Describe your research once. Every week, ScholarFeedAI scans PubMed, scores relevance with AI, and delivers a curated digest straight to your inbox.
No credit card required · Free to start
Three steps. Five minutes. Then it's automatic.
Tell us about your research interests in plain language. What you are doing, what you are looking for. We'll learn from them.
Our AI analyzes your description to identify the terms, methods, and topics that define your research profile.
Every Monday, receive a curated email with the most relevant new papers for your research, summarized and commented for you by AI.
Concise summaries, personnalized comments, and direct links - all in one email.
Weekly Digest
March 2026 · 8 papers curated for you
Transformer architectures for protein structure prediction in cryo-EM data
Nature Methods · Jan 2026 · Chen Y., Patel R., Müller K. et al.
Introduces a novel transformer-based approach that improves resolution of cryo-EM density maps by 15%, enabling more accurate atomic model building from raw electron microscopy data. [...]
🔗 Why this matters for you:
This directly extends your work on deep learning for structural biology — the attention mechanism they use on density maps mirrors your current architecture. [...]
Self-supervised contrastive learning for molecular dynamics simulations
PNAS · Jan 2026 · Okonkwo A., Zhang L., Ferreira M. et al.
Proposes a contrastive learning framework that reduces the computational cost of MD simulations by 3× while maintaining physical accuracy across benchmark protein folding datasets. [...]
🔗 Why this matters for you:
Highly relevant to your interest in ML-accelerated simulations — their open-source training code could be adapted directly to your current pipeline. [...]
+ 6 more papers in the full digest
4 free weekly digests to try. No credit card needed.
4 weekly digests for free
per month
per person / month
We designed ScholarFeedAI with one rule: your research interests belong to you.
We use your research description and keywords exclusively to generate your weekly digest. We never sell, rent, share, or monetize your personal data with any third party.
All AI processing is done via Mistral AI, the leading European sovereign AI company. Your data is processed entirely within European infrastructure.
You have the right to access, correct, export, or delete all data we hold on you at any time. Just contact us. We are fully compliant with the EU General Data Protection Regulation.
ScholarFeedAI searches PubMed/MEDLINE, one of the world's largest databases of biomedical and life sciences literature, covering over 37 million citations. We scan for new papers published in the past 7 days that match your research profile.
After you describe your research, our AI extracts 10 specific keywords. Each week, these keywords query PubMed. Papers are ranked by a relevance score based on keyword matches, and the top results are included in your digest with AI-generated summaries.
Yes, but not just any random one. Your research description and keywords are sent to Mistral AI to generate summaries and relevance notes. Mistral AI is a European sovereign AI company. Your data is processed exclusively on European servers in compliance with GDPR. Mistral AI does not use your data to train its models.
Your weekly digests will pause, but your profile, keywords, and past digests remain saved. Subscribe at any time to resume. We never charge without explicit consent.
Absolutely. Head to the Keywords section in your dashboard to add, remove, edit, or toggle any keyword. You can also re-run AI extraction from an updated research description at any time.
Every Monday at 7:00 AM UTC. Digests cover papers published during the previous week.
Yes, at any time from your account settings. You'll keep access until the end of your current billing period. No penalties or hidden fees.
Still have questions? Contact us →
Join researchers who save hours every week by letting AI curate their reading list.
Get Started Free →