arXiv cs.CL· Yiheng Su, Junyi Jessy Li, Matthew Lease·· 3 天前AI 评分22
DualView：利用广泛与输出特定证据的 LLM 多输出文本回归方法
DUALVIEW: Leveraging Broad and Output-Specific Evidence for LLM-based Multi-Output Text Regression
AI 导读
DualView 是一种参数高效的多输出文本回归方法，先从冻结的 LLM 中提取广泛与输出特定的表示，再用双分支架构生成完整分数向量。在五个异构回归任务和八种 LLM（最大 122B）上，它超越七个监督回归基线，所需拟合参数比三个最强基线少 82–1842 倍，且用较小的开源 LLM 即可超过 Gemini 3.1 Pro 的无监督预测。
正文 · 原文
Abstract:While work on LLM-based regression has largely focused on scalar prediction, multi-output text regression remains less explored. We present DualView, a parameter-efficient approach to leverage LLMs for multi-output text regression. \method{} first extracts broad and output-specific representations from a frozen LLM, then uses a dual-branch architecture to produce the full score vector. Across five heterogeneous regression tasks and eight LLMs (up to 122B), DualView outperforms seven supervised regression baselines while requiring 82--1842 times fewer fitted parameters than the three strongest baselines. These improvements extend across individual output dimensions and persist even with limited training data. Notably, DualView using a smaller open-weight LLM can outperform unsupervised prediction with Gemini 3.1 Pro. Controlled ablations show that broad and output-specific evidence provide complementary predictive benefits, while lightweight shared predictors achieve strong performance with fewer trainable parameters.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.19206 [cs.CL] |
| (or arXiv:2610.19206v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.19206 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yiheng Su [view email]
[v1]
Tue, 13 Oct 2026 21:17:19 UTC (441 KB)
来源：arXiv cs.CL · arxiv.org