Alberto Accomazzi

Tuan Dung Nguyen, Yuan-Sen Ting, Ioana Ciucă et al.

Large language models excel in many human-language tasks but often falter in highly specialized domains like scholarly astronomy. To bridge this gap, we introduce AstroLLaMA, a 7-billion-parameter model fine-tuned from LLaMA-2 using over 300,000 astronomy abstracts from arXiv. Optimized for traditional causal language modeling, AstroLLaMA achieves a 30% lower perplexity than Llama-2, showing marked domain adaptation. Our model generates more insightful and scientifically relevant text completions and embedding extraction than state-of-the-arts foundation models despite having significantly fewer parameters. AstroLLaMA serves as a robust, domain-specific model with broad fine-tuning potential. Its public release aims to spur astronomy-focused research, including automatic paper summarization and conversational agent development.

Felix Grezes, Thomas Allen, Sergi Blanco-Cuaresma et al. · cambridge, harvard

The NASA Astrophysics Data System (ADS) is an essential tool for researchers that allows them to explore the astronomy and astrophysics scientific literature, but it has yet to exploit recent advances in natural language processing. At ADASS 2021, we introduced astroBERT, a machine learning language model tailored to the text used in astronomy papers in ADS. In this work we: - announce the first public release of the astroBERT language model; - show how astroBERT improves over existing public language models on astrophysics specific tasks; - and detail how ADS plans to harness the unique structure of scientific papers, the citation graph and citation context, to further improve astroBERT.

Yuan-Sen Ting, Tuan Dung Nguyen, Tirthankar Ghosal et al.

We present a comprehensive evaluation of proprietary and open-weights large language models using the first astronomy-specific benchmarking dataset. This dataset comprises 4,425 multiple-choice questions curated from the Annual Review of Astronomy and Astrophysics, covering a broad range of astrophysical topics. Our analysis examines model performance across various astronomical subfields and assesses response calibration, crucial for potential deployment in research environments. Claude-3.5-Sonnet outperforms competitors by up to 4.6 percentage points, achieving 85.0% accuracy. For proprietary models, we observed a universal reduction in cost every 3-to-12 months to achieve similar score in this particular astronomy benchmark. open-weights models have rapidly improved, with LLaMA-3-70b (80.6%) and Qwen-2-72b (77.7%) now competing with some of the best proprietary models. We identify performance variations across topics, with non-English-focused models generally struggling more in exoplanet-related fields, stellar astrophysics, and instrumentation related questions. These challenges likely stem from less abundant training data, limited historical context, and rapid recent developments in these areas. This pattern is observed across both open-weights and proprietary models, with regional dependencies evident, highlighting the impact of training data diversity on model performance in specialized scientific domains. Top-performing models demonstrate well-calibrated confidence, with correlations above 0.9 between confidence and correctness, though they tend to be slightly underconfident. The development for fast, low-cost inference of open-weights models presents new opportunities for affordable deployment in astronomy. The rapid progress observed suggests that LLM-driven research in astronomy may become feasible in the near future.

Jinchu Li, Yuan-Sen Ting, Alberto Accomazzi et al.

We construct a concept-object knowledge graph from the full astro-ph corpus through July 2025. Using an automated pipeline, we extract named astrophysical objects from OCR-processed papers, resolve them to SIMBAD identifiers, and link them to scientific concepts annotated in the source corpus. We then test whether historical graph structure can forecast new concept-object associations before they appear in print. Because the concepts are derived from clustering and therefore overlap semantically, we apply an inference-time concept-similarity smoothing step uniformly to all methods. Across four temporal cutoffs on a physically meaningful subset of concepts, an implicit-feedback matrix factorization model (alternating least squares, ALS) with smoothing outperforms the strongest neighborhood baseline (KNN using text-embedding concept similarity) by 16.8% on NDCG@100 (0.144 vs 0.123) and 19.8% on Recall@100 (0.175 vs 0.146), and exceeds the best recency heuristic by 96% and 88%, respectively. These results indicate that historical literature encodes predictive structure not captured by global heuristics or local neighborhood voting, suggesting a path toward tools that could help triage follow-up targets for scarce telescope time.

