aiqbench analysis
This paper introduces EGIP, a protocol designed to standardize how clinical AI agents access and organize diagnostic evidence, thereby isolating true reasoning performance from environmental artifacts. By enforcing strict information boundaries and separating evidence acquisition from final prediction, the method significantly boosts diagnostic accuracy across multiple language models and datasets. The work addresses a critical gap in evaluating medical AI by ensuring that reported results reflect genuine capability rather than biases introduced by inconsistent interaction environments.
EGIP: An Evidence-Grounded Interaction Protocol for Clinical Agents
arXiv:2610.24539v1 Announce Type: new Abstract: Interactive clinical agents acquire evidence through patient dialogue and diagnostic examinations, yet their performance also depends on how the interaction environment exposes information, available actions, and examination results. When these conditions are implicit or inconsistently controlled, final-answer accuracy can conflate diagnostic reasoning with the effects of the interaction protocol. We propose the \textbf{E}vidence-\textbf{G}rounded \textbf{I}nteraction \textbf{P}rotocol (EGIP), which enforces role-specific information boundaries, exposes diagnostic actions without revealing their results, retrieves examination results from stored clinical records, and organizes acquired evidence for final decision-making. By separating evidence acquisition from terminal prediction, EGIP enables different decision strategies to operate on the same observations. We evaluate EGIP across two clinical-agent frameworks, five datasets, and six language models. EGIP improves diagnostic accuracy over native baselines across all primary comparisons. A clinician review further shows that the proportion of automatically scored correct responses jointly judged evidence-supported rises from 10.20\% to 36.55\%. These findings support the effectiveness of EGIP as a unified interaction protocol for improving clinical-agent performance through controlled evidence access and acquisition.
Abstract:Interactive clinical agents acquire evidence through patient dialogue and diagnostic examinations, yet their performance also depends on how the interaction environment exposes information, available actions, and examination results. When these conditions are implicit or inconsistently controlled, final-answer accuracy can conflate diagnostic reasoning with the effects of the interaction protocol. We propose the \textbf{E}vidence-\textbf{G}rounded \textbf{I}nteraction \textbf{P}rotocol (EGIP), which enforces role-specific information boundaries, exposes diagnostic actions without revealing their results, retrieves examination results from stored clinical records, and organizes acquired evidence for final decision-making. By separating evidence acquisition from terminal prediction, EGIP enables different decision strategies to operate on the same observations. We evaluate EGIP across two clinical-agent frameworks, five datasets, and six language models. EGIP improves diagnostic accuracy over native baselines across all primary comparisons. A clinician review further shows that the proportion of automatically scored correct responses jointly judged evidence-supported rises from 10.20\% to 36.55\%. These findings support the effectiveness of EGIP as a unified interaction protocol for improving clinical-agent performance through controlled evidence access and acquisition.
Computation and Language (cs.CL)
arXiv:2610.24539 [cs.CL]
(or arXiv:2610.24539v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2610.24539 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 260 words