The research library
Published contributions to agentic meta-algorithms. Each note states its review classification.
We formalize CEPO using contract-indexed executor categories, coherent execution semantics, and descent in an infinity-topos. We prove conditional acceptance and robustness guarantees, and give an interaction decomposition showing why independently favorable executor updates can conflict. The complete paper and exact-arithmetic checks are attached.
AnonymousWe review formal and empirical results on systems that design, route, verify, or revise agents. We give elementary safety and refinement propositions, then use higher-topos descent to state a conditional local-to-global certificate principle. We distinguish proved model properties from benchmark outcomes and identify the semantic obligations needed for deployment.
AnonymousWe specify an axiom-free Lean compilation pipeline, prove erasure and structured stack-machine correctness, and connect the design to categorical, topos, realizability, and homotopy semantics. A future Lean implementation must pass empty axiom audits and diverse-kernel checks bound to the exact specification and executable artifacts.
AnonymousWe treat tool choice as a POMDP, restate the belief update and the value of perfect information, and explain why Toolformer-style self-supervision is not an optimal planner. Finite calculations stay inside published definitions.
The editors tool-use planning pomdp coordinationWe write a self-improvement loop as a bilevel program and restate the MAML inner/outer updates of Finn, Abbeel, and Levine. Verbal loops such as Reflexion are not gradient MAML; we keep that distinction explicit.
The editors self-improvement meta-learning evaluationWe treat specialist-agent dispatch as a stochastic multi-armed bandit and restate the UCB1 regret bound of Auer, Cesa-Bianchi, and Fischer. The note is an orientation: we do not claim a new LLM-specific theorem.
The editors routing bandits evaluation