Abstract
In this work we investigate whether large language models (LLMs) ‘understand’ bridging relations and can use this knowledge effectively. We present the results obtained from two tasks: generation of texts containing bridging and filling in missing bridging spans. We show that in most of the cases LLMs fail to generate bridging in a reliable way.
- Anthology ID:
- 2025.jlcl-2.5
- Volume:
- Journal for Language Technology and Computational Linguistics, Vol. 38 No. 2: LLM Fails - Failed experiments with generative AI and what we can learn from them
- Month:
- July
- Year:
- 2025
- Address:
- Editors:
- Venue:
- [JLCL](https://aclanthology.org/venues/jlcl/)
- SIG:
- Publisher:
- German Society for Computational Linguistics and Language Technology
- Note:
- Pages:
- 77–95
- Language:
- URL:
- [https://aclanthology.org/2025.jlcl-2.5/](https://aclanthology.org/2025.jlcl-2.5/)
- DOI:
- [10.21248/jlcl.38.2025.286](https://doi.org/10.21248/jlcl.38.2025.286)
- Cite (ACL):
- Natalia Skachkova, Simon Ostermann, Josef van Genabith, and Bernd Kiefer. 2025. Do LLMs fail in bridging generation? .Journal for Language Technology and Computational Linguistics , 38(2):77–95.
- Cite (Informal):
- [Do LLMs fail in bridging generation?](https://aclanthology.org/2025.jlcl-2.5/) (Skachkova et al., JLCL 2025)
- PDF:
- [https://aclanthology.org/2025.jlcl-2.5.pdf](https://aclanthology.org/2025.jlcl-2.5.pdf)
source & further reading
aclanthology.org — original article
A Study of Errors in the Output of Large Language Models for Domain-Specific Few-Shot Named Entity Recognition
Can we Operationalize Conceptual Metaphor Cross-Lingually?
Exploring the Limits of LLMs for German Text Classification: Prompting and Fine-tuning Strategies Across Small and Medium-sized Datasets