MmAP: Multi-Modal Alignment Prompt for Cross-Domain Multi-Task Learning

Authors: Yi Xin, Junlong Du, Qiang Wang, Ke Yan, Shouhong Ding

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Comprehensive experiments on two large multi-task learning datasets demonstrate that our method achieves significant performance improvements compared to full fine-tuning while only utilizing approximately 0.09% of trainable parameters.
Researcher Affiliation Collaboration 1State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China 2Youtu Lab, Tencent xinyi@smail.nju.edu.cn, {jeffdu, albertqwang, kerwinyan, ericshding}@tencent.com
Pseudocode No The paper describes mathematical formulations and processes but does not include any pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any explicit statements about releasing source code or include a link to a code repository for its methodology.
Open Datasets Yes Following prior MTL works (Shen et al. 2021; Long et al. 2017), we consider Office-Home (Venkateswara et al. 2017) and Mini Domain Net (Zhou et al. 2021) datasets to construct our benchmark.
Dataset Splits No The paper states it 'randomly select 10% (6-shot per class) and 20% (12-shot per class) samples from Office-Home, and 1% (3shot per class) and 2% (6-shot per class) samples from Mini Domain Net for training' but does not explicitly detail the validation split proportions or methodology.
Hardware Specification Yes All experiments are conducted using the Py Torch toolkit on NVIDIA V100 GPU
Software Dependencies No The paper mentions 'Py Torch toolkit' but does not specify its version number or any other software dependencies with their versions.
Experiment Setup Yes We maintain consistent hyperparameter settings across all parameter efficient tuning methods. Specifically, we use a batch size of 16/4 and train for 5 epochs for Office-Home/Mini Domain Net. We employ the SGD optimizer with a learning rate of 0.0035.