What's in a Name? Morphological Shortcuts by LLMs in Pharmacology

• URL: http://arxiv.org/abs/2606.05616v1
• Date: Thu, 04 Jun 2026 02:40:42 GMT
• Title: What's in a Name? Morphological Shortcuts by LLMs in Pharmacology
• Abstract summary: We present a behavioral and mechanistic study of "affixs" in pharmacology.<n>We show that affix signals alone elicit class-level pharmacological responses.<n>We introduce a framework for identifying whether a model's drug semantics are driven mainly by the affix, the stem, or the drug name as a whole.
• Score: 36.7297406089205
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The morphological form of a word can often give cues to its meaning, but purely relying on these mappings can lead to overgeneralization in high-stakes domains. In the medical domain, for instance, LLMs can confidently reason about fictitious drugs from their affixes alone (e.g., wugcillin) and generate plausible-looking clinical content. We present a behavioral and mechanistic study of LLM "affix heuristics" in pharmacology. Using fictitious drug names built from real affixes, we show that affix signals alone elicit class-level pharmacological responses. We introduce a framework for identifying whether a model's drug semantics are driven mainly by the affix, the stem, or the drug name as a whole. Applied across 653 drugs, our framework reveals that models often induce drug meaning primarily through affix cues, yet rarely explicitly indicate this reliance, and sometimes incorrectly conflate properties among affix-sharing drugs. Activation patching across models further localizes this behavior to early-mid layers. These findings show that morphological shortcuts pose a subtle but measurable risk to safety.

Related papers

• DrugReason: Dynamic Multi-View Reasoning over Knowledge Graph and Language Evidence for Drug Repurposing [55.45112712244231]
  DrugReason is a multi-view reasoning framework that integrates grounded KG reasoning with LLM-generated mechanistic inference for drug repurposing.<n> Experiments on PharmaDB, DDInter, and DrugBank show that DrugReason improves average performance over strong single-view reasoning baselines.
  arXiv  Detail & Related papers  (2026-09-06T18:54:26Z)
• Tracing Pharmacological Knowledge In Large Language Models [0.0]
  We investigate how drug-group semantics are represented and retrieved within large language models.<n>Early layers play a key role in encoding drug-group knowledge.<n>Drug-group semantics are distributed across tokens and are already present in the embedding space.
  arXiv  Detail & Related papers  (2026-03-03T17:04:08Z)
• Do "New Snow Tablets" Contain Snow? Large Language Models Over-Rely on Names to Identify Ingredients of Chinese Drugs [79.00288739947406]
  Traditional Chinese Medicine (TCM) has seen increasing adoption in healthcare, with specialized Large Language Models (LLMs) emerging to support clinical applications.<n>A fundamental requirement for these models is accurate identification of TCM drug ingredients.<n>Our systematic analysis reveals consistent failure patterns: models often interpret drug names literally, overuse common herbs regardless of relevance, and exhibit erratic behaviors when faced with unfamiliar formulations.
  arXiv  Detail & Related papers  (2025-04-03T17:43:45Z)
• Trust Me, I'm Wrong: LLMs Hallucinate with Certainty Despite Knowing the Answer [51.7407540261676]
  We investigate a distinct type of hallucination, where a model can consistently answer a question correctly, but a seemingly trivial perturbation causes it to produce a hallucinated response with high certainty.<n>This phenomenon is particularly concerning in high-stakes domains such as medicine or law, where model certainty is often used as a proxy for reliability.<n>We show that CHOKE examples are consistent across prompts, occur in different models and datasets, and are fundamentally distinct from other hallucinations.
  arXiv  Detail & Related papers  (2025-02-18T15:46:31Z)
• DrugCLIP: Contrastive Drug-Disease Interaction For Drug Repurposing [4.969453745531116]
  DrugCLIP is a contrastive learning method to learn drug and disease's interaction without negative labels. We have curated a drug repurposing dataset based on real-world clinical trial records.
  arXiv  Detail & Related papers  (2024-07-02T13:41:59Z)
• Emerging Opportunities of Using Large Language Models for Translation Between Drug Molecules and Indications [6.832024637226738]
  We propose a new task, which is the translation between drug molecules and corresponding indications. The creation of molecules from indications, or vice versa, will allow for more efficient targeting of diseases.
  arXiv  Detail & Related papers  (2024-02-14T21:33:13Z)
• Large Language Model Distilling Medication Recommendation Model [58.94186280631342]
  We harness the powerful semantic comprehension and input-agnostic characteristics of Large Language Models (LLMs)<n>Our research aims to transform existing medication recommendation methodologies using LLMs.<n>To mitigate this, we have developed a feature-level knowledge distillation technique, which transfers the LLM's proficiency to a more compact model.
  arXiv  Detail & Related papers  (2024-02-05T08:25:22Z)
• Clinical Named Entity Recognition using Contextualized Token Representations [49.036805795072645]
  This paper introduces the technique of contextualized word embedding to better capture the semantic meaning of each word based on its context. We pre-train two deep contextualized language models, Clinical Embeddings from Language Model (C-ELMo) and Clinical Contextual String Embeddings (C-Flair) Explicit experiments show that our models gain dramatic improvements compared to both static word embeddings and domain-generic language models.
  arXiv  Detail & Related papers  (2021-06-23T18:12:58Z)
• Modeling Pharmacological Effects with Multi-Relation Unsupervised Graph Embedding [4.999039245939572]
  We present a method based on a multi-relation unsupervised graph embedding model that learns latent representations for drugs and diseases. Once representations for drugs and diseases are obtained we learn the likelihood of new links (that is, new indications) between drugs and diseases. Known drug indications are used for learning a model that predicts potential indications.
  arXiv  Detail & Related papers  (2020-04-30T14:51:25Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.