ContractRL: Shielded Group-Relative Policy Optimization for Auditable Tool-Call Repair

Excerpt

arXiv:2610.00328v1 Announce Type: cross Abstract: Structured tool calls often fail after only a small number of fields violate a schema or an execution contract. Regenerating the complete object enlarges the action surface and makes repeated repair difficult to audit. We introduce ContractRL, a contract-constrained sequential repair protocol that models verifier-guided JSON repair as a bounded decision process. At each step the policy observes the candidate, typed verifier feedback, JSON Pointer