{"paper":{"title":"AstroLLaMA: Towards Specialized Foundation Models in Astronomy","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["astro-ph.CO","astro-ph.GA","astro-ph.HE","cs.CL","cs.LG"],"primary_cat":"astro-ph.IM","authors_text":"Alberto Accomazzi, Alyssa Goodman, Charlie O'Neill, David Brodrick, Ernest Perkowski, Ioana Ciuc\\u{a}, Jack Miller, Jason Li, Jesse Cranney, Jill Naiman, Josh Peek, Kartheik Iyer, Kevin Schawinski, Maja Jab{\\l}o\\'nska, Pranav Khetarpal, Sandor Kruk, Sergio J. Rodr\\'iguez M\\'endez, Sharaf Zaman, Thang Bui, Tomasz R\\'o\\.za\\'nski, Tuan Dung Nguyen, UniverseTBD, Yuan-Sen Ting, Ze-Chang Sun","submitted_at":"2023-09-12T11:02:27Z","abstract_excerpt":"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 few"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.06126","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2309.06126/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}