Cross-Lingual Stability and Bias in Instruction-Tuned Language Models for Humanitarian NLP

Poli Nemkova, Amrit Adhikari, Matthew Pearson, Vamsi Krishna Sadu, Albert V. Mark


Abstract
Humanitarian organizations face a critical choice: invest in costly commercial APIs or rely on free open-weight models for multilingual human rights monitoring. While commercial systems offer reliability, open-weight alternatives lack empirical validation - especially for low-resource languages common in conflict zones. This paper presents the first systematic comparison of commercial and open-weight large language models (LLMs) for human-rights-violation detection across seven languages, quantifying the cost-reliability trade-off facing resource-constrained organizations. Across 78,000 multilingual inferences, we evaluate six models - four instruction-aligned (Claude-Sonnet-4, DeepSeek-V3, Gemini-Flash-2.0, GPT-4.1-mini) and two open-weight (LLaMA-3-8B, Mistral-7B) - using both standard classification metrics and new measures of cross-lingual reliability: Calibration Deviation (CD), Decision Bias (ΔBias), Language Robustness Score (LRS), and Language Stability Score (LSS). Results show that alignment, not scale, determines stability: aligned models maintain near-invariant accuracy and balanced calibration across typologically distant and low-resource languages (e.g., Lingala, Burmese), while open-weight models exhibit significant prompt-language sensitivity and calibration drift. These findings demonstrate that multilingual alignment enables language-agnostic reasoning and provide practical guidance for humanitarian organizations balancing budget constraints with reliability in multilingual deployment.
Anthology ID:
2026.lrec-1.346
Volume:
Proceedings of the Fifteenth Language Resources and Evaluation Conference
Month:
May
Year:
2026
Address:
Palma de Mallorca, Spain
Editors:
Stelios Piperidis, Núria Bel, Henk van den Heuvel, Nancy Ide, Simon Krek, Antonio Toral
Venue:
LREC
SIG:
Publisher:
ELRA Language Resource Association
Note:
Pages:
4423–4433
Language:
External URL:
https://lrec.elra.info/lrec2026-main-346
DOI:
10.63317/59n42irmw53u
Bibkey:
Cite (ACL):
Poli Nemkova, Amrit Adhikari, Matthew Pearson, Vamsi Krishna Sadu, and Albert V. Mark. 2026. Cross-Lingual Stability and Bias in Instruction-Tuned Language Models for Humanitarian NLP. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 4423–4433, Palma de Mallorca, Spain. ELRA Language Resource Association.
Cite (Informal):
Cross-Lingual Stability and Bias in Instruction-Tuned Language Models for Humanitarian NLP (Nemkova et al., LREC 2026)
Copy Citation: