Salvato in:
| Autori principali: | , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2507.16809 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908463579267072 |
|---|---|
| author | Lian, Da-Chen Huang, Ri-Sheng Chen, Pin-Er Lim, Chunki Lin, You-Kuan Tseng, Guan-Yu Yang, Zi-Cheng Lin, Zhen-Yu Chen, Pin-Cheng Hsieh, Shu-Kai |
| author_facet | Lian, Da-Chen Huang, Ri-Sheng Chen, Pin-Er Lim, Chunki Lin, You-Kuan Tseng, Guan-Yu Yang, Zi-Cheng Lin, Zhen-Yu Chen, Pin-Cheng Hsieh, Shu-Kai |
| contents | We propose LingBench++, a linguistically-informed benchmark and reasoning framework designed to evaluate large language models (LLMs) on complex linguistic tasks inspired by the International Linguistics Olympiad (IOL). Unlike prior benchmarks that focus solely on final answer accuracy, LingBench++ provides structured reasoning traces, stepwise evaluation protocols, and rich typological metadata across over 90 low-resource and cross-cultural languages. We further develop a multi-agent architecture integrating grammatical knowledge retrieval, tool-augmented reasoning, and deliberate hypothesis testing. Through systematic comparisons of baseline and our proposed agentic models, we demonstrate that models equipped with external knowledge sources and iterative reasoning outperform single-pass approaches in both accuracy and interpretability. LingBench++ offers a comprehensive foundation for advancing linguistically grounded, culturally informed, and cognitively plausible reasoning in LLMs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_16809 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LingBench++: A Linguistically-Informed Benchmark and Reasoning Framework for Multi-Step and Cross-Cultural Inference with LLMs Lian, Da-Chen Huang, Ri-Sheng Chen, Pin-Er Lim, Chunki Lin, You-Kuan Tseng, Guan-Yu Yang, Zi-Cheng Lin, Zhen-Yu Chen, Pin-Cheng Hsieh, Shu-Kai Computation and Language We propose LingBench++, a linguistically-informed benchmark and reasoning framework designed to evaluate large language models (LLMs) on complex linguistic tasks inspired by the International Linguistics Olympiad (IOL). Unlike prior benchmarks that focus solely on final answer accuracy, LingBench++ provides structured reasoning traces, stepwise evaluation protocols, and rich typological metadata across over 90 low-resource and cross-cultural languages. We further develop a multi-agent architecture integrating grammatical knowledge retrieval, tool-augmented reasoning, and deliberate hypothesis testing. Through systematic comparisons of baseline and our proposed agentic models, we demonstrate that models equipped with external knowledge sources and iterative reasoning outperform single-pass approaches in both accuracy and interpretability. LingBench++ offers a comprehensive foundation for advancing linguistically grounded, culturally informed, and cognitively plausible reasoning in LLMs. |
| title | LingBench++: A Linguistically-Informed Benchmark and Reasoning Framework for Multi-Step and Cross-Cultural Inference with LLMs |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2507.16809 |