Pith. sign in

REVIEW 5 cited by

Crosslingual Capabilities and Knowledge Barriers in Multilingual Large Language Models

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2406.16135 v2 pith:3TQX4GWV submitted 2024-06-23 cs.CL cs.LG

Lynn Chua , Badih Ghazi , Yangsibo Huang , Pritish Kamath , Ravi Kumar , Pasin Manurangsi , Amer Sinha , Chulin Xie
show 1 more author
Chiyuan Zhang
This is my paper · ORCID
classification cs.CLcs.LG
keywords crosslingualllmsmodelsknowledgemultilingualbenchmarklanguagelarge
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Large language models (LLMs) are typically multilingual due to pretraining on diverse multilingual corpora. But can these models relate corresponding concepts across languages, i.e., be crosslingual? This study evaluates state-of-the-art LLMs on inherently crosslingual tasks. We observe that while these models show promising surface-level crosslingual abilities on machine translation and embedding space analyses, they struggle with deeper crosslingual knowledge transfer, revealing a crosslingual knowledge barrier in both general (MMLU benchmark) and domain-specific (Harry Potter quiz and TOFU benchmark) contexts. Since simple inference-time mitigation methods offer only limited improvement, we propose fine-tuning of LLMs on mixed-language data, which effectively reduces these gaps, even when using out-of-domain datasets like WikiText. Our findings suggest the need for explicit optimization to unlock the full crosslingual potential of LLMs. Our code is publicly available at https://github.com/google-research/crosslingual-knowledge-barriers.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Unveiling Language Routing Isolation in Multilingual MoE Models for Interpretable Subnetwork Adaptation

    cs.CL 2026-04 conditional novelty 6.0 of 10

    MoE models isolate high- vs low-resource languages in expert routing; training only the isolated target subnetwork (RISE) lifts low-resource F1 by up to ~11 points with little cross-lingual loss.

  2. Rethinking Cross-lingual Gaps from a Statistical Viewpoint

    cs.CL 2025-10 conditional novelty 6.0 of 10

    Cross-lingual accuracy gaps in LLMs are dominated by higher response variance in target languages, not missing knowledge; ensembling and variance-reduction prompts shrink the gap.

  3. AI4Research: A Survey of Artificial Intelligence for Scientific Research

    cs.CL 2025-07 conditional novelty 4.0 of 10

    A survey that organizes AI-for-research work into five tasks, comprehension, survey, discovery, writing, and peer review, and compiles associated tools and benchmarks.

  4. Mitigating Negative Interference in Multilingual Sequential Knowledge Editing through Null-Space Constraints

    cs.CL 2025-06 conditional novelty 4.0 of 10

    By projecting each language's weight update into the null space of previous updates, LangEdit reports small consistent gains over AlphaEdit on a newly defined multilingual sequential editing task.

  5. SweEval: Do LLMs Really Swear? A Safety Benchmark for Testing Limits for Enterprise Use

    cs.CL 2025-05 conditional novelty 4.0 of 10

    A new cross-lingual benchmark shows large language models comply with explicit requests to use swear words far more often in Indic languages than in English, revealing a safety alignment gap.

Pith tools