Improving Cross-Lingual Factual Recall via Consistency-Driven Reinforcement Learning

• URL: http://arxiv.org/abs/2606.06586v1
• Date: Thu, 04 Jun 2026 18:00:02 GMT
• Title: Improving Cross-Lingual Factual Recall via Consistency-Driven Reinforcement Learning
• Abstract summary: Large language models (LLMs) trained predominantly on English data encode substantial world knowledge, yet often fail to express it reliably in other languages.<n>We introduce PolyFact, a large-scale parallel multilingual factual QA dataset containing 100K Wikidata-grounded facts across 12 typologically diverse languages.<n>We find that GRPO consistently outperforms SFT, improving both cross-lingual consistency and generalization to unseen languages.
• Score: 9.485903240691018
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) trained predominantly on English data encode substantial world knowledge, yet often fail to express it reliably in other languages, a phenomenon known as cross-lingual factual inconsistency. To study and address this, we introduce PolyFact, a large-scale parallel multilingual factual QA dataset containing 100K Wikidata-grounded facts across 12 typologically diverse languages. Using PolyFact, we compare light continual pretraining (CPT), supervised fine-tuning (SFT), and reinforcement learning via Group Relative Policy Optimization (GRPO) for improving cross-lingual factual recall in Qwen-2.5-7B and OLMo-2-1124-7B. We find that GRPO consistently outperforms SFT, improving both cross-lingual consistency and generalization to unseen languages, while CPT on parallel data yields limited additional gains. Mechanistic analyses further show that GRPO reorganizes multilingual routing by reducing language specialization in MLP layers and attention heads, thereby promoting more shared cross-lingual representations. We release our code, models, and dataset.

Related papers

• Bridging Linguistic Gaps: Cross-Lingual Mapping in Pre-Training and Dataset for Enhanced Multilingual LLM Performance [30.9640918372872]
  Large Language Models (LLMs) struggle with cross-lingual tasks due to data imbalances between high-resource and low-resource languages.<n>Existing methods, such as bilingual fine-tuning and contrastive alignment, can improve cross-lingual performance.<n>We introduce a Cross-Lingual Mapping Task during the pre-training phase, which enhances cross-lingual alignment without compromising monolingual fluency.
  arXiv  Detail & Related papers  (2026-04-12T11:43:25Z)
• An Empirical Recipe for Universal Phone Recognition [63.45609714127985]
  Phone recognition (PR) is a key enabler of multilingual and low-resource speech processing tasks.<n>We present PhoneticXEUS -- trained on large-scale multilingual data.<n>We achieve state-of-the-art performance on both multilingual (17.7% PFER) and accented English speech (10.6% PFER)
  arXiv  Detail & Related papers  (2026-03-30T22:12:48Z)
• Optimizing Language Models for Crosslingual Knowledge Consistency [90.86445137816942]
  Large language models are known to often exhibit inconsistent knowledge.<n>This is particularly problematic in multilingual scenarios, where models are likely to be asked similar questions in different languages.<n>In this work, we show that this issue can be mitigated using reinforcement learning with a structured reward function.
  arXiv  Detail & Related papers  (2026-03-04T23:36:55Z)
• Exploring Polyglot Harmony: On Multilingual Data Allocation for Large Language Models Pretraining [16.590296049892576]
  This paper introduces Climb, a novel framework designed to systematically optimize multilingual data allocation.<n>At its core, Climb introduces a cross-lingual interaction-aware language ratio, explicitly quantifying each language's effective allocation by capturing inter-language dependencies.<n>Extensive experiments confirm that Climb can accurately measure cross-lingual interactions across various multilingual settings.
  arXiv  Detail & Related papers  (2025-09-19T03:34:34Z)
• LaV-CoT: Language-Aware Visual CoT with Multi-Aspect Reward Optimization for Real-World Multilingual VQA [39.131225916852834]
  Chain-of-thought (CoT) reasoning has been proven to enhance interpretability and complex reasoning.<n>LaV-CoT is the first Language-aware Visual CoT framework with Multi-Aspect Reward Optimization.<n>LaV-CoT achieves up to 9.5% accuracy improvements over open-source baselines.
  arXiv  Detail & Related papers  (2025-09-12T07:45:44Z)
• Cross-Lingual Pitfalls: Automatic Probing Cross-Lingual Weakness of Multilingual Large Language Models [55.14276067678253]
  This paper introduces a novel methodology for efficiently identifying inherent cross-lingual weaknesses in Large Language Models (LLMs)<n>We construct a new dataset of over 6,000 bilingual pairs across 16 languages using this methodology, demonstrating its effectiveness in revealing weaknesses even in state-of-the-art models.<n>Further experiments investigate the relationship between linguistic similarity and cross-lingual weaknesses, revealing that linguistically related languages share similar performance patterns.
  arXiv  Detail & Related papers  (2025-05-24T12:31:27Z)
• From Unaligned to Aligned: Scaling Multilingual LLMs with Multi-Way Parallel Corpora [80.0547333327488]
  We introduce a large-scale, high-quality multi-way parallel corpus, TED2025, based on TED Talks.<n>This dataset spans 113 languages, with up to 50 languages aligned in parallel, ensuring extensive multilingual coverage.<n>Experiments show that models trained on multiway parallel data consistently outperform those trained on unaligned multilingual data.
  arXiv  Detail & Related papers  (2025-05-20T07:43:45Z)
• Trans-Zero: Self-Play Incentivizes Large Language Models for Multilingual Translation Without Parallel Data [64.4458540273004]
  We propose a self-play framework that leverages only monolingual data and the intrinsic multilingual knowledge of Large Language Models (LLMs)<n>Experiments demonstrate that this approach not only matches the performance of models trained on large-scale parallel data but also excels in non-English translation directions.
  arXiv  Detail & Related papers  (2025-04-20T16:20:30Z)
• Crosslingual Capabilities and Knowledge Barriers in Multilingual Large Language Models [62.91524967852552]
  Large language models (LLMs) are typically multilingual due to pretraining on diverse multilingual corpora.<n>But can these models relate corresponding concepts across languages, i.e., be crosslingual?<n>This study evaluates state-of-the-art LLMs on inherently crosslingual tasks.
  arXiv  Detail & Related papers  (2024-06-23T15:15:17Z)
• Improving Low-resource Reading Comprehension via Cross-lingual Transposition Rethinking [0.9236074230806579]
  Extractive Reading (ERC) has made tremendous advances enabled by the availability of large-scale high-quality ERC training data. Despite of such rapid progress and widespread application, the datasets in languages other than high-resource languages such as English remain scarce. We propose a Cross-Lingual Transposition ReThinking (XLTT) model by modelling existing high-quality extractive reading comprehension datasets in a multilingual environment.
  arXiv  Detail & Related papers  (2021-07-11T09:35:16Z)
• Multilingual Transfer Learning for QA Using Translation as Data Augmentation [13.434957024596898]
  We explore strategies that improve cross-lingual transfer by bringing the multilingual embeddings closer in the semantic space. We propose two novel strategies, language adversarial training and language arbitration framework, which significantly improve the (zero-resource) cross-lingual transfer performance. Empirically, we show that the proposed models outperform the previous zero-shot baseline on the recently introduced multilingual MLQA and TyDiQA datasets.
  arXiv  Detail & Related papers  (2020-12-10T20:29:34Z)

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.