Trained Model

Trained models, the output of machine learning processes, are central to numerous applications, with current research focusing on improving their accuracy, reliability, and efficiency. This involves developing techniques like custom transfer learning to optimize performance for specific tasks (e.g., medical image analysis, gait analysis), finetuning methods to calibrate confidence levels and mitigate overconfidence in large language models, and novel unlearning algorithms to selectively remove sensitive data. These advancements are crucial for enhancing the trustworthiness and ethical deployment of AI systems across diverse fields, from healthcare to legal document processing.

Papers

July 15, 2024

Deep-Learning-Based Markerless Pose Estimation Systems in Gait Analysis: DeepLabCut Custom Training and the Refinement Function
Giulia Panconi, Stefano Grasso, Sara Guarducci, Lorenzo Mucchi, Diego Minciacchi, Riccardo Bravi
Pose Estimation Gait Analysis Markerless Tracking Adaptive Refinement Gait Parameter Trained Model

May 31, 2024

LACIE: Listener-Aware Finetuning for Confidence Calibration in Large Language Models
Elias Stengel-Eskin, Peter Hase, Mohit Bansal
Confidence Calibration Answer Correctness Listening Effort Disentangling Confidence Score Distribution Trained Model Msi H

March 12, 2024

Efficient Knowledge Deletion from Trained Models through Layer-wise Partial Machine Unlearning
Vinay Chakravarthi Gogineni, Esmaeil S. Nadimi
Machine Unlearning Unlearning Framework Trained Model Knowledge Removal

December 12, 2023

Classifying complex documents: comparing bespoke solutions to large language models
Glen Hopkins, Kristjan Kalm
Large Language Model Classification Task Classification Algorithm Fine Tuned Large Language Model Rich Document Trained Model

October 19, 2023

Streamlining Brain Tumor Classification with Custom Transfer Learning in MRI Images
Javed Hossain, Md. Touhidul Islam, Md. Taufiqul Haque Khan Tusar
Deep Convolutional Neural Network Brain Tumor Magnetic Resonance Image Hidden Layer Pre Trained Convolutional Neural Network Trained Model

September 29, 2023

LoRA ensembles for large language model fine-tuning
Xi Wang, Laurence Aitchison, Maja Rudolph
Large Language Model Deep Ensemble Ensemble Approach Trained Model

December 28, 2022

Customizing Knowledge Graph Embedding to Improve Clinical Study Recommendation
Xiong Liu, Iya Khalil, Murthy Devarakonda
Knowledge Graph Jina Embeddings Knowledge Graph Embeddings Graph Embeddings Trained Model

June 30, 2022

COVID Detection and Severity Prediction with 3D-ConvNeXt and Custom Pretrainings
Daniel Kienzle, Julian Lorenz, Robin Schön, Katja Ludwig, Rainer Lienhart
Covid 19 COVID 19 Detection Severity Prediction ConvNeXt Model Lung CT Trained Model

February 8, 2022

Trained Model in Supervised Deep Learning is a Conditional Risk Minimizer
Yutong Xie, Dufan Wu, Bin Dong, Quanzheng Li
Unsupervised Learning ImageNet Dataset Supervised Deep Learning Supervised Learning Problem Trained Model Competing Risk Conditional Stochastic Optimization

December 13, 2021

Ex-Model: Continual Learning from a Stream of Trained Models
Antonio Carta, Andrea Cossu, Vincenzo Lomonaco, Davide Bacciu
Full Model Continual LEArning Continual Learning Link Stream Continual Learning Algorithm Model Paradigm Trained Model

Trained Model

Papers

Deep-Learning-Based Markerless Pose Estimation Systems in Gait Analysis: DeepLabCut Custom Training and the Refinement Function

LACIE: Listener-Aware Finetuning for Confidence Calibration in Large Language Models

Efficient Knowledge Deletion from Trained Models through Layer-wise Partial Machine Unlearning

Classifying complex documents: comparing bespoke solutions to large language models

Streamlining Brain Tumor Classification with Custom Transfer Learning in MRI Images

LoRA ensembles for large language model fine-tuning

Customizing Knowledge Graph Embedding to Improve Clinical Study Recommendation

COVID Detection and Severity Prediction with 3D-ConvNeXt and Custom Pretrainings

Trained Model in Supervised Deep Learning is a Conditional Risk Minimizer

Ex-Model: Continual Learning from a Stream of Trained Models