# Bridging Lexical Divergence: LLM-Assisted, Cost-Efficient, Zero-shot Scientific Entity Linking

arXiv: https://arxiv.org/abs/2609.00228 · HTML: https://mlsanity.com/paper/2609.00228/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 28 references that are in the map.
- First posted on arXiv 31 Aug 2026.

## Builds on (most cited first, 12 of 12)

- [Autoregressive Entity Retrieval](https://mlsanity.com/paper/2010.00904/) (2020)
- [Self-Alignment Pretraining for Biomedical Entity Representations](https://mlsanity.com/paper/2010.11784/) (2020)
- [Zero-Shot Entity Linking by Reading Entity Descriptions](https://mlsanity.com/paper/1906.07348/) (2019)
- [Learning Dense Representations for Entity Retrieval](https://mlsanity.com/paper/1909.10506/) (2019)
- [Biomedical Entity Representations with Synonym Marginalization](https://mlsanity.com/paper/2005.00239/) (2020)
- [Joint Learning of the Embedding of Words and Entities for Named Entity Disambiguation](https://mlsanity.com/paper/1601.01343/) (2016)
- [Generative Biomedical Entity Linking via Knowledge Base-Guided Pre-training and Synonyms-Aware Fine-tuning](https://mlsanity.com/paper/2204.05164/) (2022)
- [LLMAEL: Large Language Models are Good Context Augmenters for Entity Linking](https://mlsanity.com/paper/2407.04020/) (2024)
- [A Lightweight Neural Model for Biomedical Entity Linking](https://mlsanity.com/paper/2012.08844/) (2020)
- [A Read-and-Select Framework for Zero-shot Entity Linking](https://mlsanity.com/paper/2310.12450/) (2023)
- [Learning from Negative Samples in Biomedical Generative Entity Linking](https://mlsanity.com/paper/2408.16493/) (2024)
- [Re-Examine Distantly Supervised NER: A New Benchmark and a Simple Approach](https://mlsanity.com/paper/2402.14948/) (2024)

## Abstract (arXiv)

Scientific domain entity linking (EL) differs from general domain EL because mentions and entity names often lack lexical overlap. Another challenge is that specialized terminology is used in the scientific domain, which is rarely encountered in models pretrained on general domains. Therefore, models trained on general domains transfer poorly to scientific domains. To address this, in-domain fine-tuning is the natural remedy. However, many scientific domains lack expert-annotated data, motivating the need for a zero-human-annotation approach. Existing zero-shot methods heavily rely on LLMs to generate aliases across entire mention corpora, which incurs substantial computational cost, and those methods provide no mechanism to filter out noise from LLMs. To address these challenges, we propose Sci-ZSEL, a framework that selectively generates entity aliases with an LLM to control computational cost, and applies an ontology-aware filter to remove aliases that semantically drift toward ontology neighbors. Then, filtered aliases are used to construct pseudo-labeled mention-entity pairs for fine-tuning. To enable evaluation of EL under low lexical overlap, we also release a new animal science EL benchmark linked to three livestock trait ontologies, where mentions and entities exhibit substantially lower lexical overlap than in existing benchmarks. Across five benchmarks, Sci-ZSEL outperforms the non-fine-tuned baseline, is most useful on nonoverlapping mentions, and combining it with curated synonyms gives the best performance in most settings.