Prompt Tuning without Labeled Samples for Zero-Shot Node Classification in Text-Attributed Graphs Conference: Proceedings of the Nineteenth ACM International Conference on Web Search and Data Mining; February 22–26, 2026; Boise, ID, USA Proceedings of the Nineteenth ACM International Conference on W

Sethupathy Parameswaran, Suresh Sundaram, Yuan Fang · 2026

Read Prompt Tuning without Labeled Samples for Zero-Shot Node Classification in Text-Attributed Graphs Conference: Proceedings of the Nineteenth ACM International Conference on Web Search and Data Mining; February 22–26, 2026; Boise, ID, USA Proceedings of the Nineteenth ACM International Conference on W by Sethupathy Parameswaran, Suresh Sundaram, Yuan Fang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-03793.txt.

Description

Sethupathy Parameswaran, Suresh Sundaram, and Yuan Fang tackle zero-shot node classification in text-attributed graphs, networks whose nodes carry written descriptions, like papers with abstracts or products with blurbs. With no labeled examples at all, sorting nodes into classes is hard. Their Zero-shot Prompt Tuning framework leans on a universal bimodal conditional generator to manufacture the missing training data.

First, a graph-language model is pretrained to capture both network structure and node text. A conditional generative model then learns the joint pattern of graph and text embeddings, producing synthetic samples for each class from its name alone. Those samples support continuous prompt tuning for the classification task. Synthetic stand-ins for labels could help cold-start problems in social networks, publishing, and online retail alike.

Catalog record

CREATORSethupathy Parameswaran, Suresh Sundaram, Yuan Fang
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE52 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-03793

Browse the full library · About the Tower of Babel