LangProBe: a Language Program Benchmark

Shangyin Tan, Lakshya A Agrawal, Arnav Singhvi, Liheng Lai, Michael J Ryan, Dan Klein, Omar Khattab, Koushik Sen, Matei Zaharia


Abstract
Composing language models (LMs) into multi-step language programs and automatically optimizing their modular prompts is now a mainstream paradigm for building AI systems, but the tradeoffs in this space have only scarcely been studied before. We introduce LangProBe, the first large-scale benchmark for evaluating the architectures and optimization strategies for language programs, with over 2000 combinations of tasks, architectures, optimizers, and choices of LMs. Using LangProBe, we are the first to study the impact of program architectures and optimizers (and their compositions together and with different models) on tradeoffs of quality and cost. We find that optimized language programs offer strong cost-quality Pareto improvement over raw calls to models, but simultaneously demonstrate that human judgment (or empirical decisions) about which compositions to pursue is still necessary for best performance.
Anthology ID:
2025.findings-emnlp.1172
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2025
Month:
November
Year:
2025
Address:
Suzhou, China
Editors:
Christos Christodoulopoulos, Tanmoy Chakraborty, Carolyn Rose, Violet Peng
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
21489–21509
Language:
URL:
https://aclanthology.org/2025.findings-emnlp.1172/
DOI:
Bibkey:
Cite (ACL):
Shangyin Tan, Lakshya A Agrawal, Arnav Singhvi, Liheng Lai, Michael J Ryan, Dan Klein, Omar Khattab, Koushik Sen, and Matei Zaharia. 2025. LangProBe: a Language Program Benchmark. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 21489–21509, Suzhou, China. Association for Computational Linguistics.
Cite (Informal):
LangProBe: a Language Program Benchmark (Tan et al., Findings 2025)
Copy Citation:
PDF:
https://aclanthology.org/2025.findings-emnlp.1172.pdf
Checklist:
 2025.findings-emnlp.1172.checklist.pdf