Rui Pan, Tuan Dung Nguyen, Hardik Arora et al.

Continual pretraining of large language models on domain-specific data has been proposed to enhance performance on downstream tasks. In astronomy, the previous absence of astronomy-focused benchmarks has hindered objective evaluation of these specialized LLM models. Leveraging a recent initiative to curate high-quality astronomical MCQs, this study aims to quantitatively assess specialized LLMs in astronomy. We find that the previously released AstroLLaMA series, based on LLaMA-2-7B, underperforms compared to the base model. We demonstrate that this performance degradation can be partially mitigated by utilizing high-quality data for continual pretraining, such as summarized text from arXiv. Despite the observed catastrophic forgetting in smaller models, our results indicate that continual pretraining on the 70B model can yield significant improvements. However, the current supervised fine-tuning dataset still constrains the performance of instruct models. In conjunction with this study, we introduce a new set of models, AstroLLaMA-3-8B and AstroLLaMA-2-70B, building upon the previous AstroLLaMA series.

Sergi Blanco-Cuaresma, Ioana Ciucă, Alberto Accomazzi et al. · cambridge, harvard

Open-source Large Language Models enable projects such as NASA SciX (i.e., NASA ADS) to think out of the box and try alternative approaches for information retrieval and data augmentation, while respecting data copyright and users' privacy. However, when large language models are directly prompted with questions without any context, they are prone to hallucination. At NASA SciX we have developed an experiment where we created semantic vectors for our large collection of abstracts and full-text content, and we designed a prompt system to ask questions using contextual chunks from our system. Based on a non-systematic human evaluation, the experiment shows a lower degree of hallucination and better responses when using Retrieval Augmented Generation. Further exploration is required to design new features and data augmentation processes at NASA SciX that leverages this technology while respecting the high level of trust and quality that the project holds.

Bishwaranjan Bhattacharjee, Aashka Trivedi, Masayasu Muraoka et al.

Large language models (LLMs) trained on general domain corpora showed remarkable results on natural language processing (NLP) tasks. However, previous research demonstrated LLMs trained using domain-focused corpora perform better on specialized tasks. Inspired by this insight, we developed INDUS, a comprehensive suite of LLMs tailored for the closely-related domains of Earth science, biology, physics, heliophysics, planetary sciences and astrophysics, and trained using curated scientific corpora drawn from diverse data sources. The suite of models include: (1) an encoder model trained using domain-specific vocabulary and corpora to address NLP tasks, (2) a contrastive-learning based text embedding model trained using a diverse set of datasets to address information retrieval tasks and (3) smaller versions of these models created using knowledge distillation for applications which have latency or resource constraints. We also created three new scientific benchmark datasets, CLIMATE-CHANGE NER (entity-recognition), NASA-QA (extractive QA) and NASA-IR (IR) to accelerate research in these multi-disciplinary fields. We show that our models outperform both general-purpose (RoBERTa) and domain-specific (SCIBERT) encoders on these new tasks as well as existing tasks in the domains of interest. Furthermore, we demonstrate the use of these models in two industrial settings -- as a retrieval model for large-scale vector search applications and in automatic content tagging systems.

Tijmen de Haan, Yuan-Sen Ting, Tirthankar Ghosal et al.

