Source-Grounded Semantic Reinforcement Learning for Low-Resource Target-Language Generation

• URL: http://arxiv.org/abs/2605.29502v1
• Date: Thu, 28 May 2026 07:27:16 GMT
• Title: Source-Grounded Semantic Reinforcement Learning for Low-Resource Target-Language Generation
• Abstract summary: Source-Grounded Semantic Reinforcement Learning (SG-SRL) is a resource-utilization framework that converts source-language monolingual data into cross-lingual semantic supervision for target-language generation.<n>SG-SRL performs reference-free reinforcement learning (RL) on source-language data using a cross-lingual semantic reward model.<n>Experiments on Chinese-to-Thai generation show that SG-SRL improves semantic grounding and factual coverage over cold-start SFT.
• Score: 19.750295463954853
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Low-resource target-language generation is often limited by scarce parallel data, while high-resource source-language monolingual data is abundant but difficult to use with standard supervised fine-tuning. We propose Source-Grounded Semantic Reinforcement Learning (SG-SRL), a resource-utilization framework that converts source-language monolingual data into cross-lingual semantic supervision for target-language generation. SG-SRL performs reference-free reinforcement learning (RL) on source-language data using a cross-lingual semantic reward model, instantiated by a cross-lingual reranker that scores the semantic relevance between the source input and the target-language generation. While this induces severe verbosity-based reward hacking, a lightweight recovery stage using a small parallel corpus restores fluency, conciseness, and task format while preserving the semantic gains. Experiments on Chinese-to-Thai generation show that SG-SRL improves semantic grounding and factual coverage over cold-start SFT. Additional analyses on long-form transfer and Tibetan embedding-based rewards clarify the generalization behavior of SG-SRL and show that an encoder-based semantic reward can substitute for an LLM-based reranker in a realistic low-resource language setting.

Related papers

• Reinforcement Learning Elicits Contextual Learning of Unseen Language Translation [35.3382718505629]
  We propose a reinforcement learning (RL) approach to unseen language translation given rich linguistic context.<n>Despite the lightweight reward, our RL-trained models effectively extract and apply relevant linguistic information from the provided context.<n>Our analyses suggest that outcome-based RL can extend beyond conventional reasoning tasks like math and coding to serve as a recipe for language learning from context.
  arXiv  Detail & Related papers  (2026-06-04T17:32:06Z)
• Reasoning over Grammar: Can Synthetic Linguistic Reasoning Traces Enhance Low-Resource Machine Translation? [49.7935995447581]
  We investigate whether low-resource machine translation can benefit from structured intermediate steps of linguistic analysis and grammatical reasoning.<n>We propose a pipeline for automatically generating step-by-step linguistic reasoning traces from Universal Dependencies treebanks, dictionaries, and grammar-rule banks.<n>Our results show that linguistic reasoning traces are most effective as inference-time guidance.
  arXiv  Detail & Related papers  (2026-06-02T15:36:12Z)
• Semantic Aware Linear Transfer by Recycling Pre-trained Language Models for Cross-lingual Transfer [5.990773821761297]
  SALT is a novel cross-lingual transfer technique that recycles embeddings from target language Pre-trained Language Models (PLMs)<n>Our experiments show that SALT significantly outperforms other transfer methods and achieves lower loss with accelerating faster convergence during language adaptation.
  arXiv  Detail & Related papers  (2025-05-16T07:30:22Z)
• Think Carefully and Check Again! Meta-Generation Unlocking LLMs for Low-Resource Cross-Lingual Summarization [108.6908427615402]
  Cross-lingual summarization ( CLS) aims to generate a summary for the source text in a different target language.<n>Currently, instruction-tuned large language models (LLMs) excel at various English tasks.<n>Recent studies have shown that LLMs' performance on CLS tasks remains unsatisfactory even with few-shot settings.
  arXiv  Detail & Related papers  (2024-10-26T00:39:44Z)
• Cross-lingual Back-Parsing: Utterance Synthesis from Meaning Representation for Zero-Resource Semantic Parsing [6.074150063191985]
  Cross-Lingual Back-Parsing is a novel data augmentation methodology designed to enhance cross-lingual transfer for semantic parsing. Our methodology effectively performs cross-lingual data augmentation in challenging zero-resource settings.
  arXiv  Detail & Related papers  (2024-10-01T08:53:38Z)
• SMILE: Speech Meta In-Context Learning for Low-Resource Language Automatic Speech Recognition [55.2480439325792]
  Speech Meta In-Context LEarning (SMILE) is an innovative framework that combines meta-learning with speech in-context learning (SICL)<n>We show that SMILE consistently outperforms baseline methods in training-free few-shot multilingual ASR tasks.
  arXiv  Detail & Related papers  (2024-09-16T16:04:16Z)
• MoE-CT: A Novel Approach For Large Language Models Training With Resistance To Catastrophic Forgetting [53.77590764277568]
  We introduce a novel MoE-CT architecture that separates the base model's learning from the multilingual expansion process. Our design freezes the original LLM parameters, thus safeguarding its performance in high-resource languages, while an appended MoE module, trained on diverse language datasets, augments low-resource language proficiency.
  arXiv  Detail & Related papers  (2024-06-25T11:03:45Z)
• Enhancing Multilingual Capabilities of Large Language Models through Self-Distillation from Resource-Rich Languages [60.162717568496355]
  Large language models (LLMs) have been pre-trained on multilingual corpora. Their performance still lags behind in most languages compared to a few resource-rich languages.
  arXiv  Detail & Related papers  (2024-02-19T15:07:32Z)
• MoSECroT: Model Stitching with Static Word Embeddings for Crosslingual Zero-shot Transfer [50.40191599304911]
  We introduce MoSECroT Model Stitching with Static Word Embeddings for Crosslingual Zero-shot Transfer. In this paper, we present the first framework that leverages relative representations to construct a common space for the embeddings of a source language PLM and the static word embeddings of a target language. We show that although our proposed framework is competitive with weak baselines when addressing MoSECroT, it fails to achieve competitive results compared with some strong baselines.
  arXiv  Detail & Related papers  (2024-01-09T21:09:07Z)
• Learning Cross-lingual Mappings for Data Augmentation to Improve Low-Resource Speech Recognition [31.575930914290762]
  Exploiting cross-lingual resources is an effective way to compensate for data scarcity of low resource languages. We extend the concept of learnable cross-lingual mappings for end-to-end speech recognition. The results show that any source language ASR model can be used for a low-resource target language recognition.
  arXiv  Detail & Related papers  (2023-06-14T15:24:31Z)
• Cross-Lingual Semantic Role Labeling with High-Quality Translated Training Corpus [41.031187560839555]
  Cross-lingual semantic role labeling is one promising way to address the problem. We propose a novel alternative based on corpus translation, constructing high-quality training datasets for the target languages. Experimental results on Universal Proposition Bank show that the translation-based method is highly effective.
  arXiv  Detail & Related papers  (2020-04-14T04:16:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.