EpiBench: Can LLMs Understand Epitopes for Antibody Drug Discovery?
Summary
EpiBench is a new closed-book, sequence-based benchmark for evaluating how well LLMs understand epitopes across five antibody-drug-discovery tasks, finding that current models capture partial signals but struggle with antibody-specific reasoning.
View Cached Full Text
Cached at: 08/07/26, 07:52 AM
# EpiBench: Can LLMs Understand Epitopes for Antibody Drug Discovery? Source: [https://arxiv.org/abs/2608.06022](https://arxiv.org/abs/2608.06022) [View PDF](https://arxiv.org/pdf/2608.06022) > Abstract:Epitopes determine where antibodies bind antigens and shape downstream therapeutic properties such as functional blockade and escape resistance, making epitope understanding central to antibody drug discovery\. Although large language models \(LLMs\) have shown strong biomedical reasoning ability, it remains unclear whether they can infer epitope information directly from antigen and antibody sequences\. Existing epitope resources typically focus on isolated prediction tasks or rely on specialized structural settings, while general protein benchmarks do not evaluate epitope\-centered decisions across the antibody development workflow\. To address this gap, we introduce EpiBench, a closed\-book, sequence\-based, and automatically scorable benchmark for evaluating epitope reasoning in LLMs\. EpiBench contains 1,609 curated samples grounded in structural antibody\-\-antigen contacts, curated functional B\-cell assays, and deep mutational scanning escape measurements\. It covers five connected tasks: targetable region discovery, antibody\-conditioned epitope identification, epitope binning, functional epitope assessment, and antibody escape assessment, with controlled sampling to reduce shortcut\-based evaluation artifacts\. We evaluate nine general\-purpose LLMs and analyze their behavior through task\-specific baselines, antigen length stratification, explicit\-reasoning comparison, and failure\-mode inspection\. The results show that current LLMs capture partial epitope\-related signals but remain limited in antibody\-specific sequence grounding, long\-context residue localization, and biologically grounded reasoning\. Therefore, EpiBench provides a diagnostic testbed for measuring and improving sequence\-aware biomedical LLMs toward reliable LLM\-assisted antibody discovery\. ## Submission history From: Jiaqi Wang \[[view email](https://arxiv.org/show-email/d9b8b306/2608.06022)\] **\[v1\]**Thu, 6 Aug 2026 13:29:53 UTC \(1,307 KB\)
Similar Articles
EHRBench: An Automated and Reliable EHR-based Benchmark for Clinical Decision Making with LLMs
EHRBench is an automated and reliable benchmark for evaluating LLMs on clinical decision-making tasks using real-world electronic health records, covering nearly 1M QA items across diagnosis, treatment, and prognosis tasks.
Measuring Epistemic Resilience of LLMs Under Misleading Medical Context
Introduces MedMisBench to measure LLMs' ability to maintain correct medical reasoning under misleading context. Shows that accuracy drops sharply from 71.1% to 38.0% under adversarial conditions, with potential harm flagged by clinical panel.
Are LLMs Ready for Scientific Discovery? A Capability-Oriented Benchmark for AI Scientists
Introduces SDABench, a benchmark evaluating LLMs on six scientific analysis capabilities across five domains, finding models struggle with tasks requiring assumption selection and mechanistic reasoning.
Knowledge before Reasoning: EC-Reason-Bench, a Training-Free Diagnostic Benchmark for LLM Enzyme Classification
This paper introduces EC-Reason-Bench, a training-free diagnostic benchmark to analyze why general LLMs fail on enzyme EC number prediction. It finds that external knowledge is decisive and must precede reasoning, and that reasoning over evidence acts as an arbiter of conflicting nearest neighbors rather than a source of new knowledge.
Before the Action: Benchmarking LLMs on Prospective Hypothesis Discovery
The paper introduces HypoArena, a benchmark for evaluating LLMs' ability to proactively construct hypothesis spaces from incomplete evidence, and experiments on 15 frontier LLMs reveal capability stratification.