Back to research overview

Research program

AI for Adjuvant Design

This program studies four components of AI-assisted adjuvant research: infrastructure, domain models, neuro-symbolic verification, and scientific agents.

Four-stage visual roadmap progressing from adjuvant research infrastructure to a domain model, neuro-symbolic verification, and an autonomous research agent
The program proceeds from problem formulation and data infrastructure to domain modeling, verification, and agent-based research workflows.

Roadmap

Four connected stages

01Foundation

Benchmarks & Infrastructure

Formal task definitions, structured domain knowledge, datasets, and evaluation protocols for adjuvant research.

View benchmark
02Model building

Adjuvant Domain Model

Multi-agent synthetic-data generation and domain-specific fine-tuning for adjuvant knowledge and reasoning.

03Verification

Neuro-Symbolic Verifier

SAVANT combines literature evidence, immune outcomes, and symbolic mechanism chains to inspect generated designs.

View SAVANT · EMNLP 2026 Findings
04Agentic research

Auto Adjuvant

An adjuvant research agent that connects evidence retrieval, candidate design, verification, and iterative planning.

Direction

Verification across the design workflow

The framework treats evidence retrieval, candidate generation, mechanistic verification, and iterative planning as explicit, inspectable components.

View publications Discuss collaboration