Zero-Shot Contextual Embeddings via Offline Synthetic Corpus Generation

Philip Lippmann, Jie Yang


Abstract
Context-aware embedding methods boost retrieval accuracy by conditioning on corpus statistics (e.g., term co-occurrence and topical patterns) extracted from neighboring documents. However, this context-aware approach requires access to the target corpus or requires domain-specific finetuning, posing practical barriers in privacy-sensitive or resource-constrained settings. We present ZEST, a zero-shot contextual adaptation framework that replaces real corpus access with a one-time offline synthesis of a compact proxy. Given only a handful of exemplar documents representative of the general target domain, we use a multi-step hierarchical procedure to generate a synthetic context corpus of several hundred documents that aims to emulate key domain-specific distributions. At inference, the frozen context-aware encoder uses this proxy corpus – without any finetuning or target corpus access – to produce domain-adapted embeddings. Across the MTEB benchmark, ZEST’s zero-shot synthetic context adaptation using only five example documents performs within 0.5% of models leveraging full target corpus access – demonstrating remarkable efficacy without any retraining. ZEST thus provides a practical method for deploying high-performance, adaptable embeddings in constrained environments.
Anthology ID:
2025.findings-emnlp.111
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:
2089–2104
Language:
URL:
https://aclanthology.org/2025.findings-emnlp.111/
DOI:
Bibkey:
Cite (ACL):
Philip Lippmann and Jie Yang. 2025. Zero-Shot Contextual Embeddings via Offline Synthetic Corpus Generation. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 2089–2104, Suzhou, China. Association for Computational Linguistics.
Cite (Informal):
Zero-Shot Contextual Embeddings via Offline Synthetic Corpus Generation (Lippmann & Yang, Findings 2025)
Copy Citation:
PDF:
https://aclanthology.org/2025.findings-emnlp.111.pdf
Checklist:
 2025.findings-emnlp.111.checklist.pdf