Kasi Vegesana
@h0rcrux.bsky.social
Social media parasite. Always browsing, never interacting.
Reposted by Kasi Vegesana
DeepSeek R1 works remarkably well for the biology/immunology/oncology/genomics queries I've thrown at it. The change-of-thought (which is funny to read) seems to really operationalize a form of reasoning
Not sure if it's better than o3 (haven't used it) but seems to best o1/Sonnet 3.5/Gemini 2.0
Not sure if it's better than o3 (haven't used it) but seems to best o1/Sonnet 3.5/Gemini 2.0
January 24, 2025 at 8:33 PM
DeepSeek R1 works remarkably well for the biology/immunology/oncology/genomics queries I've thrown at it. The change-of-thought (which is funny to read) seems to really operationalize a form of reasoning
Not sure if it's better than o3 (haven't used it) but seems to best o1/Sonnet 3.5/Gemini 2.0
Not sure if it's better than o3 (haven't used it) but seems to best o1/Sonnet 3.5/Gemini 2.0
Reposted by Kasi Vegesana
Reposted by Kasi Vegesana
How to learn generalizable rules across complex sequence-function maps? 💡In our new preprint we propose a framework for learning two-point statistics and use it to discover biophysical rules of TCR co-specificity that generalize to unseen ligands ! ✨ arxiv.org/abs/2412.13722
December 19, 2024 at 11:14 AM
How to learn generalizable rules across complex sequence-function maps? 💡In our new preprint we propose a framework for learning two-point statistics and use it to discover biophysical rules of TCR co-specificity that generalize to unseen ligands ! ✨ arxiv.org/abs/2412.13722