-
Harvard
- Harvard, USA
- https://zitniklab.hms.harvard.edu
- @marinkazitnik
Stars
Kempner Institute Computing Handbook
SEMPER: Multimodal protein representations across human brain cell types
STRAND: Sequence-Conditioned Transport for Single-Cell Perturbations
Latent Decision Flow Rewards for Clinical Guideline Reasoning
When do biological reasoning models use their biological inputs? Strong benchmark performance can create a mirage of reliance on biological foundation model representations, even when those represe…
Do LLMs Know What to Ask and When? Evaluating Multi-Turn Information Seeking
A protocol for event-level logging of clinical AI.
How Post-Training Shapes Biological Reasoning Models
ATHENA-R1: AI agent for treatment reasoning over a biomedical tool universe
[ICML'26] SPATIA: Multimodal Generation and Prediction of Spatial Cell Phenotypes
[ICML 2026] ARTIST: Adaptive Time Series Reasoning via Segment Selection
Evaluating Relational Reasoning in LLMs with REL
AutoScientists: Self-Organizing Agent Teams for Long-Running Scientific Experimentation
Autonomous Knowledge Graph Exploration with Adaptive Breadth-Depth Retrieval
A modern multimodal knowledge graph with type-specific metadata across biomedical domains.
Qworld: Question-Specific Evaluation Criteria for LLMs
Terminal based chat implementation of ARK agent
Act or Defer: Error-Controlled Decision Policies for Medical Foundation Models
Medea: An AI agent for therapeutic reasoning across biological contexts
Environment for developing substructure-aware protein models
Graph AI generates neurological hypotheses validated in molecular, organoid, and clinical systems
Precision Medicine Knowledge Graph (PrimeKG)
ProCyon: A multimodal foundation model for protein phenotypes
Geometric Byte Pair Encoding of Protein Structure (ICLR 2026)
Generative modeling of spatially resolved perturbation responses with CONCERT
ClinVec: Unified Embeddings of Clinical Codes Enable Knowledge-Grounded AI in Medicine
Combinatorial prediction of therapeutic perturbations using causally-inspired neural networks