Search Results for author:
Found 28 papers, 17 papers with code
AvaTaR: Optimizing LLM Agents for Tool Usage via Contrastive Reasoning
Large language model (LLM) agents have demonstrated impressive capabilities in utilizing external tools and knowledge to boost accuracy and reduce hallucinations.
STaRK: Benchmarking LLM Retrieval on Textual and Relational Knowledge Bases
To address the gap, we develop STARK, a large-scale Semi-structure retrieval benchmark on Textual and Relational Knowledge Bases.
PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning
We present PyTorch Frame, a PyTorch-based framework for deep learning over multi-modal tabular data.
Ranked #1 on
Binary Classification
on kickstarter
GraphMETRO: Mitigating Complex Graph Distribution Shifts via Mixture of Aligned Experts
GraphMETRO employs a Mixture-of-Experts (MoE) architecture with a gating model and multiple expert models, where each expert model targets a specific distributional shift to produce a referential representation w. r. t.
TpuGraphs: A Performance Prediction Dataset on Large Tensor Computational Graphs
TpuGraphs provides 25x more graphs than the largest graph property prediction dataset (with comparable graph sizes), and 770x larger graphs on average compared to existing performance prediction datasets on machine learning programs.
Ranked #2 on
Runtime ranking
on TpuGraphs Layout mean
Communication-Free Distributed GNN Training with Vertex Cut
Here, we introduce CoFree-GNN, a novel distributed GNN training framework that significantly speeds up the training process by implementing communication-free training.
Ranked #2 on
Node Classification
on Reddit
Learning Large Graph Property Prediction via Graph Segment Training
Here we propose Graph Segment Training (GST), a general framework that utilizes a divide-and-conquer approach to allow learning large graph property prediction with a constant memory footprint.
AutoTransfer: AutoML with Knowledge Transfer -- An Application to Graph Neural Networks
Experiments demonstrate that (i) our proposed task embedding can be computed efficiently, and that tasks with similar embeddings have similar best-performing architectures; (ii) AutoTransfer significantly improves search efficiency with the transferred design priors, reducing the number of explored architectures by an order of magnitude.
Relational Multi-Task Learning: Modeling Relations between Data and Tasks
Here we introduce a novel relational multi-task learning setting where we leverage data point labels from auxiliary tasks to make more accurate predictions on the new task.
TuneUp: A Simple Improved Training Strategy for Graph Neural Networks
Extensive evaluation of TuneUp on five diverse GNN architectures, three types of prediction tasks, and both transductive and inductive settings shows that TuneUp significantly improves the performance of the base GNN on tail nodes, while often even improving the performance on head nodes.
Learning Large-scale Subsurface Simulations with a Hybrid Graph Network Simulator
To model complex reservoir dynamics at both local and global scale, HGNS consists of a subsurface graph neural network (SGNN) to model the evolution of fluid flows, and a 3D-U-Net to model the evolution of pressure.
Learning Backward Compatible Embeddings
We formalize the problem where the goal is for the embedding team to keep updating the embedding version, while the consumer teams do not have to retrain their models.
Open-World Semi-Supervised Learning
Here, we introduce a novel open-world semi-supervised learning setting that formalizes the notion that novel classes may appear in the unlabeled test data.
Ranked #2 on
Open-World Semi-Supervised Learning
on ImageNet-100
Coresets for Robust Training of Deep Neural Networks against Noisy Labels
Modern neural networks have the capacity to overfit noisy labels frequently found in real-world datasets.
Coresets for Robust Training of Neural Networks against Noisy Labels
Modern neural networks have the capacity to overfit noisy labels frequently found in real-world datasets.
Ranked #38 on
Image Classification
on mini WebVision 1.0
Concept Learners for Few-Shot Learning
Developing algorithms that are able to generalize to a novel task given only a few labeled examples represents a fundamental challenge in closing the gap between machine- and human-level performance.
Heteroskedastic and Imbalanced Deep Learning with Adaptive Regularization
Real-world large-scale datasets are heteroskedastic and imbalanced -- labels have varying levels of uncertainty and label distributions are long-tailed.
Ranked #11 on
Image Classification
on WebVision-1000
Learning Temporal Action Proposals With Fewer Labels
Most current methods for training action proposal modules rely on fully supervised approaches that require large amounts of annotated temporal action intervals in long video sequences.
Ranked #3 on
Semi-Supervised Action Detection
on ActivityNet-1.3
Delving Deep Into Hybrid Annotations for 3D Human Recovery in the Wild
Specifically, we focus on the challenging task of in-the-wild 3D human recovery from single images when paired 3D annotations are not fully available.
Few-Shot Video Classification via Temporal Alignment
In this paper, we propose Temporal Alignment Module (TAM), a novel few-shot learning framework that can learn to classify a previous unseen video.
Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss
Deep learning algorithms can fare poorly when the training dataset suffers from heavy class-imbalance but the testing criterion requires good generalization on less frequent classes.
Ranked #4 on
Long-tail learning with class descriptors
on CUB-LT
Disentangling Content and Style via Unsupervised Geometry Distillation
It is challenging to disentangle an object into two orthogonal spaces of content and style since each can influence the visual observation differently and unpredictably.
TransGaGa: Geometry-Aware Unsupervised Image-to-Image Translation
Extensive experiments demonstrate the superior performance of our method to other state-of-the-art approaches, especially in the challenging near-rigid and non-rigid objects translation tasks.
CariGANs: Unpaired Photo-to-Caricature Translation
Facial caricature is an art form of drawing faces in an exaggerated way to convey humor or sarcasm.
Unsupervised Disentangling Structure and Appearance
It is challenging to disentangle an object into two orthogonal spaces of structure and appearance since each can influence the visual observation in a different and unpredictable way.
DNN Dataflow Choice Is Overrated
Many DNN accelerators have been proposed and built using different microarchitectures and program mappings.
Distributed, Parallel, and Cluster Computing
Pose-Robust Face Recognition via Deep Residual Equivariant Mapping
However, many contemporary face recognition models still perform relatively poor in processing profile faces compared to frontal faces.
Ranked #1 on
Face Identification
on IJB-A
Merge or Not? Learning to Group Faces via Imitation Learning
Given a large number of unlabeled face images, face grouping aims at clustering the images into individual identities present in the data.
