LOGISTAR — Scheda Oggetto

Oggetto salvato correttamente.

Dati oggetto

URL

Apri

Titolo / Descrizione

Sito / Azienda

Estratto / Body

&lt;p&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;ContentParagraph ContentParagraph_align_left&quot; data-testid=&quot;content-paragraph&quot;&gt;&lt;span class=&quot;ContentText ContentText_variant_bodyNormal&quot; data-testid=&quot;content-text&quot;&gt;AI researchers from Stanford and Turing award winner Yoshua Bengio have trained a large language model on human genome data to better predict DNA profiles.&lt;/span&gt;&lt;/p&gt;
&lt;p class=&quot;ContentParagraph ContentParagraph_align_left&quot; data-testid=&quot;content-paragraph&quot;&gt;&lt;span class=&quot;ContentText ContentText_variant_bodyNormal&quot; data-testid=&quot;content-text&quot;&gt;Researchers created &lt;/span&gt;&lt;span class=&quot;ContentText ContentText_variant_bodyNormal&quot; data-testid=&quot;content-text&quot;&gt;&lt;a class=&quot;ContentText-BodyTextChunk ContentText-BodyTextChunk_link&quot; href=&quot;https://huggingface.co/papers/2306.15794&quot; target=&quot;_blank&quot; data-feathr-click-track=&quot;true&quot; data-feathr-link-aids=&quot;[&amp;quot;64db786b15f34666fc299753&amp;quot;]&quot;&gt;HyenaDNA &lt;/a&gt;&lt;/span&gt;&lt;span class=&quot;ContentText ContentText_variant_bodyNormal&quot; data-testid=&quot;content-text&quot;&gt;by combining the &lt;/span&gt;&lt;span class=&quot;ContentText ContentText_variant_bodyNormal&quot; data-testid=&quot;content-text&quot;&gt;&lt;a class=&quot;ContentText-BodyTextChunk ContentText-BodyTextChunk_link&quot; href=&quot;https://arxiv.org/abs/2302.10866&quot; target=&quot;_blank&quot; data-feathr-click-track=&quot;true&quot; data-feathr-link-aids=&quot;[&amp;quot;64db786b15f34666fc299753&amp;quot;]&quot;&gt;Hyena large language model&lt;/a&gt;&lt;/span&gt;&lt;span class=&quot;ContentText ContentText_variant_bodyNormal&quot; data-testid=&quot;content-text&quot;&gt; and pre-training it on human reference genomic sequences of up to one million tokens. Prior models have typically used context lengths of 512 to 4,000 token, or less than 0.001% of the human genome.&lt;/span&gt;&lt;/p&gt;
&lt;p class=&quot;ContentParagraph ContentParagraph_align_left&quot; data-testid=&quot;content-paragraph&quot;&gt;&lt;span class=&quot;ContentText ContentText_variant_bodyNormal&quot; data-testid=&quot;content-text&quot;&gt;Stanford&amp;rsquo;s researchers contend that most of the work on long context models have focused on natural language and code and that biology is &amp;ldquo;inherently made of ultralong sequences.&amp;rdquo;&lt;/span&gt;&lt;/p&gt;

Note

Voto oggetto (1-10)

Precisione (1-10)

Tag (8)

✎ Modifica tag 🤖 Suggerisci AI

           genoma,llm,dev,nn,tool,gpt        

dev 5 gen 5 genoma 5 gpt 5 llm 5 nn 5 tool 5 wpg 5

ID: 343 • Tipo: WPG • Reg: 2023-10-09 • Mod: 29/07/2026 20:16