Papers by Bayram Ayadi
Probing Bias Formation in Medical LLMs through Activation Steering (2026.acl-srw)
Copied to clipboard
| Challenge: | Large Language Models specialized for the medical domain achieve high performance on static benchmarks, but are vulnerable to sycophantic confabulation. |
| Approach: | They propose a framework toward clinical AI systems that are more robust and aligned with expert medical logic. |
| Outcome: | The proposed framework outperforms static global interventions on a medical prompt with cluster-conditioned dynamic steering. |
The Clinical Fingerprint: Comparing the Rhetorical Integrity and Epistemic Safety of Human Physicians and Large Language Models (2026.eacl-srw)
Copied to clipboard
| Challenge: | Large Language Models demonstrate expert proficiency on medical benchmarks, but clinical encounter requires a sophisticated rhetorical performance of care. |
| Approach: | They compare the rhetorical performance of large language models with human physicians . they find that generic models often bury critical advice under layers of linguistic recursion . |
| Outcome: | The proposed models lack the ethical integrity needed to deliver clinical advice, the authors argue . they show that generic models often bury critical advice under layers of complex linguistic recursion . |