Jimeng Sun

Nov 18, 2024 | nature.com | Qiao Jin |Charalampos S Floudas |Fangyuan Chen |Yifan Yang |Jimeng Sun |Changlin Gong

AbstractPatient recruitment is challenging for clinical trials. We introduce TrialGPT, an end-to-end framework for zero-shot patient-to-trial matching with large language models. TrialGPT comprises three modules: it first performs large-scale filtering to retrieve candidate trials (TrialGPT-Retrieval); then predicts criterion-level patient eligibility (TrialGPT-Matching); and finally generates trial-level scores (TrialGPT-Ranking).

Nov 17, 2024 | flipboard.com | Qiao Jin |Zifeng Wang |Charalampos S Floudas |Fangyuan Chen |Changlin Gong |Dara Bracken-Clarke | +4 more

4 hours agoHas AI Killed Graphic Design Jobs? The rapid advancement of generative AI technologies has undeniably transformed various sectors, including the field of graphic design.

Nov 21, 2023 | nature.com | Jimeng Sun

Correction to: Nature Communications https://doi.org/10.1038/s41467-023-41093-0, published online 31 August 2023The original version of this article contained errors in Table 3 and in equations 7 and 8. In the first column of Table 3, the text ‘er’ appears before HALO-Coarse; this should only read HALO-Coarse. The error equation 7 only appears in the PDF version, and not in the HTML version; it incorrectly reads and it should read In equation 8, there is a missing comma after the first colon.

Oct 18, 2023 | arxiv.org | Zhen Yuan Lin |Jimeng Sun

[Submitted on 18 Oct 2023] Title:Conformal Drug Property Prediction with Density Estimation under Covariate Shift Download a PDF of the paper titled Conformal Drug Property Prediction with Density Estimation under Covariate Shift, by Siddhartha Laghuvarapu and Zhen Lin and Jimeng Sun Download PDF Abstract:In drug discovery, it is vital to confirm the predictions of pharmaceutical properties from computational models using costly wet-lab experiments.

Aug 31, 2023 | nature.com | Jimeng Sun

AbstractSynthetic electronic health records (EHRs) that are both realistic and privacy-preserving offer alternatives to real EHRs for machine learning (ML) and statistical analysis. However, generating high-fidelity EHR data in its original, high-dimensional form poses challenges for existing methods.