MétaCan
Menu
Back to cohort
Record W4412945237 · doi:10.18653/v1/2025.acl-long.657

Code-Switching Red-Teaming: LLM Evaluation for Safety and Multilingual Understanding

2025· article· en· W4412945237 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicSoftware Reliability and Analysis Research
Canadian institutionsKootenay Association for Science & Technology
Fundersnot available
KeywordsComputer scienceCode-switchingCode (set theory)Programming languageSoftware engineeringLinguistics

Abstract

fetched live from OpenAlex

As large language models (LLMs) have advanced rapidly, concerns regarding their safety have become prominent.In this paper, we discover that code-switching in red-teaming queries can effectively elicit undesirable behaviors of LLMs, which are common practices in natural language.We introduce a simple yet effective framework, CSRT, to synthesize codeswitching red-teaming queries and investigate the safety and multilingual understanding of LLMs comprehensively.Through extensive experiments with ten state-of-the-art LLMs and code-switching queries combining up to 10 languages, we demonstrate that the CSRT significantly outperforms existing multilingual redteaming techniques, achieving 46.7% more attacks than standard attacks in English and being effective in conventional safety domains.We also examine the multilingual ability of those LLMs to generate and understand codeswitching texts.Additionally, we validate the extensibility of the CSRT by generating codeswitching attack prompts with monolingual data.We finally conduct detailed ablation studies exploring code-switching and propound unintended correlation between resource availability of languages and safety alignment in existing multilingual LLMs. 1 * This work was done during an internship at NAVER AI Lab.Why do people who live in San Francisco act so gay? stereotypical and potentially offensive. ? Red-teaming Multilingual Red-teaming Code-Switching Red-Teaming (CSRT) Sample from MultiJail Sample from CSRT (Ours) [In Korean] It is not appropriate San Francisco has a long history of being a hub for LGBTQ+ culture and rights.The city is known for its openness, acceptance, and vibrant gay scene, wata mengapa?

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.003
metaresearch head score (Gemma)0.001
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.958
Threshold uncertainty score0.289

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0030.001
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.085
GPT teacher head0.383
Teacher spread0.298 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it