Retrieval-Augmented Reasoning for Extremely Low-Resource Language Decipherment

Xiaoyu Zhu, Jinan Xu, Long Yuan, Fengran Mo, Kaiyu Huang

Keywords: Low-Resource Language Reasoning, Large Language Model, Decipherment

Abstract: Inspired by linguistic Olympiads, extremely low-resource language reasoning presents a unique challenge that enables models to solve problems without prior knowledge. This task mirrors the Rosetta Stone decipherment process, where the goal is to induce and apply linguistic rules from minimal context. Existing methods mainly rely on naive in-context learning that fails to handle the complexity and diversity of language rules. To mitigate this issue, we propose a framework that combines dynamic knowledge construction with task-aware retrieval augmentation. First, we use large language models (LLMs) to generate a diverse set of task-specific examples that instantiate potential linguistic rules for the target low-resource language. Second, we apply a semantic retrieval mechanism to select the most relevant examples for each test query, preventing context overload and ensuring focused, analogical reasoning. Our method shifts from learning language distributions to dynamically discovering and applying rules. Experimental results on the LINGOLY and Linguini benchmark show that our approach achieves competitive performance across various LLMs, outperforming existing baselines. More importantly, our framework advances extremely low-resource reasoning and provides a generalizable framework for rule induction under knowledge constraints.

Primary Area: other topics in machine learning (i.e., none of the above)

Submission Number: 9343