From Cross-Task Examples to In-Task Prompts: A Graph-Based Pseudo-Labeling Framework for In-context Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Zihan, Wang, Song, Fu, Xingbo, Shi, Chengshuai, Lei, Zhenyu, Shen, Cong, Li, Jundong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918175943163904
author Chen, Zihan
Wang, Song
Fu, Xingbo
Shi, Chengshuai
Lei, Zhenyu
Shen, Cong
Li, Jundong
author_facet Chen, Zihan
Wang, Song
Fu, Xingbo
Shi, Chengshuai
Lei, Zhenyu
Shen, Cong
Li, Jundong
contents The capability of in-context learning (ICL) enables large language models (LLMs) to perform novel tasks without parameter updates by conditioning on a few input-output examples. However, collecting high-quality examples for new or challenging tasks can be costly and labor-intensive. In this work, we propose a cost-efficient two-stage pipeline that reduces reliance on LLMs for data labeling. Our approach first leverages readily available cross-task examples to prompt an LLM and pseudo-label a small set of target task instances. We then introduce a graph-based label propagation method that spreads label information to the remaining target examples without additional LLM queries. The resulting fully pseudo-labeled dataset is used to construct in-task demonstrations for ICL. This pipeline combines the flexibility of cross-task supervision with the scalability of LLM-free propagation. Experiments across five tasks demonstrate that our method achieves strong performance while lowering labeling costs.
format Preprint
id arxiv_https___arxiv_org_abs_2510_24528
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle From Cross-Task Examples to In-Task Prompts: A Graph-Based Pseudo-Labeling Framework for In-context Learning
Chen, Zihan
Wang, Song
Fu, Xingbo
Shi, Chengshuai
Lei, Zhenyu
Shen, Cong
Li, Jundong
Artificial Intelligence
The capability of in-context learning (ICL) enables large language models (LLMs) to perform novel tasks without parameter updates by conditioning on a few input-output examples. However, collecting high-quality examples for new or challenging tasks can be costly and labor-intensive. In this work, we propose a cost-efficient two-stage pipeline that reduces reliance on LLMs for data labeling. Our approach first leverages readily available cross-task examples to prompt an LLM and pseudo-label a small set of target task instances. We then introduce a graph-based label propagation method that spreads label information to the remaining target examples without additional LLM queries. The resulting fully pseudo-labeled dataset is used to construct in-task demonstrations for ICL. This pipeline combines the flexibility of cross-task supervision with the scalability of LLM-free propagation. Experiments across five tasks demonstrate that our method achieves strong performance while lowering labeling costs.
title From Cross-Task Examples to In-Task Prompts: A Graph-Based Pseudo-Labeling Framework for In-context Learning
topic Artificial Intelligence
url https://arxiv.org/abs/2510.24528