The paper investigates whether domain-specific fine‑tuning benefits open‑ended scientific reasoning in astronomy. Using a curated 300‑question QA benchmark from 2017–2026 Olympiad‑style materials, the authors compare open‑weight, API‑served general‑purpose, multimodal, and astronomy‑specialized language models. Results show that strong general‑purpose models set the highest correctness baseline, but variations in metric agreement, judge sensitivity, benchmark composition, and modality suggest that domain specialization is task‑ and deployment‑dependent and that domain‑specific evaluation is crucial for scientific workflows.
By Vanessa Lama, Sanjay Das, Emily Herron, Yuan-Sen Ting, Tijmen de Haan, Junqi Yin, Tirthankar Ghosal, Feiyi Wang
arXiv:2606. 28446v1 Announce Type: cross Abstract: Light curves describe temporal variations in the brightness of celestial objects.
By Yicheng Rui
arXiv:2606. 00384v1 Announce Type: new Abstract: Fitting quantitative models to data is a central step in scientific workflows, yet it remains one of the least automated.
By William Rudman, Abhishek Divekar, Kanishk Jain, Sebastian Joseph, Stella S. R. Offner, Matthew Lease, Kyle Mahowald, Greg Durrett, Junyi Jessy Li
arXiv:2608. 13267v1 Announce Type: cross Abstract: Existing vision-language model (VLM) benchmarks emphasize perception and reasoning accuracy (how well VLMs describe and reason about what they see in an image), with limited attention to behavioral reliability under uncertainty (how they behave when visual evidence is missing or misleading).
By Paul Osemudiame Oamen, Owusu-Banahene Osei, Ananya Mukherjee, Christian Greisinger, Steffen Eger, Pius Onobhayedo, Wei Zhao
SciFigBench is a new benchmark for evaluating Vision Language Models on scientific method figure understanding. It focuses on three fine‑grained capabilities—dense content perception, visual relation comprehension, and information‑flow tracing—using 4,352 curated figures and 40,670 QA pairs. The authors also introduce SciFigCue, a training‑free visual prompting module that improves performance on several tasks without updating model parameters.
By Lijia Wu, Zheng Li, Yuxuan Liu, Jingwen Zhang, Penghai Zhao, Xiang Li
arXiv:2607. 00228v1 Announce Type: cross Abstract: Modern time-domain surveys such as the Zwicky Transient Facility (ZTF) generate hundreds of thousands of alerts each night, making real-time decisions for follow-up observations a central challenge in time-domain astronomy.
By Ved G. Shah, Nabeel Rehemtulla, Adam A. Miller, Sushant Sharma Chaudhary, Michael W. Coughlin, Antoine Le Calloch, Matthew J. Graham, Joahan Castaneda Jaimes, Theophile Jegou du Laz, Ashish A. Mahabal, Frank J. Masci, Josiah Purdum, Reed Riddle, Jesper Sollerman, Anastasia Wei, Mansi M. Kasliwal
arXiv:2606. 13020v1 Announce Type: new Abstract: Three paradigmatic forms of inference recur across scientific reasoning: deduction, induction, and causal abduction.
By Pierre Beckmann, Marco Valentino, Andre Freitas
arXiv:2609.07102v1 Announce Type: cross
Abstract: Astronomical spectra encode rich physical information, but drawing scientific conclusions from spectral features typically requires expert interpreta...
By Jinghang Shi, Yanxia Zhang, Ali Luo, Changhua Li, Xiao Kong
arXiv:2609.13514v1 Announce Type: new
Abstract: Ensuring the reliability of black-box machine learning models in safety-critical space missions remains a significant challenge, particularly when grou...
By Nikki Grens, Lu\'is F. Sim\~oes, Kai Hou Yip, Theresa Lueftinger
The study audits the AION-1 foundation model, a 39‑modality transformer trained on over 200 million astronomical objects, and finds that its reliance on a survey detection channel—specifically the segmentation map—introduces a severe systematic bias. By keeping image tokens unchanged and editing only the segmentation map, all model outputs (flux, size, ellipticity, redshift) shift by factors of 110–4400 compared to a placebo, revealing that the model’s predictions are driven more by detection gating than by the actual light distribution. This bias propagates into cosmological analyses, shifting tomographic mean redshifts by a median 0.71 × the LSST DESC requirement and exceeding it in multiple assignments, while removing the detection channel eliminates the effect without measurable cost.
whyItMatters":"The bias in the detection channel directly inflates errors in key astronomical measurements, potentially compromising the precision of cosmological studies that rely on accurate redshift estimates."
By Ihor Kendiukhov
The paper introduces a framework for evaluating large language model agents by attempting to end‑to‑end reproduce published astronomy studies, separating execution from verification and distinguishing computational failures from methodological ambiguities. Applying this to fourteen papers—one from The Astrophysical Journal and thirteen from Nature—revealed that eleven contained ambiguities that prevented a uniquely specified reproduction path. In a controlled case study, twelve different analysis paths produced distance estimates ranging from 2.16 to 3.53 kpc, with only one matching the published value of ~2.70 kpc, demonstrating that matching outcomes does not guarantee that the agent has reconstructed the underlying reasoning.
whyItMatters":"The study shows that end‑to‑end reproduction can expose gaps in implicit scientific knowledge within AI systems, highlighting the need for better integration of causal relevance in LLM agents."
By Yuehui Wang, Xinyu Qi, Guirong Xue, Cheng Wang, Yangbin Xie, Xiaoyu Tang, Cong Sun
arXiv:2609.33399v2 Announce Type: replace
Abstract: In realistic education, a solution is often expressed not only in words but in a drawing--a circuit, a geometric construction, a function plot--and...
By Jiali Chen, Zhengteng Lin, Zuqi Wang, Shirong Lin, Xi Yu, Xusen Hei, DingBa Fu, Jiayuan Xie, Yi Cai