Papers by Marc Felix Brinner
SemCSE: Semantic Contrastive Sentence Embeddings Using LLM-Generated Summaries For Scientific Abstracts (2025.emnlp-main)
Copied to clipboard
| Challenge: | SemCSE is an unsupervised method for learning semantic embeddings of scientific texts . |
| Approach: | They propose an unsupervised method for learning semantic embeddings of scientific texts that leverages LLM-generated scientific summaries to train a model that positions semantically related summary closer together in the embeddable space. |
| Outcome: | The proposed method achieves state-of-the-art performance on the SciRepEval benchmark for scientific text embeddings, highlighting the benefits of a semantically focused training approach. |
Enhancing Domain-Specific Encoder Models with LLM-Generated Data: How to Leverage Ontologies, and How to Do Without Them (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a new method for continual pretraining transformer encoder models is proposed for specialized domains with limited training data. |
| Approach: | They propose to use LLM-generated data to enrich domain-specific ontologies and pretrain transformer encoder models as an ontology-informed embedding model for concept definitions. |
| Outcome: | The proposed method improves on standard MLM pretraining on invasion biology domains. |