Papers with few-shot adaptation

6 papers
Systematic Analysis for Pretrained Language Model Priming for Parameter-Efficient Fine-tuning (2024.naacl-srw)

Copied to clipboard

Challenge: Parameter-efficient (PE) methods for adapting pre-trained language models to downstream tasks are still lacking in many cases.
Approach: They propose a general PE priming framework to enhance few-shot adaptation and generalization ability of PE methods.
Outcome: The proposed framework reveals that the best priming strategy facilitates adaptation to target tasks.
Prompts Can Play Lottery Tickets Well: Achieving Lifelong Information Extraction via Lottery Prompt Tuning (2023.acl-long)

Copied to clipboard

Challenge: Existing research on information extraction tasks focuses on one specific task, but in real-world scenarios, new data of different IE tasks and domains come in a stream over time.
Approach: They propose a parameter- and deployment-efficient prompt tuning method to evaluate the UIE system under a “lifelong learning” setting.
Outcome: The proposed method is able to learn new tasks without forgetting old ones and expand knowledge and functionalities without retraining the whole system.
On Measuring the Intrinsic Few-Shot Hardness of Datasets (2022.emnlp-main)

Copied to clipboard

Challenge: Recent work has shown that few-shot learning is successful for pre-trained models, but there is no concrete understanding of when and why it is successful.
Approach: They propose a simple metric that estimates few-shot hardness for a given dataset . they propose metric which exploits feature-space invariances between training and test samples .
Outcome: The proposed metric better accounts for few-shot hardness compared to existing notions and is 8-100x faster to compute.
How to Solve Few-Shot Abusive Content Detection Using the Data We Actually Have (2024.lrec-main)

Copied to clipboard

Challenge: Existing datasets for abusive language detection are expensive and lack of knowledge about the target is a challenge.
Approach: They propose to build models cheaply for a new target label set and/or language, using only a few training examples of the target domain.
Outcome: The proposed model improves monolingually and across languages using existing datasets and only a few-shots of the target domain.
Impressions: Visual Semiotics and Aesthetic Impact Understanding (2023.emnlp-main)

Copied to clipboard

Challenge: Existing image captioning and conditional generation models struggle to simulate plausible human responses to images.
Approach: They propose a dataset to investigate the semiotics of images and how visual features and design choices can elicit specific emotions, thoughts and beliefs.
Outcome: The proposed dataset improves existing models for image captioning and conditional generation.
HyperLoRA: Efficient Cross-task Generalization via Constrained Low-Rank Adapters Generation (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing approaches to adapt pre-trained language models (PLMs) to emerging tasks are costly and inefficient.
Approach: They propose a meta-network that generates task-specific weights without any optimization.
Outcome: The proposed approach has flexible generalization ability and superior performance over hypenetworks.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations