Workflow as Knowledge: Semantic Persistence for LLM-Mediated Workflows (arXiv 2507.08709)

Forum topic · 小凯 · 2026-07-13

Summary

This arXiv paper (2507.08709) by Emanuele Quinto, Carlo Andrea Rozzi, and Francesco Zanitti proposes a conceptual model for LLM applications that use explicit workflows for tool use, retrieval, branching, checkpointing, and human approval. The authors introduce a Lisp-inspired but language-independent framework in which symbolic forms, object identity, and live-image thinking serve as explanatory lenses rather than implementation commitments. In this model, workflow definitions, workflow instances, inference records, context snapshots, and dependency relations are represented as persistent knowledge objects within a shared knowledge substrate. The central contribution is a semantic distinction between derive and infer: derive refers to deterministic computation over available state, while infer refers to LLM-mediated judgment performed under declared context and executor-controlled capability policy. Published July 12, 2025, the paper aims to give AI systems a principled way to treat workflows as first-class, persistent knowledge, improving auditability, reproducibility, and governance of LLM-mediated automation pipelines.

Paper Overview

Field: AI Systems Authors: Emanuele Quinto, Carlo Andrea Rozzi, Francesco Zanitti Published: 2025-07-12 arXiv: 2507.08709

Abstract (Original)

Large language model (LLM) applications increasingly use explicit workflows for tool use, retrieval, branching, checkpointing, and human approval. This paper proposes a Lisp-inspired but language-independent conceptual model: symbolic forms, object identity, and live-image thinking are used as explanatory lenses, not implementation commitments. In this model, workflow definitions, workflow instances, inference records, context snapshots, and dependency relations are represented as persistent knowledge objects in a shared knowledge substrate. Its central semantic distinction is between derive and infer: derive is deterministic computation over available state; infer is mediated LLM judgment under declared context and executor-controlled capability policy.

Key Ideas

Workflows as persistent knowledge objects: workflow definitions, instances, inference records, context snapshots, and dependency relations all live in a shared knowledge substrate.
Lisp-inspired, language-independent model: symbolic forms, object identity, and live-image thinking are used as explanatory lenses, not implementation commitments.
Core semantic distinction:
derive — deterministic computation over available state.
infer — LLM-mediated judgment under declared context and executor-controlled capability policy.

Relevance

The model provides a principled foundation for building auditable and governable LLM-mediated automation pipelines, where each inference step can be recorded, checkpointed, and subject to human approval.