Nobody Understands Me, Statistically Speaking

Line-by-line explanation of why many arguments against the use of LLMs in psychology are NOT plausible.

Have no resources to use the commercial models? Use OSS! You can even launch them on your laptop. And they ARE good.

An important conclusion and point from this research is that you don't need a frontier model to achieve acceptable results!

Has some nice theoretical stuff.

An attempt to defend the usage of LLMs. A good attempt but it's too polite and does not focus on the real gaps in Jowsey's arguments.

Debates, debates, debates... However, how strong are the arguments? (spoiler: nope; why? because the authors do not realize what LLMs really is and how to use it...)

Grounded theory application tutorial, though the methods could be better.

If you read one methods paper before coding with an LLM, read this one. It takes both the hermeneutics and the engineering seriously — a rare combination.

Note the fact that the models are outdated. Newer ones can narrow the gap.

Quantifies coding results

Machine Learning for Small Bodies in the Solar System is the first book devoted entirely to the application of artificial intelligence and machine learning to the study of asteroids, comets, and trans-Neptunian objects. Covering topics from celestial mechanics and asteroid family identification to object detection, spectroscopy, and autonomous data analysis, it combines methodological foundations with practical implementations, including code examples and publicly available repositories. The volume is intended both as an introduction for researchers entering the field and as a reference for experienced planetary scientists seeking to incorporate modern AI techniques into their research. As the era of data-intensive surveys led by facilities such as the Vera C. Rubin Observatory begins, the methods presented in this book provide an essential foundation for the next generation of solar system science.

This paper demonstrates the transformative potential of the Vera C. Rubin Observatory for the discovery of long-period and hyperbolic comets, showing that next-generation surveys could dramatically increase detection rates. Beyond its scientific results, it provides valuable insight into how modern survey capabilities and data-driven methods will reshape studies of the distant Solar System. An important reference for researchers interested in the future of small-body discovery and survey science.

A landmark paper in autonomous spacecraft operations, this work demonstrated that AI planning and execution systems could reliably control a spacecraft in a real mission environment. By validating the Remote Agent architecture beyond laboratory experiments, it established a foundation for autonomous space exploration and remains an essential reference for researchers interested in AI-driven mission planning, spacecraft autonomy, and intelligent control.

An innovative application of AI to spacecraft guidance and control, this work introduces an adaptive control strategy for stable hovering around asteroids with uncertain gravity. It highlights the growing role of intelligent algorithms in enabling autonomous proximity operations, inspection, and future sampling and landing missions.

One of the early applications of machine learning to the dynamical classification of trans-Neptunian objects, this work illustrates how AI can streamline the analysis of complex orbital behavior. Its methodology remains relevant as modern surveys continue to expand the known population of distant Solar System bodies, making automated classification increasingly important.

This work establishes the first comprehensive benchmark for evaluating multimodal large language models on resonance classification in celestial mechanics. Its standardized datasets, evaluation methodology, and comparison of commercial and open-source models provide a foundation for future research on generative AI in dynamical astronomy. Essential reading for anyone interested in the intersection of large language models and Solar System dynamics.

Yep, that's the point: no training, no code, just use. Similar to Smirnov (2024).

Bla-bla-bla. I'd suggest to start with not cheating on the data in psychology...

Prompt engineering was important... a few years ago. Today it's better to focus on the methods that can overcome this limitation.

Wrong models, wrong method, wrong results...