Job Summary
We are seeking a highly motivated Research Associate in NLP, Mechanistic Interpretability and Trustworthy Generative AI. This role offers an exciting opportunity to contribute to AIRIS: Mechanism-Informed Generative AI for Causal and Dynamical Modelling in Multimodal Biomedical Research, a Horizon Europe project developing mechanism-informed generative AI for biomedical research. The post holder will be based in the Department of Computer Science and the National Centre for Text Mining.
Responsibilities
- Develop biomedical NLP methods for mechanism extraction
- Design and implement methods for extracting biomedical entities, events, relations, causal claims and mechanistic pathways from scientific literature.
- Build pipelines that link textual evidence to biomedical ontologies, knowledge graphs, causal models and semantic representations.
- Develop methods to assess AI-generated hypotheses for plausibility, novelty, evidence support, contradiction, uncertainty and explainability.
- Publish in leading NLP, AI and biomedical informatics venues including ACL, EMNLP, NAACL, NeurIPS, ICLR, AAAI, AMIA and relevant biomedical journals and present research at project meetings, conferences and stakeholder workshops.
About You
We encourage applications from individuals with a wide range of backgrounds and experiences. You should demonstrate:
Essential Criteria
- A PhD in Computer Science, specialising in Artificial Intelligence, Natural Language Processing
- Strong research background in NLP, LLMs.
- Experience with mechanistic interpretability, explainable AI, causal representation learning, model probing, attribution methods or faithfulness evaluation.
- Strong programming skills in Python and experience with modern machine learning frameworks such as PyTorch, JAX or TensorFlow.
- Experience with bias and fairness evaluation in NLP, LLMs, biomedical AI or clinical prediction systems.
Desirable Criteria
- Experience in biomedical NLP, clinical NLP, biomedical literature mining, event extraction, relation extraction, pathway extraction or causal knowledge extraction.
- Knowledge of factuality evaluation, hallucination detection, claim verification, evidence retrieval or scientific hypothesis evaluation.
- Experience with graph‑based methods, knowledge graphs, Graph‑RAG, semantic search or neuro‑symbolic AI.
- Familiarity with multimodal biomedical data, including clinical records, omics, imaging reports, patient‑reported outcomes or longitudinal health data.
- Interest in trustworthy AI, responsible AI, AI safety, research integrity and translational biomedical applications.
Benefits
- Generous employer contribution pension
- 29 days annual leave plus bank holidays, along with Christmas closure
- Ride to work and EV car scheme available
All vacancies are open to all candidates regardless of immigration status.
Applications close at midnight on the closing date.
Research Associate In Nlp, Mechanistic Interpretability And Trustworthy Generative Ai in Manchester employer: The University of Manchester
The University of Manchester is an exceptional employer, offering a dynamic work culture that prioritises innovation and collaboration. With a strong commitment to employee growth, you will have access to comprehensive benefits including leading pension schemes and health services, all while contributing to the strategic direction of a prestigious institution in a vibrant city known for its rich cultural heritage and academic excellence.
Contact Details:
The University of Manchester Recruitment Team
