Towards Better Multi-task Learning: A Framework for Optimizing Dataset Combinations in Large Language Models
To efficiently select optimal dataset combinations for enhancing multi-task learning (MTL) performance in large language models, we proposed a novel framework that leverages a neural network to predict the best dataset combinations. The framework iteratively refines the selection, greatly improving efficiency, while being model-, dataset-, and domain-independent. Through experiments on 12 biomedical datasets across four tasks - named entity recognition, relation extraction, event extraction, and text classification-we demonstrate that our approach effectively identifies better combinations, even for tasks that may seem unpromising from a human perspective. This verifies that our framework provides a promising solution for maximizing MTL potential.
Code (0)
등록된 구현이 없습니다.
Tasks
Event ExtractionMulti-Task Learningnamed-entity-recognitionNamed Entity RecognitionRelation Extractiontext-classificationText ClassificationSimilar Papers 제목 키워드 기반
A Generative Adversarial Framework for Optimizing Image Matting and Harmonization Simultaneously
Image matting and image harmonization are two important tasks in image composition. Image matting, aiming to achieve foreground boundary details, and image harmonization, aiming to make the background compatible with the…
Image HarmonizationImage MattingLearning Interpretable Models Through Multi-Objective Neural Architecture Search
Monumental advances in deep learning have led to unprecedented achievements across various domains. While the performance of deep neural networks is indubitable, the architectural design and interpretability of such mode…
Explainable Artificial Intelligence (XAI)image-classificationImage ClassificationNeural Architecture SearchCost-Sensitive Self-Training for Optimizing Non-Decomposable Metrics
Self-training based semi-supervised learning algorithms have enabled the learning of highly accurate deep neural networks, using only a fraction of labeled data. However, the majority of work on self-training has focused…
Task-Conditional Flow Matching for Balanced Multilingual Text Embedding Adaptation
Multilingual text embedding models are commonly adapted using a single training objective across diverse tasks, despite different tasks requiring fundamentally different optimization strategies. We introduce Task-Conditi…
Optimizing Byte-level Representation for End-to-end ASR
We propose a novel approach to optimizing a byte-level representation for end-to-end automatic speech recognition (ASR). Byte-level representation is often used by large scale multilingual ASR systems when the character …
Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Quantizationspeech-recognition+1