POaaS: Minimal-Edit Prompt Optimization as a Service to Lift Accuracy and Cut Hallucinations on On-Device sLLMs

Jungwoo Shim, Dae Won Kim, Sunwook Kim, Sooyoung Kim, Myungcheol Lee, Jaegeun Cha, Hyunhwa Choi


Abstract
Small language models (sLLMs) are increasingly deployed on-device, where imperfect user prompts–typos, unclear intent, or missing context–can trigger factual errors and hallucinations. Existing automatic prompt optimization (APO) methods were designed for large cloud LLMs and rely on search that often produces long, structured instructions; when executed under an on-device constraint where the same small model must act as optimizer and solver, these pipelines can waste context and even hurt accuracy. We propose POaaS, a minimal-edit prompt optimization layer that routes each query to lightweight specialists (Cleaner, Paraphraser, Fact-Adder) and merges their outputs under strict drift and length constraints, with a conservative skip policy for well-formed prompts. Under a strict fixed-model setting with Llama-3.2-3B and Llama-3.1-8B, POaaS improves both task accuracy and factuality while representative APO baselines degrade them, and POaaS recovers up to +7.4% under token deletion and mixup. Overall, per-query conservative optimization is a practical alternative to search-heavy APO for on-device sLLMs.
Anthology ID:
2026.fever-1.2
Volume:
Proceedings of the Ninth Fact Extraction and VERification Workshop (FEVER)
Month:
March
Year:
2026
Address:
Rabat, Morocco
Editors:
Mubashara Akhtar, Rami Aly, Rui Cao, Christos Christodoulopoulos, Oana Cocarascu, Zhijiang Guo, Arpit Mittal, Michael Schlichtkrull, James Thorne, Andreas Vlachos
Venues:
FEVER | WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
13–27
Language:
URL:
https://aclanthology.org/2026.fever-1.2/
DOI:
Bibkey:
Cite (ACL):
Jungwoo Shim, Dae Won Kim, Sunwook Kim, Sooyoung Kim, Myungcheol Lee, Jaegeun Cha, and Hyunhwa Choi. 2026. POaaS: Minimal-Edit Prompt Optimization as a Service to Lift Accuracy and Cut Hallucinations on On-Device sLLMs. In Proceedings of the Ninth Fact Extraction and VERification Workshop (FEVER), pages 13–27, Rabat, Morocco. Association for Computational Linguistics.
Cite (Informal):
POaaS: Minimal-Edit Prompt Optimization as a Service to Lift Accuracy and Cut Hallucinations on On-Device sLLMs (Shim et al., FEVER 2026)
Copy Citation:
PDF:
https://aclanthology.org/2026.fever-1.2.pdf