HydroSophyTech/BEAKER
Viewer • Updated • 2.01k • 2 • 1
HydroSophy
non-profit
AI & ML interests
AI for Science, Molecular Modeling, Large Language Models, Multimodal Learning, Dataset Benchmarking, Drug Design, and Structural Biology.
Recent Activity
Organization Card
HydroSophy
About
We work on AI for Science — mostly molecular modeling, multimodal foundation models, and the benchmarks / datasets / evaluation tools that go with them.
Projects
⚗️ BEAKER · NeurIPS 2026 Spotlight
BEAKER: An Expert-Curated Benchmark for Embodied Brains in Self-Driving Chemical Laboratories
A benchmark for MLLMs as embodied brains in self-driving chemical laboratories — 1,000 expert-curated questions, 11 image and video tasks across Actionability, Knowledge, Experimentation, and Reasoning, with 28 MLLMs evaluated.
🧪 ToxiMol · KDD 2026 Oral
Breaking Bad Molecules: Are MLLMs Ready for Structure-Level Molecular Detoxification?
A benchmark for molecular toxicity repair with MLLMs — 660 toxic molecules, 11 repair tasks, multimodal inputs (SMILES + 2D structures), plus the companion ToxiEval evaluation framework.
🔐 ToxiEval-ZKP · IFAC BIKA Best Paper Award
ToxiEval-ZKP: A Structure-Private Verification Framework for Molecular Toxicity Repair Tasks
A structure-private, verifiable evaluation framework for molecular toxicity repair built on zero-knowledge proofs.
📄 Paper | 🔬 Code (coming soon)
Contact
- GitHub: @HydroSophy
- Email: feilin@ieee.org
- Location: Macau, China
models 0
None public yet