Workflow as Knowledge: Semantic Persistence for LLM-Mediated Workflows

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Emanuele Quinto, Carlo Andrea Rozzi, Francesco Zanitti
• Venue: arXiv: cs.AI
• Published: 2026-07-09
• Collected: 2026-07-10
• URL: http://arxiv.org/abs/2607.08740v1
• PDF: https://arxiv.org/pdf/2607.08740v1
• DOI: UNKNOWN

Plain-Language Brief

The paper proposes a conceptual model for large language model (LLM) applications that uses explicit workflows for various tasks. This model aims to represent workflow definitions, instances, and other related elements as persistent knowledge objects in a shared substrate. The central distinction is between 'derive' and 'infer', with derive being deterministic computation and infer being mediated by the LLM under certain contexts.

What the Document Says

• The paper discusses a conceptual model for LLM applications using workflows.
• The model uses symbolic forms, object identity, and live-image thinking as explanatory lenses.
• Workflow definitions, instances, inference records, context snapshots, and dependency relations are represented as persistent knowledge objects.
• The central semantic distinction is between 'derive' (deterministic computation) and 'infer' (mediated by LLM under declared context and executor-controlled capability policy).
• The result is a preliminary conceptual account of semantic persistence for workflows.

Key Concepts Mentioned

• Large Language Model (LLM)
• Workflow
• Symbolic Forms
• Object Identity
• Live-Image Thinking
• Derive
• Infer

Claims Made by the Document

• This model can represent workflows as persistent knowledge objects.
• The model distinguishes between 'derive' and 'infer'.

Evidence Used

• The text explains that the proposed model represents workflow definitions, instances, etc., as persistent knowledge objects.
• The text clarifies the central semantic distinction between 'derive' and 'infer'.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Large language model (LLM) applications increasingly use explicit workflows for tool use, retrieval, branching, checkpointing, and human approval. Existing workflow systems already address many execution concerns. 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. The result is a preliminary conceptual account of semantic persistence: workflows do not merely produce knowledge and leave traces, but can themselves be represented as inspectable, resumable, and reviewable knowledge objects, while formal transition semantics remain future work.