{"id":463398,"date":"2026-02-09T08:19:08","date_gmt":"2026-02-09T08:19:08","guid":{"rendered":"https:\/\/www.newsbeep.com\/ca\/463398\/"},"modified":"2026-02-09T08:19:08","modified_gmt":"2026-02-09T08:19:08","slug":"openscholar-beats-chatgpt-in-scientific-citation-accuracy","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/ca\/463398\/","title":{"rendered":"OpenScholar Beats ChatGPT In Scientific Citation Accuracy"},"content":{"rendered":"<p>            <a href=\"https:\/\/www.opensourceforu.com\/wp-content\/uploads\/2026\/02\/openscholar-e1770621045200.jpg\" data-caption=\"OpenScholar Trained On 45 Million Papers Outperforms ChatGPT And Other LLMs In Scientific Citation Accuracy And Research Usefulness&#010;\" rel=\"nofollow noopener\" target=\"_blank\"><img loading=\"lazy\" decoding=\"async\" width=\"696\" height=\"487\" class=\"entry-thumb td-modal-image\" src=\"https:\/\/www.newsbeep.com\/ca\/wp-content\/uploads\/2026\/02\/openscholar-696x487.jpg\"   alt=\"OpenScholar Trained On 45 Million Papers Outperforms ChatGPT And Other LLMs In Scientific Citation Accuracy And Research Usefulness\" title=\"OpenScholar Trained On 45 Million Papers Outperforms ChatGPT And Other LLMs In Scientific Citation Accuracy And Research Usefulness\"\/><\/a>OpenScholar Trained On 45 Million Papers Outperforms ChatGPT And Other LLMs In Scientific Citation Accuracy And Research Usefulness<\/p>\n<p>University of Washington researchers release OpenScholar, an open source scientific LLM, that beats proprietary tools like ChatGPT in citation accuracy and trusted literature synthesis.<\/p>\n<p>OpenScholar, an open source large language model built specifically for scientific literature search and synthesis, has outperformed proprietary systems such as ChatGPT, GPT-4o and Perplexity in citation accuracy and answer usefulness. The research, published in Nature, positions the transparent tool as a credible alternative to black-box generative AI for science.<\/p>\n<p>Developed by computer scientists Hannaneh Hajishirzi and Akari Asai at the University of Washington, the model was trained exclusively on 45 million open access scientific papers. It uses retrieval-augmented generation (RAG) to incorporate new information beyond its training data, reducing hallucinations, outdated responses and irrelevant citations.<\/p>\n<p>Automatic testing showed higher citation accuracy than competing models. In manual evaluations, 16 domain experts compared AI responses with human-written answers. OpenScholar\u2019s outputs were rated more useful more than 50 per cent of the time, largely because they were more comprehensive and typically twice as detailed.<\/p>\n<p>Demand surfaced quickly after an early demo release. \u201cquicky, we got a lot of queries, far more than we\u2019d expected. It really speaks to the need for this sort of open-source, transparent system that can synthesize research,\u201d said Hajishirzi. She added, \u201cbut the big question ultimately is whether we can trust that its answers are correct,\u201d highlighting concerns with general-purpose AI.<\/p>\n<p>Asai noted, \u201cIt might cite some research papers that weren\u2019t the most relevant or cite just one paper or pull from a blog post randomly,\u201d adding, \u201cWe\u2019ve already seen a lot of scientists using OpenScholar because it\u2019s open source. Others are building on this research and already improving on our results.\u201d<\/p>\n<p>The team is now developing Deep Research Tulu to deliver even more comprehensive scientific responses.<\/p>\n","protected":false},"excerpt":{"rendered":"OpenScholar Trained On 45 Million Papers Outperforms ChatGPT And Other LLMs In Scientific Citation Accuracy And Research Usefulness&hellip;\n","protected":false},"author":2,"featured_media":463399,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[7],"tags":[49,48,66],"class_list":["post-463398","post","type-post","status-publish","format-standard","has-post-thumbnail","category-science","tag-ca","tag-canada","tag-science"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/posts\/463398","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/comments?post=463398"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/posts\/463398\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/media\/463399"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/media?parent=463398"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/categories?post=463398"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/tags?post=463398"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}