Skip to content
AI News HubLIVE
Original source2 min read

A Systematic Evaluation of Cross-Lingual Consistency Enhancement Methods in Multilingual Language Models

Summary

This paper presents a unified evaluation of methods aimed at improving cross-lingual consistency in multilingual language models for question answering. The authors compare inference-time interventions and post-training approaches across three model families and three closed-form benchmarks. Post-training methods, especially direct distribution alignment, prove more reliable, while cross-domain transfer remains limited. Tests on culturally diverse QA show no systematic closed-form degradation, but open-ended generation occasionally loses accuracy for non-English responses.

SourcearXiv Computational LinguisticsAuthor: Jirui Qi, Mingyang Wang, Hinrich Sch\"utze, Raquel Fern\'andez, Arianna Bisazza
A Systematic Evaluation of Cross-Lingual Consistency Enhancement Methods in Multilingual Language Models
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 3 Sep 2026]

Title:A Systematic Evaluation of Cross-Lingual Consistency Enhancement Methods in Multilingual Language Models

View a PDF of the paper titled A Systematic Evaluation of Cross-Lingual Consistency Enhancement Methods in Multilingual Language Models, by Jirui Qi and 4 other authors

View PDF HTML (experimental)

Abstract:Multilingual language models often produce inconsistent answers to semantically equivalent questions across languages, motivating methods to improve cross-lingual consistency (CLC). However, existing methods are typically evaluated using different models, tasks, and protocols, leaving their relative strengths unclear. In this work, we present a unified evaluation of representative CLC-enhancement methods for question answering, spanning inference-time interventions and post-training approaches across three model families and three closed-form benchmarks. The results show that post-training methods are generally more reliable, with direct distribution alignment consistently improving CLC across all model-dataset combinations, while other methods are more sensitive to answer format and the breadth of language coverage. Notably, cross-domain transfer is limited unless source and target tasks share similar output formats. We further investigate whether CLC enhancement hurts models' ability to respond differently *when needed*, that is, when asked culture-dependent questions. Across two benchmarks of culturally diverse question answering, we find no systematic degradation in controlled closed-form evaluation, whereas open-ended generation reveals occasional accuracy reductions, particularly for non-English responses. Our work highlights the need to evaluate CLC enhancement for both cross-domain robustness and culturally appropriate variation, informing future work in post-training and benchmark development.

Comments: Preprint. All code and datasets will be released upon publication

Subjects:

Computation and Language (cs.CL); Artificial Intelligence (cs.AI)

Cite as: arXiv:2609.04409 [cs.CL]

(or arXiv:2609.04409v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2609.04409

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Jirui Qi [view email] [v1] Thu, 3 Sep 2026 19:18:44 UTC (1,645 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled A Systematic Evaluation of Cross-Lingual Consistency Enhancement Methods in Multilingual Language Models, by Jirui Qi and 4 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CL

new | recent | 2026-09

Change to browse by:

cs cs.AI

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

InvestorsAdvanced

Key points

  • A unified benchmark compares inference-time and post-training methods for cross-lingual consistency. IgnoreIgnore
  • Post-training with direct distribution alignment improves consistency across all tested models and datasets. Ignore
  • Cross-domain transfer is limited unless source and target tasks share output formats.
  • Closed-form culturally diverse QA shows no systematic degradation, but open-ended generation can hurt non-English accuracy.

Highlights and analysis are generated automatically and may contain errors. Check the original source.