Abstract:As real-world datasets become more complex and heterogeneous, supervised learning is often bottlenecked by input representation design. Modeling multimodal data, such as time-series, free text, and structured records, often requires non-trivial domain expertise. We propose an agentic pipeline to streamline this process. First, an LLM analyzes a small but diverse subset of text-serialized input examples in-context to synthesize a global rubric, which acts as a programmatic specification for extracting and organizing evidence. This rubric is then used to transform naive text-serializations of inputs into a more standardized format for downstream models. We also describe local rubrics, which are task-conditioned interpretive summaries generated by an LLM. Across 15 clinical tasks from the EHRSHOT benchmark, our rubric approaches significantly outperform count-feature models, naive LLM baselines, and a clinical foundation model pretrained on orders of magnitude more data. Beyond performance, rubrics offer operational advantages such as being easy to audit, cost-effectiveness at scale, and facilitating tabular representations.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2603.11679 [cs.AI] |
| (or arXiv:2603.11679v3 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2603.11679 arXiv-issued DOI via DataCite |
From: Ilker Demirel [view email]
[v1]
Thu, 12 Mar 2026 08:44:06 UTC (1,217 KB)
[v2]
Sat, 21 Mar 2026 05:47:26 UTC (1,223 KB)
[v3]
Wed, 20 May 2026 19:19:40 UTC (8,113 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。