analiza aiqbench
This research exposes critical flaws in current multilingual embedding benchmarks, particularly for Slavic languages, where dataset scarcity undermines the reliability of model rankings and generalization claims. By introducing an Evidence Strength Score, the authors provide a necessary metric to distinguish between robust findings and those derived from sparse or correlated data sources. Ultimately, the study argues that trustworthy cross-lingual evaluation requires addressing fundamental gaps in low-resource language datasets before drawing definitive conclusions about model performance.
Dataset Scarcity Limits Robust Evaluation of Multilingual Embedding Models: A Case Study of Slavic Languages
arXiv:2608.24477v1 Announce Type: new Abstract: Multilingual text embedding models enable cross-lingual transfer of knowledge across a wide range of NLP tasks, but their evaluation remains highly uneven across high-, mid- and low-resource languages. In this paper, we propose a two-dimensional framework, specifically tailored for analyzing multilingual embedding benchmarks under dataset scarcity, and apply it on the Slavic-language subset of the MTEB benchmark. The framework distinguishes between task-specific and cross-task evaluation, while jointly analyzing three complementary aspects: (1) ranking robustness, (2) model consistency, and (3) evidence strength. At the task-specific level, we evaluate the stability of model rankings under changes in ranking methodology and benchmark dataset composition. At the cross-task level, we assess the ability of models to generalize across diverse tasks within a language. To quantify the reliability of benchmark conclusions, we introduce an Evidence Strength Score that accounts for dataset availability, diversity, and robustness assessability. Our analysis reveals severe benchmark sparsity, with many Slavic language-task pairs relying on a single dataset or highly correlated benchmark collections, limiting the ability to draw robust conclusions. The cross-task analysis reveals a small group of highly transferable models, most notably llama-embed-nemotron-8b, multilingual-e5-large-instruct, and Qwen3-Embedding variants, that consistently perform well across Slavic languages and tasks. Overall, the results demonstrate that benchmark rankings and robustness conclusions must be interpreted jointly with certain notation of their evidence strength and highlight benchmark scarcity as a major obstacle to trustworthy multilingual evaluation.
Abstract:Multilingual text embedding models enable cross-lingual transfer of knowledge across a wide range of NLP tasks, but their evaluation remains highly uneven across high-, mid- and low-resource languages. In this paper, we propose a two-dimensional framework, specifically tailored for analyzing multilingual embedding benchmarks under dataset scarcity, and apply it on the Slavic-language subset of the MTEB benchmark. The framework distinguishes between task-specific and cross-task evaluation, while jointly analyzing three complementary aspects: (1) ranking robustness, (2) model consistency, and (3) evidence strength. At the task-specific level, we evaluate the stability of model rankings under changes in ranking methodology and benchmark dataset composition. At the cross-task level, we assess the ability of models to generalize across diverse tasks within a language. To quantify the reliability of benchmark conclusions, we introduce an Evidence Strength Score that accounts for dataset availability, diversity, and robustness assessability. Our analysis reveals severe benchmark sparsity, with many Slavic language-task pairs relying on a single dataset or highly correlated benchmark collections, limiting the ability to draw robust conclusions. The cross-task analysis reveals a small group of highly transferable models, most notably llama-embed-nemotron-8b, multilingual-e5-large-instruct, and Qwen3-Embedding variants, that consistently perform well across Slavic languages and tasks. Overall, the results demonstrate that benchmark rankings and robustness conclusions must be interpreted jointly with certain notation of their evidence strength and highlight benchmark scarcity as a major obstacle to trustworthy multilingual evaluation.
Computation and Language (cs.CL)
arXiv:2608.24477 [cs.CL]
(or arXiv:2608.24477v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2608.24477 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
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.
Text integral sub licența deschisă a publicației · 324 cuvinte
Subiecte conexe
Enabling Language-specific Reasoning in Multilingual Models with Reinforcement Learning
AMD · 24 iul. 2026
Introducing Embed 5—A new family of frontier embedding models
Cohere · 30 sept. 2026
ChemCLIR-Bench: Benchmarking Cross-Lingual Information Retrieval in Multilingual Chemical Patents
arXiv · 22 sept. 2026
Can Voice Agents Handle Bilingual Customers? Benchmarking Frontier ASR on Code-Switched Speech
Hugging Face · 9 iun. 2026