General-purpose large language models, despite their broad capabilities, often struggle with specialized domain knowledge, a limitation particularly pronounced in more accessible, lower-parameter versions. This gap hinders their deployment as effective agents in demanding fields such as astronomy. Building on our prior work with AstroSage-8B, this study introduces AstroSage-70B, a significantly larger and more advanced domain-specialized natural-language AI assistant. It is designed for research and education across astronomy, astrophysics, space science, astroparticle physics, cosmology, and astronomical instrumentation. Developed from the Llama-3.1-70B foundation, AstroSage-70B underwent extensive continued pre-training on a vast corpus of astronomical literature, followed by supervised fine-tuning and model merging. Beyond its 70-billion parameter scale, this model incorporates refined datasets, judiciously chosen learning hyperparameters, and improved training procedures, achieving state-of-the-art performance on complex astronomical tasks. Notably, we integrated reasoning chains into the SFT dataset, enabling AstroSage-70B to either answer the user query immediately, or first emit a human-readable thought process. Evaluated on the AstroMLab-1 benchmark -- comprising 4,425 questions from literature withheld during training -- AstroSage-70B achieves state-of-the-art performance. It surpasses all other tested open-weight and proprietary models, including leading systems like o3, Gemini-2.5-Pro, Claude-3.7-Sonnet, Deepseek-R1, and Qwen-3-235B, even those with API costs two orders of magnitude higher. This work demonstrates that domain specialization, when applied to large-scale models, can enable them to outperform generalist counterparts in specialized knowledge areas like astronomy, thereby advancing the frontier of AI capabilities in the field.

Atilla Kaan Alkan, Felix Grezes, Jennifer Lynn Bartlett et al.

We present our participation in the SOMD 2026 shared task on cross-document software mention coreference resolution, where our systems ranked second across all three subtasks. We compare two fine-tuning-free approaches: Fuzzy Matching (FM), a lexical string-similarity method, and Context Aware Representations (CAR), which combines mention-level and document-level embeddings. Both achieve competitive performance across all subtasks (CoNLL F1 of 0.94-0.96), with CAR consistently outperforming FM by 1 point on the official test set, consistent with the high surface regularity of software names, which reduces the need for complex semantic reasoning. A controlled noise-injection study reveals complementary failure modes: as boundary noise increases, CAR loses only 0.07 F1 points from clean to fully corrupted input, compared to 0.20 for FM, whereas under mention substitution, FM degrades more gracefully (0.52 vs. 0.63). Our inference-time analysis shows that FM scales superlinearly with corpus size, whereas CAR scales approximately linearly, making CAR the more efficient choice at large scale. These findings suggest that system selection should be informed by both the noise profile of the upstream mention detector and the scale of the target corpus. We release our code to support future work on this underexplored task.

Atilla Kaan Alkan, Felix Grezes, Sergi Blanco-Cuaresma et al.

Scientific multi-label text classification suffers from extreme class imbalance, where specialized terminology exhibits severe power-law distributions that challenge standard classification approaches. Existing scientific corpora lack comprehensive controlled vocabularies, focusing instead on broad categories and limiting systematic study of extreme imbalance. We introduce AstroConcepts, a corpus of English abstracts from 21,702 published astrophysics papers, labeled with 2,367 concepts from the Unified Astronomy Thesaurus. The corpus exhibits severe label imbalance, with 76% of concepts having fewer than 50 training examples. By releasing this resource, we enable systematic study of extreme class imbalance in scientific domains and establish strong baselines across traditional, neural, and vocabulary-constrained LLM methods. Our evaluation reveals three key patterns that provide new insights into scientific text classification. First, vocabulary-constrained LLMs achieve competitive performance relative to domain-adapted models in astrophysics classification, suggesting a potential for parameter-efficient approaches. Second, domain adaptation yields relatively larger improvements for rare, specialized terminology, although absolute performance remains limited across all methods. Third, we propose frequency-stratified evaluation to reveal performance patterns that are hidden by aggregate scores, thereby making robustness assessment central to scientific multi-label evaluation. These results offer actionable insights for scientific NLP and establish benchmarks for research on extreme imbalance.

