🤖 AI Summary
This work addresses the tension between the flexibility of natural language interaction and the need for determinism, traceability, and controllability in scientific workflows—a challenge inadequately met by existing systems. The authors propose a schema-gated orchestration architecture that elevates execution boundaries to the cross-step workflow level, decoupling dialogue understanding from execution authority through machine-verifiable schema specifications. Three operational principles are introduced: clarification-first interaction, constrained plan-execution coordination, and tool-to-workflow-level gating. Evaluated via a multi-model protocol spanning three LLM families and 15 conversational turns, and benchmarked against 20 systems, the approach demonstrates effective unification of high flexibility and high determinism. Multi-model scores show strong agreement with expert assessments (Krippendorff’s α = 0.80–0.98).
📝 Abstract
Large language models (LLMs) can now translate a researcher's plain-language goal into executable computation, yet scientific workflows demand determinism, provenance, and governance that are difficult to guarantee when an LLM decides what runs. Semi-structured interviews with 18 experts across 10 industrial R&D stakeholders surface 2 competing requirements--deterministic, constrained execution and conversational flexibility without workflow rigidity--together with boundary properties (human-in-the-loop control and transparency) that any resolution must satisfy. We propose schema-gated orchestration as the resolving principle: the schema becomes a mandatory execution boundary at the composed-workflow level, so that nothing runs unless the complete action--including cross-step dependencies--validates against a machine-checkable specification. We operationalize the 2 requirements as execution determinism (ED) and conversational flexibility (CF), and use these axes to review 20 systems spanning 5 architectural groups along a validation-scope spectrum. Scores are assigned via a multi-model protocol--15 independent sessions across 3 LLM families--yielding substantial-to-near-perfect inter-model agreement (Krippendorff a=0.80 for ED and a=0.98 for CF), demonstrating that multi-model LLM scoring can serve as a reusable alternative to human expert panels for architectural assessment. The resulting landscape reveals an empirical Pareto front--no reviewed system achieves both high flexibility and high determinism--but a convergence zone emerges between the generative and workflow-centric extremes. We argue that a schema-gated architecture, separating conversational from execution authority, is positioned to decouple this trade-off, and distill 3 operational principles--clarification-before-execution, constrained plan-act orchestration, and tool-to-workflow-level gating--to guide adoption.