arXiv cs.CL· Chun Chet Ng, Jia Yu Lim, Zhen Hao Chu, Yixi Zhou, Wei Zeng Low·· 13 小时前AI 评分31
PRISM：面向金融检索的提示词精炼上下文系统建模框架
PRISM: Prompt-Refined In-Context System Modelling for Financial Retrieval
AI 导读
PRISM 是一个无需微调的金融文档检索框架，组合提示词工程、上下文学习与候选精炼模块，在 FiQA-2018 上将 Recall@100 从 0.7783 提升至 0.8326，NDCG@10 达 0.6197。
正文 · 原文
Abstract:Deploying LLMs for financial information retrieval demands accuracy, cost efficiency, and reproducibility, yet most approaches require expensive fine-tuning. PRISM is a training-free framework that composes three deployable modules (prompt engineering, in-context learning, and a candidate refinement module) with a boundary-condition study of multi-agent coordination for financial document retrieval. On FiQA-2018, stacked reranking and $k$-reciprocal encoding raise Recall@100 from 0.7783 to 0.8326. On FinAgentBench, a Cohere-refined top-50 pool yields a higher observed end-to-end NDCG@5, below this evaluation's minimum detectable effect, while cutting per-query cost by about 35%. Added complexity does not always pay off: $k$-reciprocal encoding stops helping once the upstream stage is already high-recall, and multi-agent coordination is not a default for fine-grained chunk ranking. Matched pairs that hold prompt, model, and exemplars fixed lose 0.0702 and 0.0681 combined NDCG@5 to their single-path controls, and the whole gap sits at the chunk stage. We release an expert-annotated FinAgentBench evaluation set, together with a label-robustness sweep over 16 alternative ground-truth constructions, restoring a reproducible evaluation path after the original evaluator went offline. Without fine-tuning, PRISM placed third on the official FinAgentBench private leaderboard (leaderboard basis, NDCG@5 0.71181), reaches NDCG@10 0.6197 on FiQA-2018, and 98% accuracy on FinanceBench. We report per-query latency, token, and dollar accounting across roughly 40 configurations, present the quality-cost-latency Pareto frontier, and introduce FACET, a transparent SLA-specific scalarization for choosing among non-dominated configurations. Code and dataset are released at this https URL.
| Comments: | Accepted to the EMNLP 2026 Industry Track |
| Subjects: | Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE); Computation and Language (cs.CL); Information Retrieval (cs.IR) |
| Cite as: | arXiv:2511.14130 [cs.AI] |
| (or arXiv:2511.14130v3 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2511.14130 arXiv-issued DOI via DataCite |
Submission history
From: Chun Chet Ng [view email]
[v1]
Tue, 18 Nov 2025 04:30:52 UTC (1,909 KB)
[v2]
Mon, 6 Apr 2026 14:09:13 UTC (3,254 KB)
[v3]
Wed, 21 Oct 2026 08:59:25 UTC (1,275 KB)
来源：arXiv cs.CL · arxiv.org