CoolSci 2026: Presentations from STScI Postdocs

Value of the Information & Strength of the Argument

The presentations offer substantial value by introducing novel computational tools and statistical frameworks for astronomical data analysis. Fernandez’s ASPECT algorithm addresses the need for automated spectral feature detection in large surveys, with a clear demonstration of its utility. Kim’s Bayesian pipeline tackles the challenges of UV absorption spectroscopy, providing a rigorous approach to component fitting and model comparison. Gibson’s framework for analyzing stellar populations in nearby galaxies enables detailed studies of galactic structure and evolution. Soltis’s foundation model for Gaia data represents a promising step toward leveraging large datasets for diverse astronomical tasks. The arguments are well-supported by technical details and references to ongoing research, though some presentations are more advanced than others. The overall scientific value is high, with each talk contributing to the advancement of data analysis techniques in astronomy.

Scientific Rigor, Source Quality, Title Accuracy

The scientific rigor is generally high, with presenters referencing specific surveys, instruments, and their own publications. Fernandez mentions the CEERS and CAPERS surveys and the LiMe package, while Kim references UV spectroscopy techniques and Bayesian inference. Gibson discusses APOGEE and Hubble/JWST data, and Soltis mentions Gaia DR3. The sources are credible and relevant, though the video does not provide detailed citations for all claims. The title accurately reflects the content, as it is a compilation of postdoc presentations. The presentations are well-structured and technically sound, with appropriate caveats and future directions. However, the lack of peer-reviewed publication details for some works slightly reduces the overall rigor.

The title accurately reflects the content: a series of presentations by STScI postdocs at the CoolSci 2026 event.

8/10

The video presents four research talks by STScI postdocs, each detailing methodological developments in astronomical data analysis. The content is technical and grounded in ongoing research, with references to specific surveys and instruments. The presentations are clear and well-structured, though the video is a recording of a live colloquium with minor audio issues. The scientific rigor is high, but the lack of peer-reviewed publication details for some works limits the score.

Markers derived by PSI from the transcript: the creator did not define chapters.

• 02:00Introduction to the colloquium and first speaker Vital Fernandez.
• 05:00Fernandez explains the motivation for ASPECT: identifying spectral features and managing bad data.
• 10:00Demonstration of ASPECT on a spectrum of NGC 62A, showing classification of continuum, lines, and cosmic rays.
• 15:00Fernandez discusses the training of the machine learning model and the parameter space for line features.
• 20:00Implementation of ASPECT on real data, including redshift measurement and line flux estimation.
• 25:00Q&A session for Fernandez's talk.
• 30:00Introduction of second speaker Doyeon Avery Kim and her talk on UV absorption spectroscopy.
• 35:00Kim explains the challenges of component fitting and the need for a scalable Bayesian pipeline.
• 40:00Kim presents the dynamic nested sampling approach and its application to iron absorption lines.
• 45:00Q&A for Kim's talk, discussing prior choices and model comparison.
• 50:00Introduction of third speaker Benjamin Gibson and his talk on stellar populations in nearby galaxies.
• 55:00Gibson discusses his framework for analyzing integrated light spectra and its application to M31 and other galaxies.
• 1:00:00Introduction of fourth speaker John Soltis and his talk on a Gaia foundation model.

• CEERS survey (Finkelstein et al 2024) — Mentioned by Fernandez as a survey where ASPECT was applied.
• CAPERS survey (Dickinson et al 2025) — Mentioned by Fernandez as a survey where ASPECT was applied.
• LiMe package (Fernández et al 2024) — Fernandez discusses integrating ASPECT into the LiMe package for line analysis.
• APOGEE survey — Gibson mentions using high-resolution integrated light spectra from APOGEE for M31 analysis.
• Gaia Data Release 3 — Soltis discusses developing a foundation model using Gaia DR3 data.

• CEERS survey (Finkelstein et al 2024) — Fernandez's application of ASPECT to CEERS data aligns with the survey's goals of studying early galaxies.
• CAPERS survey (Dickinson et al 2025) — Similar to CEERS, CAPERS provides data for ASPECT validation.