Optimizing Language Models for Crosslingual Knowledge Consistency

• URL: http://arxiv.org/abs/2603.04678v1
• Date: Wed, 04 Mar 2026 23:36:55 GMT
• Title: Optimizing Language Models for Crosslingual Knowledge Consistency
• Authors: Tianyu Liu, Jirui Qi, Mrinmaya Sachan, Ryan Cotterell, Raquel Fernández, Arianna Bisazza,
• Abstract summary: 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.
• Score: 90.86445137816942
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models are known to often exhibit inconsistent knowledge. This is particularly problematic in multilingual scenarios, where models are likely to be asked similar questions in different languages, and inconsistent responses can undermine their reliability. In this work, we show that this issue can be mitigated using reinforcement learning with a structured reward function, which leads to an optimal policy with consistent crosslingual responses. We introduce Direct Consistency Optimization (DCO), a DPO-inspired method that requires no explicit reward model and is derived directly from the LLM itself. Comprehensive experiments show that DCO significantly improves crosslingual consistency across diverse LLMs and outperforms existing methods when training with samples of multiple languages, while complementing DPO when gold labels are available. Extra experiments demonstrate the effectiveness of DCO in bilingual settings, significant out-of-domain generalizability, and controllable alignment via direction hyperparameters. Taken together, these results establish DCO as a robust and efficient solution for improving knowledge consistency across languages in multilingual LLMs. All code, training scripts, and evaluation benchmarks are released at https://github.com/Betswish/ConsistencyRL.

Related papers

• Language-Coupled Reinforcement Learning for Multilingual Retrieval-Augmented Generation [73.54930910609328]
  We propose LcRL, a multilingual search-augmented reinforcement learning framework.<n>LcRL integrates a language-coupled Group Relative Policy Optimization into the policy and reward models.<n>We adopt the language-coupled group sampling in the rollout module to reduce knowledge bias, and regularize an auxiliary anti-consistency penalty in the reward models to mitigate the knowledge conflict.
  arXiv  Detail & Related papers  (2026-01-21T11:32:32Z)
• Aligning LLMs for Multilingual Consistency in Enterprise Applications [39.24743635297897]
  Large language models (LLMs) remain unreliable for global enterprise applications.<n>We propose a practical, batch-wise alignment strategy for fine-tuning LLMs.<n>This approach improves non-English accuracy by up to 23.9% without compromising English performance, model reasoning, or retrieval quality.
  arXiv  Detail & Related papers  (2025-09-28T05:51:22Z)
• CausalAbstain: Enhancing Multilingual LLMs with Causal Reasoning for Trustworthy Abstention [9.76878200328024]
  Large Language Models (LLMs) often exhibit knowledge disparities across languages.<n>We introduce textitCausalAbstain, a method that helps LLMs determine whether to utilize multiple generated feedback responses.<n>Experiments demonstrate that textitCausalAbstain effectively selects helpful feedback and enhances abstention decisions with interpretability.
  arXiv  Detail & Related papers  (2025-05-31T11:35:31Z)
• 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)
• CALM: Unleashing the Cross-Lingual Self-Aligning Ability of Language Model Question Answering [42.92810049636768]
  Large Language Models (LLMs) are pretrained on extensive multilingual corpora to acquire both language-specific cultural knowledge and general knowledge.<n>We explore the Cross-Lingual Self-Aligning ability of Language Models (CALM) to align knowledge across languages.<n>We employ direct preference optimization (DPO) to align the model's knowledge across different languages.
  arXiv  Detail & Related papers  (2025-01-30T16:15:38Z)
• P-MMEval: A Parallel Multilingual Multitask Benchmark for Consistent Evaluation of LLMs [84.24644520272835]
  We introduce P-MMEval, a large-scale benchmark covering effective fundamental and capability-specialized datasets.<n>P-MMEval delivers consistent language coverage across various datasets and provides parallel samples.<n>We conduct extensive experiments on representative multilingual model series to compare performances across models and tasks.
  arXiv  Detail & Related papers  (2024-11-14T01:29:36Z)
• 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)
• Getting More from Less: Large Language Models are Good Spontaneous Multilingual Learners [67.85635044939836]
  Large Language Models (LLMs) have shown impressive language capabilities. In this work, we investigate the spontaneous multilingual alignment improvement of LLMs. We find that LLMs instruction-tuned on the question translation data (i.e. without annotated answers) are able to encourage the alignment between English and a wide range of languages.
  arXiv  Detail & Related papers  (2024-05-22T16:46:19Z)
• Bridging the Language Gap: Dynamic Learning Strategies for Improving Multilingual Performance in LLMs [15.911445732909849]
  Large language models (LLMs) have revolutionized various domains but still struggle with non-Latin scripts and low-resource languages.<n>We introduce a novel dynamic learning approach that optimize prompt strategy, embedding model, and LLM per query at runtime.<n>We show our approach results in 10-15% improvements in multilingual performance over pre-trained models and 4x gains compared to fine-tuned, language-specific models.
  arXiv  Detail & Related papers  (2023-05-28T14:48:38Z)

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.