Matryoshka representation learning
Learned representations are a central component in modern ML systems, serving a
multitude of downstream tasks. When training such representations, it is often the case that …
multitude of downstream tasks. When training such representations, it is often the case that …
Node feature extraction by self-supervised multi-scale neighborhood prediction
Learning on graphs has attracted significant attention in the learning community due to
numerous real-world applications. In particular, graph neural networks (GNNs), which take …
numerous real-world applications. In particular, graph neural networks (GNNs), which take …
Fast multi-resolution transformer fine-tuning for extreme multi-label text classification
Extreme multi-label text classification~(XMC) seeks to find relevant labels from an extreme
large label collection for a given text input. Many real-world applications can be formulated …
large label collection for a given text input. Many real-world applications can be formulated …
Extreme multi-label learning for semantic matching in product search
We consider the problem of semantic matching in product search: given a customer query,
retrieve all semantically related products from a huge catalog of size 100 million, or more …
retrieve all semantically related products from a huge catalog of size 100 million, or more …
Cascadexml: Rethinking transformers for end-to-end multi-resolution training in extreme multi-label classification
Abstract Extreme Multi-label Text Classification (XMC) involves learning a classifier that can
assign an input with a subset of most relevant labels from millions of label choices. Recent …
assign an input with a subset of most relevant labels from millions of label choices. Recent …
The effect of metadata on scientific literature tagging: A cross-field cross-model study
Due to the exponential growth of scientific publications on the Web, there is a pressing need
to tag each paper with fine-grained topics so that researchers can track their interested fields …
to tag each paper with fine-grained topics so that researchers can track their interested fields …
Finger: Fast inference for graph-based approximate nearest neighbor search
Approximate K-Nearest Neighbor Search (AKNNS) has now become ubiquitous in modern
applications, such as a fast search procedure with two-tower deep learning models. Graph …
applications, such as a fast search procedure with two-tower deep learning models. Graph …
Linear classifier: An often-forgotten baseline for text classification
Large-scale pre-trained language models such as BERT are popular solutions for text
classification. Due to the superior performance of these advanced methods, nowadays …
classification. Due to the superior performance of these advanced methods, nowadays …
Self-paced unified representation learning for hierarchical multi-label classification
Hierarchical Multi-Label Classification (HMLC) is a well-established problem that aims at
assigning data instances to multiple classes stored in a hierarchical structure. Despite its …
assigning data instances to multiple classes stored in a hierarchical structure. Despite its …
Dual-encoders for extreme multi-label classification
Dual-encoder (DE) models are widely used in retrieval tasks, most commonly studied on
open QA benchmarks that are often characterized by multi-class and limited training data. In …
open QA benchmarks that are often characterized by multi-class and limited training data. In …