paper
Reasoning over Grammar: Can Synthetic Linguistic Reasoning Traces Enhance Low-Resource Machine Translation?
paperactiveprovisional
reasoning-over-grammar-can-synthetic-linguistic-reasoning-traces-enhance-low-resource-machine-translation--379ebca7·1 events·first seen 13d agoAliases: Reasoning over Grammar: Can Synthetic Linguistic Reasoning Traces Enhance Low-Resource Machine Translation?
Co-occurring entities
More like this (12)
Reasoning Language ModelsBeyond Accuracy: Community Perspectives on Machine TranslationData Synthesis and Parameter-Efficient Fine-Tuning for Low-Resource NMT: A Case Study on Q'eqchi' MayanReasoning EnhancementDoes Reasoning Preserve Alignment? On the Trustworthiness of Large Reasoning ModelsLong-context Reasoning BenchmarksLearning to Reason by Analogy via Retrieval-Augmented Reinforcement Fine-TuningNatural Language InferenceReinforcement Learning Elicits Contextual Learning of Unseen Language TranslationFailed Reasoning Traces Tell You What Is Fixable (But Not by Reading Them)Reasoning as Pattern Matching: Shared Mechanisms in Human and LLM Everyday ReasoningWhen English Isn't the Best Teacher: Source Language Effects in Cross-Lingual In-Context Learning
Recent events (1)
Synthetic linguistic reasoning traces improve low-resource machine translation via in-context learning
Researchers propose a pipeline that generates step-by-step linguistic reasoning traces from Universal Dependencies treebanks, dictionaries, and grammar-rule banks to assist LLMs in translating extremely low-resource languages. Evaluated on Xibe and Chintang across ICL, SFT, and RFT settings, the traces prove most effective as inference-time guidance rather than training data. Models can leverage reliable grammatical analyses at inference time but struggle to learn to generate accurate traces themselves, identifying trace generation quality as the key bottleneck.