On Conditional and Compositional Language Model Differentiable Prompting

Authors: Jonathan Pilault, Can Liu, Mohit Bansal, Markus Dreyer

IJCAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We present extensive empirical and theoretical analysis and show that PROPS consistently surpasses other PLM adaptation techniques, and often improves upon fully fine-tuned models, on compositional generalization tasks, controllable summarization and multilingual translation, while needing fewer trainable parameters.
Researcher Affiliation Collaboration Jonathan Pilault1 , Can Liu2, Mohit Bansal3, Markus Dreyer2 1Mila Qu ebec AI Institute, Polytechnique Montr eal 2Amazon Alexa, 3University of North Carolina at Chapel Hill
Pseudocode Yes As summarized in Algorithm 1, each condition sequence SC = ct|t {1, . . . , TC} C, where TC is the sequence length of condition C C and C is the set of conditions, is first encoded by a Condition Encoder f( ).
Open Source Code No The code and datasets will be made publicly available.
Open Datasets Yes We study four Conditional Natural Language Generation (CNLG) datasets SCAN [Lake and Baroni, 2018], Europarl [Koehn, 2005], XSum [Narayan et al., 2018] and Topic-CNN-DM [Mrini et al., 2021a].
Dataset Splits No Each language pair direction has 1M training and 100K testing examples.
Hardware Specification No The paper does not provide specific hardware details such as GPU/CPU models, processor types, or memory amounts used for running experiments.
Software Dependencies No The paper does not provide specific software dependency details with version numbers (e.g., Python, PyTorch, or CUDA versions) needed to replicate the experiment.
Experiment Setup Yes We describe our datasets, training and evaluation setup in Appendix E.