Text Encoders Lack Knowledge: Leveraging Generative LLMs for Domain-Specific Semantic Textual Similarity
Joseph Gatto, Omar Sharif, Parker Seegmiller, Philip Bohlman, Sarah Preum
Abstract
Amidst the sharp rise in the evaluation of large language models (LLMs) on various tasks, we find that semantic textual similarity (STS) has been under-explored. In this study, we show that STS can be cast as a text generation problem while maintaining strong performance on multiple STS benchmarks. Additionally, we show generative LLMs significantly outperform existing encoder-based STS models when characterizing the semantic similarity between two texts with complex semantic relationships dependent on world knowledge. We validate this claim by evaluating both generative LLMs and existing encoder-based STS models on three newly-collected STS challenge sets which require world knowledge in the domains of Health, Politics, and Sports. All newly-collected data is sourced from social media content posted after May 2023 to ensure the performance of closed-source models like ChatGPT cannot be credited to memorization. Our results show that, on average, generative LLMs outperform the best encoder-only baselines by an average of 22.3% on STS tasks requiring world knowledge. Our results suggest generative language models with STS-specific prompting strategies achieve state-of-the-art performance in complex, domain-specific STS tasks.- Anthology ID:
- 2023.gem-1.23
- Volume:
- Proceedings of the Third Workshop on Natural Language Generation, Evaluation, and Metrics (GEM)
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Sebastian Gehrmann, Alex Wang, João Sedoc, Elizabeth Clark, Kaustubh Dhole, Khyathi Raghavi Chandu, Enrico Santus, Hooman Sedghamiz
- Venues:
- GEM | WS
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 277–288
- Language:
- URL:
- https://aclanthology.org/2023.gem-1.23
- DOI:
- Cite (ACL):
- Joseph Gatto, Omar Sharif, Parker Seegmiller, Philip Bohlman, and Sarah Preum. 2023. Text Encoders Lack Knowledge: Leveraging Generative LLMs for Domain-Specific Semantic Textual Similarity. In Proceedings of the Third Workshop on Natural Language Generation, Evaluation, and Metrics (GEM), pages 277–288, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- Text Encoders Lack Knowledge: Leveraging Generative LLMs for Domain-Specific Semantic Textual Similarity (Gatto et al., GEM-WS 2023)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2023.gem-1.23.pdf