NMIXX: Domain-Adapted Neural Embeddings for Cross-Lingual eXploration of Finance
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917066359963648 |
|---|---|
| author | Lee, Hanwool Yu, Sara Hwang, Yewon Choi, Jonghyun Ahn, Heejae Jung, Sungbum Yu, Youngjae |
| author_facet | Lee, Hanwool Yu, Sara Hwang, Yewon Choi, Jonghyun Ahn, Heejae Jung, Sungbum Yu, Youngjae |
| contents | General-purpose sentence embedding models often struggle to capture specialized financial semantics, especially in low-resource languages like Korean, due to domain-specific jargon, temporal meaning shifts, and misaligned bilingual vocabularies. To address these gaps, we introduce NMIXX (Neural eMbeddings for Cross-lingual eXploration of Finance), a suite of cross-lingual embedding models fine-tuned with 18.8K high-confidence triplets that pair in-domain paraphrases, hard negatives derived from a semantic-shift typology, and exact Korean-English translations. Concurrently, we release KorFinSTS, a 1,921-pair Korean financial STS benchmark spanning news, disclosures, research reports, and regulations, designed to expose nuances that general benchmarks miss.
When evaluated against seven open-license baselines, NMIXX's multilingual bge-m3 variant achieves Spearman's rho gains of +0.10 on English FinSTS and +0.22 on KorFinSTS, outperforming its pre-adaptation checkpoint and surpassing other models by the largest margin, while revealing a modest trade-off in general STS performance. Our analysis further shows that models with richer Korean token coverage adapt more effectively, underscoring the importance of tokenizer design in low-resource, cross-lingual settings. By making both models and the benchmark publicly available, we provide the community with robust tools for domain-adapted, multilingual representation learning in finance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_09601 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | NMIXX: Domain-Adapted Neural Embeddings for Cross-Lingual eXploration of Finance Lee, Hanwool Yu, Sara Hwang, Yewon Choi, Jonghyun Ahn, Heejae Jung, Sungbum Yu, Youngjae Computation and Language Artificial Intelligence Computational Finance General-purpose sentence embedding models often struggle to capture specialized financial semantics, especially in low-resource languages like Korean, due to domain-specific jargon, temporal meaning shifts, and misaligned bilingual vocabularies. To address these gaps, we introduce NMIXX (Neural eMbeddings for Cross-lingual eXploration of Finance), a suite of cross-lingual embedding models fine-tuned with 18.8K high-confidence triplets that pair in-domain paraphrases, hard negatives derived from a semantic-shift typology, and exact Korean-English translations. Concurrently, we release KorFinSTS, a 1,921-pair Korean financial STS benchmark spanning news, disclosures, research reports, and regulations, designed to expose nuances that general benchmarks miss. When evaluated against seven open-license baselines, NMIXX's multilingual bge-m3 variant achieves Spearman's rho gains of +0.10 on English FinSTS and +0.22 on KorFinSTS, outperforming its pre-adaptation checkpoint and surpassing other models by the largest margin, while revealing a modest trade-off in general STS performance. Our analysis further shows that models with richer Korean token coverage adapt more effectively, underscoring the importance of tokenizer design in low-resource, cross-lingual settings. By making both models and the benchmark publicly available, we provide the community with robust tools for domain-adapted, multilingual representation learning in finance. |
| title | NMIXX: Domain-Adapted Neural Embeddings for Cross-Lingual eXploration of Finance |
| topic | Computation and Language Artificial Intelligence Computational Finance |
| url | https://arxiv.org/abs/2507.09601 |