mathematical Reasoning Evaluation Beyond High-Resource Languages

Mathematical reasoning has become a central task for evaluating and tuning reasoning Large Language Models (LLMs), yet existing benchmarks remain heavily biased toward high-resource languages, with English and Chinese dominating both pre-training corpora and evaluation suites. The recently released PolyMath (Wang et al., 2025) dataset represents a significant step forward, yet its coverage is still limited to 18 only high-resource languages. To address this gap, we introduce PluraMath, an extension of PolyMath to 18 additional {underrepresented languages spanning 6 language families -- ranging from mid-resource to extreme low-resource settings. We constructed the dataset through a human-curated pipeline, where native speakers thoroughly validated pre-computed translations. Using PluraMath, we then benchmark 27 reasoning LLMs across four model scales -- small, mid-size, large, and closed-source ensembles -- probing the multilingual mathematical reasoning capabilities of state-of-the-art models under diverse linguistic conditions. Our fine-grained analysis confirms a persistent gap in mathematical reasoning performance between high-resource and underrepresented languages, with stronger results largely associated with better instruction-following ability. We fully open-source our dataset, data acquisition pipeline, and evaluation framework, with the goal of lowering the barrier to multilingual benchmark development for underrepresented communities.

Daryna Dementieva, N. Babakov, Kathy Hammerl et al. · 0 citations