| title | SearchJev: A Fast and Calibrated System-1 Model for Search Agents |
|---|---|
| authors | Congfeng Cao, Lipeng Zuo, Konstantinos Papakostas, Qiwei Xu, Songwei Xu, Lun Zhou, Zhaochun Ren, Yougang Lyu, Xiaohui Yan |
| date | 2026-10-04 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2610.05107 |
| upvotes | 32 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| abstract | Search agents repeatedly make short decisions about relevance, evidence sufficiency, and search actions. Using generative language models for these decisions introduces latency and unreliable confidence. We present SearchJev, a fast and calibrated System-1 model that separates search decisions from System-2 reasoning and generation. Given a search state and a decision schema, SearchJev directly scores legal options without autoregressive output generation. We propose Soft-Label Learning for Calibrated Decisions (SLCD) to learn decision probabilities from uncertain supervision and calibrate their confidence. In a dual-system search agent, SearchJev handles short decisions and delegates uncertain judgments to System 2, which retains planning, query generation, and answer composition. We also introduce SearchDecision-Bench, a benchmark unifying six types of search decisions for training and evaluation. On SearchDecision-Bench, SEARCHJEV improves decision quality over same-size Qwen3.5 autoregressive models, achieves 5.2-5.3 times faster decisions, and reduces average expected calibration error by 41-74%. On BrowseComp-Plus, the dual-system agents achieve a 3.7-4.7 times speedup in active search time while improving answer accuracy from 45% to up to 54%. |
cite
@misc{cao2026searchjev,
title={SearchJev: A Fast and Calibrated System-1 Model for Search Agents},
author={Congfeng Cao and Lipeng Zuo and Konstantinos Papakostas and Qiwei Xu and Songwei Xu and Lun Zhou and Zhaochun Ren and Yougang Lyu and Xiaohui Yan},
year={2026},
eprint={2610.05107},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2610.05107}
}