Home Research Papers Datasets State of the Art Pricing

Discover, visualize, and connect AI research papers. Explore the latest trends and insights in artificial intelligence research.

Product

Home
Research Papers
About

Support

Contact
Terms of Service
Privacy Policy

© 2026 Papersgraph. All rights reserved.

methodology-8

Extreme Multi-Label Classification

3260 papers • 126 benchmarks • 313 datasets

Extreme Multi-Label Classification is a supervised learning problem where an instance may be associated with multiple labels. The two main problems are the unbalanced labels in the dataset and the amount of different labels.

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in extreme-multi-label-classification-41

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find extreme-multi-label-classification-41 models and implementations

3 papers 489

Datasets

EXTREME CLASSIFICATION

EXTREME CLASSIFICATION

LSHTC

Subtasks

No subtasks available.

Most implemented papers

Node Feature Extraction by Self-Supervised Multi-scale Neighborhood Prediction

Cho-Jui Hsieh, Wei-Cheng Chang, Hsiang-Fu Yu, I. Dhillon, Eli Chien, Jiong Zhang, O. Milenkovic•Thu Oct 28 2021

A new self-supervised learning framework, Graph Information Aided Node feature exTraction (GIANT), which makes use of the eXtreme Multi-label Classification (XMC) formalism, which is crucial for fine-tuning the language model based on graph information, and scales to large datasets.

171

Content

Introduction Benchmarks Datasets Subtasks Libraries Papers

mwydmuch/napkinXC

2 papers 58

0

Taming Pretrained Transformers for Extreme Multi-label Text Classification

Kai Zhong, Wei-Cheng Chang, Hsiang-Fu Yu, Yiming Yang, I. Dhillon•Mon May 06 2019

X-Transformer is proposed, the first scalable approach to fine-tuning deep transformer models for the XMC problem and achieves new state-of-the-art results on four XMC benchmark datasets.

253 0

Probabilistic Label Trees for Extreme Multi-label Classification

Marek Wydmuch, Kalina Jasinska-Kobus, K. Dembczynski, Mikhail Kuznetsov, R. Busa-Fekete•Tue Sep 22 2020

This paper thoroughly investigates probabilistic label trees (PLTs) which can be treated as a generalization of hierarchical softmax for multi-label problems and proves the consistency of PLTs for a wide spectrum of performance metrics.

24 0

Generalized test utilities for long-tail performance in extreme multi-label classification

Marek Wydmuch, Erik Schultheis, Wojciech Kotlowski, Rohit Babbar, Krzysztof Dembczy'nski•Wed Nov 08 2023

This paper analyzes generalized metrics budgeted at k in the expected test utility (ETU) framework and derives optimal prediction rules and construct computationally efficient approximations with provable regret guarantees and robustness against model misspecification.

5 0

In-Context Learning for Extreme Multi-Label Classification

O. Khattab, Thomas Demeester, Karel D’Oosterlinck, François Remy, Chris Develder, Christopher Potts•Sun Jan 21 2024

This work proposes a general program, $\texttt{Infer--Retrieve--Rank}$, that defines multi-step interactions between LMs and retrievers to efficiently tackle multi-label classification problems and attains state-of-the-art results across three benchmarks.

7 0

Deep Extreme Multi-label Learning

Junchi Yan, Wenjie Zhang, Liwei Wang, Xiangfeng Wang, H. Zha•Tue Apr 11 2017

This paper proposes a practical deep embedding method for extreme multi-label classification, which harvests the ideas of non-linear embedding and graph priors-based label space modeling simultaneously.

127 0

Revisiting the Vector Space Model: Sparse Weighted Nearest-Neighbor Method for Extreme Multi-Label Classification

Tatsuhiro Aoshima, Kei Kobayashi, Mihoko Minami•Sun Feb 11 2018

A novel approach to XMLC termed the Sparse Weighted Nearest-Neighbor Method, which can be derived as a fast implementation of state-of-the-art (SOTA) one-versus-rest linear classifiers for very sparse datasets with superior performance to the SOTA models on a dataset with 3 million labels.

0 0

Adversarial Extreme Multi-label Classification

B. Scholkopf, Rohit Babbar•Sun Mar 04 2018

This work poses the learning task in extreme classification with large number of tail-labels as learning in the presence of adversarial perturbations to motivate a robust optimization framework and equivalence to a corresponding regularized objective.

16 0

Adding a benchmark result helps the community track progress.

Extreme Multi-Label Classification | State-of-the-Art