Model spider: Learning to rank pre-trained models efficiently

YK Zhang, TJ Huang, YX Ding… - Advances in Neural …, 2024 - proceedings.neurips.cc
Abstract Figuring out which Pre-Trained Model (PTM) from a model zoo fits the target task is
essential to take advantage of plentiful model resources. With the availability of numerous …

Leveraging small-scale datasets for additive manufacturing process modeling and part certification: Current practice and remaining gaps

D Fullington, E Yangue, MM Bappy, C Liu… - Journal of Manufacturing …, 2024 - Elsevier
Additive manufacturing (AM) provides a data-rich environment for collecting a variety of
process data. These crucial data can be used to develop effective machine learning (ML) …

Beimingwu: A learnware dock system

ZH Tan, JD Liu, XD Bi, P Tan, QC Zheng… - Proceedings of the 30th …, 2024 - dl.acm.org
The learnware paradigm proposed by Zhou (2016) aims to enable users to leverage
numerous existing high-performing models instead of building machine learning models …

[PDF][PDF] Handling Learnwares Developed from Heterogeneous Feature Spaces without Auxiliary Data.

P Tan, ZH Tan, Y Jiang, ZH Zhou - IJCAI, 2023 - lamda.nju.edu.cn
The learnware paradigm proposed by Zhou [2016] devotes to constructing a market of
numerous wellperformed models, enabling users to solve problems by reusing existing …

Identifying helpful learnwares without examining the whole market

Y Xie, ZH Tan, Y Jiang, ZH Zhou - ECAI 2023, 2023 - ebooks.iospress.nl
The learnware paradigm aims to construct a market of numerous well-performing machine
learning models, which enables users to leverage these models to accomplish specific tasks …

Learning Only When It Matters: Cost-Aware Long-Tailed Classification

YC He, YX Ding, HJ Ye, ZH Zhou - … of the AAAI Conference on Artificial …, 2024 - ojs.aaai.org
Most current long-tailed classification approaches assume the cost-agnostic scenario, where
the training distribution of classes is long-tailed while the testing distribution of classes is …

Which Model to Transfer? A Survey on Transferability Estimation

Y Ding, B Jiang, A Yu, A Zheng, J Liang - arXiv preprint arXiv:2402.15231, 2024 - arxiv.org
Transfer learning methods endeavor to leverage relevant knowledge from existing source
pre-trained models or datasets to solve downstream target tasks. With the increase in the …

[PDF][PDF] Working With What You've Got: Leveraging Mislabeled Datasets And Improving Imperfect Pretrained Models

W Gerych - 2023 - digital.wpi.edu
Resources such as OpenML and HuggingFace have made large datasets and powerful
pretrained models more accessible than ever for deep learning practitioners and …

Handling Learnwares from Heterogeneous Feature Spaces with Explicit Label Exploitation

P Tan, HT Liu, ZH Tan, ZH Zhou - The Thirty-eighth Annual Conference on … - openreview.net
The learnware paradigm aims to help users leverage numerous existing high-performing
models instead of starting from scratch, where a learnware consists of a well-trained model …

LLM4GCL: CAN LARGE LANGUAGE MODEL EM-POWER GRAPH CONTRASTIVE LEARNING?

Y Fang, X Tang, L Shang, D Fan, D Zha, Q Tan - openreview.net
Graph contrastive learning (GCL) has made significant strides in pre-training graph neural
networks (GNNs) without requiring human annotations. Previous GCL efforts have primarily …