Verity: Neuro-Symbolic Synthesis of Verified Distributed Systems
CE2P translates formal verification failures into structured LLM feedback. Benefit inversely correlated with model capability - weaker models gain 34-39pp.
Papers
Preprints and technical disclosures
Research on AI trust, verification, and agent reliability.
Published
Field-Theoretic Memory for AI Agents
Continuous Dynamics for Context Preservation
Memory architecture treating stored information as continuous fields governed by partial differential equations. Semantic diffusion, thermodynamic decay, and field coupling for persistent, composable agent memory.
+116%
F1 on multi-session reasoning
>99.8%
Collective intelligence
2026
Adversarial Testing for AI Systems
Evolutionary approaches where attack and defense strategies improve through competitive pressure.
Trust-Based Decision Routing
Formal framework for ROI-based decision routing in multi-tier verification systems.
In Progress
Attack Taxonomy for Agentic AI Systems
Comprehensive threat model covering input, state, tool, planning, and coordination attacks. Empirical evaluation across agent architectures.
Activation-Based Attack Detection for Autonomous Agents
Interpretability methods to detect adversarial attacks on agentic systems. Extends metacognitive probing to security.
Calibrated Uncertainty in LLM Reasoning Chains
Uncertainty quantification tracking actual accuracy. Activation-based estimation and propagation through multi-step reasoning chains.
Prior Work
Defensive publications by the founding team during prior work at Google. Published via TD Commons, CC BY 4.0.
Collaborate
Work with us
For Researchers
Co-author a paper
We welcome collaborators across AI safety, interpretability, formal methods, and adversarial ML.
Get in touch
For Organizations
Pilot our tools
Deploy trust infrastructure in production. All packages are open source, with enterprise support via Rotascale.
View packages
Open Source
Contribute on GitHub
All core research, benchmarks, and tools are open source. Issues, PRs, and discussions welcome.
GitHub