# LLMs are not Zero-Shot Reasoners for Biomedical Information Extraction

arXiv: https://arxiv.org/abs/2408.12249 · HTML: https://mlsanity.com/paper/2408.12249/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning.
- Builds on 25 of its 89 references that are in the map.
- First posted on arXiv 22 Aug 2024.

## Placed in

- LLM reasoning / [Clinical and biomedical applications](https://mlsanity.com/lr/clinical-and-biomedical-applications/)

## Builds on (most cited first, 25 of 25)

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [The Faiss library](https://mlsanity.com/paper/2401.08281/) (2024)
- [Benchmarking Retrieval-Augmented Generation for Medicine](https://mlsanity.com/paper/2402.13178/) (2024)
- [BioMistral: A Collection of Open-Source Pretrained Large Language Models for Medical Domains](https://mlsanity.com/paper/2402.10373/) (2024)
- [MEDITRON-70B: Scaling Medical Pretraining for Large Language Models](https://mlsanity.com/paper/2311.16079/) (2023)
- [Chain-of-Thought Reasoning Without Prompting](https://mlsanity.com/paper/2402.10200/) (2024)
- [MedDialog: Two Large-scale Medical Dialogue Datasets](https://mlsanity.com/paper/2004.03329/) (2020)
- [To Believe or Not to Believe Your LLM](https://mlsanity.com/paper/2406.02543/) (2024)
- [BioMedLM: A 2.7B Parameter Language Model Trained On Biomedical Text](https://mlsanity.com/paper/2403.18421/) (2024)
- [Functional Benchmarks for Robust Evaluation of Reasoning Performance, and the Reasoning Gap](https://mlsanity.com/paper/2402.19450/) (2024)
- [Uncertainty Estimation and Quantification for LLMs: A Simple Supervised Approach](https://mlsanity.com/paper/2404.15993/) (2024)
- [RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Horizon Generation](https://mlsanity.com/paper/2403.05313/) (2024)
- [UniversalNER: Targeted Distillation from Large Language Models for Open Named Entity Recognition](https://mlsanity.com/paper/2308.03279/) (2023)
- [Locate and Label: A Two-stage Identifier for Nested Named Entity Recognition](https://mlsanity.com/paper/2105.06804/) (2021)
- [Empirical Study of Zero-Shot NER with ChatGPT](https://mlsanity.com/paper/2310.10035/) (2023)
- [MedAdapter: Efficient Test-Time Adaptation of Large Language Models towards Medical Reasoning](https://mlsanity.com/paper/2405.03000/) (2024)
- [JMLR: Joint Medical LLM and Retrieval Training for Enhancing Reasoning and Professional Question Answering Capability](https://mlsanity.com/paper/2402.17887/) (2024)
- [OLAPH: Improving Factuality in Biomedical Long-form Question Answering](https://mlsanity.com/paper/2405.12701/) (2024)
- [Improving accuracy of GPT-3/4 results on biomedical data using a retrieval-augmented language model](https://mlsanity.com/paper/2305.17116/) (2023)
- [Me LLaMA: Foundation Large Language Models for Medical Applications](https://mlsanity.com/paper/2402.12749/) (2024)
- [On-the-fly Definition Augmentation of LLMs for Biomedical NER](https://mlsanity.com/paper/2404.00152/) (2024)
- [LLM-based Privacy Data Augmentation Guided by Knowledge Distillation with a Distribution Tutor for Medical Text Classification](https://mlsanity.com/paper/2402.16515/) (2024)
- [MedReadMe: A Systematic Study for Fine-grained Sentence Readability in Medical Domain](https://mlsanity.com/paper/2405.02144/) (2024)
- [How Important is Domain Specificity in Language Models and Instruction Finetuning for Biomedical Relation Extraction?](https://mlsanity.com/paper/2402.13470/) (2024)

## Built on by (most cited first, 9 of 9)

- [TarDiff: Target-Oriented Diffusion Guidance for Synthetic Electronic Health Record Time Series Generation](https://mlsanity.com/paper/2504.17613/) (2025)
- [Evaluating Differentially Private Generation of Domain-Specific Text](https://mlsanity.com/paper/2508.20452/) (2025)
- [Generating Synthetic Data with Formal Privacy Guarantees: State of the Art and the Road Ahead](https://mlsanity.com/paper/2503.20846/) (2025)
- [Clinical trial cohort selection using Large Language Models on n2c2 Challenges](https://mlsanity.com/paper/2501.11114/) (2025)
- [Generating Realistic Multi-Beat ECG Signals](https://mlsanity.com/paper/2505.18189/) (2025)
- [Toward Federated Large Language Models in Medicine: A Parameter-Efficient Framework for Privacy-Preserving, Multi-Institutional Adaptation](https://mlsanity.com/paper/2601.22124/) (2026)
- [RexDrug: Reliable Multi-Drug Combination Extraction through Reasoning-Enhanced LLMs](https://mlsanity.com/paper/2603.08166/) (2026)
- [Beyond Agreement: Scoring Panel-Surfaced Biomedical Entity Candidates for Curator Triage](https://mlsanity.com/paper/2605.30826/) (2026)
- [A Knowledge-Injection Framework for Zero-Shot Adaptation of LLMs to Delirium Prediction](https://mlsanity.com/paper/2607.20453/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) are increasingly adopted for applications in healthcare, reaching the performance of domain experts on tasks such as question answering and document summarisation. Despite their success on these tasks, it is unclear how well LLMs perform on tasks that are traditionally pursued in the biomedical domain, such as structured information extraction. To bridge this gap, in this paper, we systematically benchmark LLM performance in Medical Classification and Named Entity Recognition (NER) tasks. We aim to disentangle the contribution of different factors to the performance, particularly the impact of LLMs' task knowledge and reasoning capabilities, their (parametric) domain knowledge, and addition of external knowledge. To this end, we evaluate various open LLMs - including BioMistral and Llama-2 models - on a diverse set of biomedical datasets, using standard prompting, Chain of-Thought (CoT) and Self Consistency based reasoning as well as Retrieval-Augmented Generation (RAG) with PubMed and Wikipedia corpora. Counter intuitively, our results reveal that standard prompting consistently outperforms more complex techniques across both tasks, laying bare the limitations in the current application of CoT, self-consistency and RAG in the biomedical domain. Our findings suggest that advanced prompting methods developed for knowledge- or reasoning-intensive tasks, such as CoT or RAG, are not easily portable to biomedical tasks where precise structured outputs are required. This highlights the need for more effective integration of external knowledge and reasoning mechanisms in LLMs to enhance their performance in real-world biomedical applications.