Search papers, labs, and topics across Lattice.
This paper introduces Natural Language Declarative Prompting (NLD-P) as a modular governance method for prompt design to address the challenge of model drift in large language models (LLMs). NLD-P formalizes prompt engineering as a declarative control abstraction, separating provenance, constraints, task content, and evaluation, all encoded in natural language. The authors demonstrate NLD-P's utility as a governance framework accessible to non-developers, emphasizing its potential for maintaining stable and interpretable control amidst evolving LLM ecosystems.
Stop wrestling with prompt engineering as an ad-hoc craft: NLD-P offers a modular, natural language approach to govern prompts and maintain control amidst LLM drift.
The rapid evolution of large language models (LLMs) has transformed prompt engineering from a localized craft into a systems-level governance challenge. As models scale and update across generations, prompt behavior becomes sensitive to shifts in instruction-following policies, alignment regimes, and decoding strategies, a phenomenon we characterize as GPT-scale model drift. Under such conditions, surface-level formatting conventions and ad hoc refinement are insufficient to ensure stable, interpretable control. This paper reconceptualizes Natural Language Declarative Prompting (NLD-P) as a declarative governance method rather than a rigid field template. NLD-P is formalized as a modular control abstraction that separates provenance, constraint logic, task content, and post-generation evaluation, encoded directly in natural language without reliance on external orchestration code. We define minimal compliance criteria, analyze model-dependent schema receptivity, and position NLD-P as an accessible governance framework for non-developer practitioners operating within evolving LLM ecosystems. Portions of drafting and editorial refinement employed a schema-bound LLM assistant configured under NLD-P. All conceptual framing, methodological claims, and final revisions were directed, reviewed, and approved by the human author under a documented human-in-the-loop protocol. The paper concludes by outlining implications for declarative control under ongoing model evolution and identifying directions for future empirical validation.
