Unlocking LLM Safeguards for Low-Resource Languages via Reasoning and Alignment with Minimal Training Data
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911206258769920 |
|---|---|
| author | Chen, Zhuowei Zhang, Bowei Lin, Nankai Hou, Tian Wang, Lianxi |
| author_facet | Chen, Zhuowei Zhang, Bowei Lin, Nankai Hou, Tian Wang, Lianxi |
| contents | Recent advances in LLMs have enhanced AI capabilities, but also increased the risk posed by malicious requests, highlighting the need for effective LLM safeguards to detect such queries. Existing approaches largely rely on classifier-based methods that lack interpretability and perform poorly on low-resource languages. To address these limitations, we propose ConsistentGuard, a novel reasoning-based multilingual safeguard, which enhances explainability via reasoning and boosts knowledge transfer between languages through alignment. With only 1,000 training samples, our method demonstrates superior performance on three datasets across six languages, outperforming larger models trained with significantly more data, and exhibits strong interpretability and generalization ability. We also contribute a multilingual benchmark extension and release our codes to support future research. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_10677 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Unlocking LLM Safeguards for Low-Resource Languages via Reasoning and Alignment with Minimal Training Data Chen, Zhuowei Zhang, Bowei Lin, Nankai Hou, Tian Wang, Lianxi Computation and Language Recent advances in LLMs have enhanced AI capabilities, but also increased the risk posed by malicious requests, highlighting the need for effective LLM safeguards to detect such queries. Existing approaches largely rely on classifier-based methods that lack interpretability and perform poorly on low-resource languages. To address these limitations, we propose ConsistentGuard, a novel reasoning-based multilingual safeguard, which enhances explainability via reasoning and boosts knowledge transfer between languages through alignment. With only 1,000 training samples, our method demonstrates superior performance on three datasets across six languages, outperforming larger models trained with significantly more data, and exhibits strong interpretability and generalization ability. We also contribute a multilingual benchmark extension and release our codes to support future research. |
| title | Unlocking LLM Safeguards for Low-Resource Languages via Reasoning and Alignment with Minimal Training Data |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2510.10677 |