Saved in:
Bibliographic Details
Main Authors: Vir, Reya, Shankar, Shreya, Chase, Harrison, Fu-Hinthorn, Will, Parameswaran, Aditya
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.14738
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910915327164416
author Vir, Reya
Shankar, Shreya
Chase, Harrison
Fu-Hinthorn, Will
Parameswaran, Aditya
author_facet Vir, Reya
Shankar, Shreya
Chase, Harrison
Fu-Hinthorn, Will
Parameswaran, Aditya
contents Large language models (LLMs) are increasingly deployed in specialized production data processing pipelines across diverse domains -- such as finance, marketing, and e-commerce. However, when running them in production across many inputs, they often fail to follow instructions or meet developer expectations. To improve reliability in these applications, creating assertions or guardrails for LLM outputs to run alongside the pipelines is essential. Yet, determining the right set of assertions that capture developer requirements for a task is challenging. In this paper, we introduce PROMPTEVALS, a dataset of 2087 LLM pipeline prompts with 12623 corresponding assertion criteria, sourced from developers using our open-source LLM pipeline tools. This dataset is 5x larger than previous collections. Using a hold-out test split of PROMPTEVALS as a benchmark, we evaluated closed- and open-source models in generating relevant assertions. Notably, our fine-tuned Mistral and Llama 3 models outperform GPT-4o by 20.93% on average, offering both reduced latency and improved performance. We believe our dataset can spur further research in LLM reliability, alignment, and prompt engineering.
format Preprint
id arxiv_https___arxiv_org_abs_2504_14738
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PROMPTEVALS: A Dataset of Assertions and Guardrails for Custom Production Large Language Model Pipelines
Vir, Reya
Shankar, Shreya
Chase, Harrison
Fu-Hinthorn, Will
Parameswaran, Aditya
Computation and Language
Large language models (LLMs) are increasingly deployed in specialized production data processing pipelines across diverse domains -- such as finance, marketing, and e-commerce. However, when running them in production across many inputs, they often fail to follow instructions or meet developer expectations. To improve reliability in these applications, creating assertions or guardrails for LLM outputs to run alongside the pipelines is essential. Yet, determining the right set of assertions that capture developer requirements for a task is challenging. In this paper, we introduce PROMPTEVALS, a dataset of 2087 LLM pipeline prompts with 12623 corresponding assertion criteria, sourced from developers using our open-source LLM pipeline tools. This dataset is 5x larger than previous collections. Using a hold-out test split of PROMPTEVALS as a benchmark, we evaluated closed- and open-source models in generating relevant assertions. Notably, our fine-tuned Mistral and Llama 3 models outperform GPT-4o by 20.93% on average, offering both reduced latency and improved performance. We believe our dataset can spur further research in LLM reliability, alignment, and prompt engineering.
title PROMPTEVALS: A Dataset of Assertions and Guardrails for Custom Production Large Language Model Pipelines
topic Computation and Language
url https://arxiv.org/abs/2504.14738