Golnaz Shapurian, Michael J Kurtz, Alberto Accomazzi

The automatic identification of planetary feature names in astronomy publications presents numerous challenges. These features include craters, defined as roughly circular depressions resulting from impact or volcanic activity; dorsas, which are elongate raised structures or wrinkle ridges; and lacus, small irregular patches of dark, smooth material on the Moon, referred to as "lake" (Planetary Names Working Group, n.d.). Many feature names overlap with places or people's names that they are named after, for example, Syria, Tempe, Einstein, and Sagan, to name a few (U.S. Geological Survey, n.d.). Some feature names have been used in many contexts, for instance, Apollo, which can refer to mission, program, sample, astronaut, seismic, seismometers, core, era, data, collection, instrument, and station, in addition to the crater on the Moon. Some feature names can appear in the text as adjectives, like the lunar craters Black, Green, and White. Some feature names in other contexts serve as directions, like craters West and South on the Moon. Additionally, some features share identical names across different celestial bodies, requiring disambiguation, such as the Adams crater, which exists on both the Moon and Mars. We present a multi-step pipeline combining rule-based filtering, statistical relevance analysis, part-of-speech (POS) tagging, named entity recognition (NER) model, hybrid keyword harvesting, knowledge graph (KG) matching, and inference with a locally installed large language model (LLM) to reliably identify planetary names despite these challenges. When evaluated on a dataset of astronomy papers from the Astrophysics Data System (ADS), this methodology achieves an F1-score over 0.97 in disambiguating planetary feature names.

Timothy W. Hostetler, Shinyi Chen, Sergi Blanco-Cuaresma et al.

The NASA Astrophysics Data System (ADS), a critical research service for the astrophysics community, strives to provide the most accessible and inclusive environment for the discovery and exploration of the astronomical literature. Part of this goal involves creating a digital platform that can accommodate everybody, including those with disabilities that would benefit from alternative ways to present the information provided by the website. NASA ADS follows the official Web Content Accessibility Guidelines (WCAG) standard for ensuring accessibility of all its applications, striving to exceed this standard where possible. Through the use of both internal audits and external expert review based on these guidelines, we have identified many areas for improving accessibility in our current web application, and have implemented a number of updates to the UI as a result of this. We present an overview of some current web accessibility trends, discuss our experience incorporating these trends in our web application, and discuss the lessons learned and recommendations for future projects.

Felix Grezes, Sergi Blanco-Cuaresma, Alberto Accomazzi et al.

The existing search tools for exploring the NASA Astrophysics Data System (ADS) can be quite rich and empowering (e.g., similar and trending operators), but researchers are not yet allowed to fully leverage semantic search. For example, a query for "results from the Planck mission" should be able to distinguish between all the various meanings of Planck (person, mission, constant, institutions and more) without further clarification from the user. At ADS, we are applying modern machine learning and natural language processing techniques to our dataset of recent astronomy publications to train astroBERT, a deeply contextual language model based on research at Google. Using astroBERT, we aim to enrich the ADS dataset and improve its discoverability, and in particular we are developing our own named entity recognition tool. We present here our preliminary results and lessons learned.

Sergi Blanco-Cuaresma, Alberto Accomazzi, Michael J. Kurtz et al.

The Agile manifesto encourages us to value individuals and interactions over processes and tools, while Scrum, the most adopted Agile development methodology, is essentially based on roles, events, artifacts, and the rules that bind them together (i.e., processes). Moreover, it is generally proclaimed that whenever a Scrum project does not succeed, the reason is because Scrum was not implemented correctly and not because Scrum may have its own flaws. This grants irrefutability to the methodology, discouraging deviations to fit the actual needs and peculiarities of the developers. In particular, the members of the NASA ADS team are highly creative and autonomous whose motivation can be affected if their freedom is too strongly constrained. We present our experience following Agile principles, reusing certain Scrum elements and seeking the satisfaction of the team members, while rapidly reacting/keeping the project in line with our stakeholders expectations.

Kriste Krstovski, Michael J. Kurtz, David A. Smith et al.

Scientific publications have evolved several features for mitigating vocabulary mismatch when indexing, retrieving, and computing similarity between articles. These mitigation strategies range from simply focusing on high-value article sections, such as titles and abstracts, to assigning keywords, often from controlled vocabularies, either manually or through automatic annotation. Various document representation schemes possess different cost-benefit tradeoffs. In this paper, we propose to model different representations of the same article as translations of each other, all generated from a common latent representation in a multilingual topic model. We start with a methodological overview on latent variable models for parallel document representations that could be used across many information science tasks. We then show how solving the inference problem of mapping diverse representations into a shared topic space allows us to evaluate representations based on how topically similar they are to the original article. In addition, our proposed approach provides means to discover where different concept vocabularies require improvement.