Probing factual knowledge in Pre-trained Language Models (PLMs) using prompts has indirectly implied that language models (LMs) can be treated as knowledge bases. To this end, this phenomena has been effective, especially when these LMs are fine-tuned towards not just data, but also to the style or linguistic pattern of the prompts themselves. We observe that, satisfying a particular linguistic pattern in prompts is an unsustainable, time-consuming constraint in the probing task, especially because, they are often manually designed and the range of possible prompt template patterns can vary depending on the prompting task. To alleviate this constraint, we propose using a position-attention mechanism to capture positional information of each word in a prompt relative to the mask to be filled, hence avoiding the need to re-construct prompts when the prompts’ linguistic pattern changes. Using our approach, we demonstrate the ability of eliciting answers (in a case study on health outcome generation) to not only common prompt templates like Cloze and Prefix, but also rare ones too, such as Postfix and Mixed patterns whose masks are respectively at the start and in multiple random places of the prompt. More so, using various biomedical PLMs, our approach consistently outperforms a baseline in which the default PLMs representation is used to predict masked tokens.


Position-based Prompting for Health Outcome Generation

Jointly detecting and classifying outcomes (measurements or observations used to capture and assess the effect of a treatment in clinical trials) leverages the relationship between an outcome span mentioned in a clinical trial abstract and it's category or type. Treating the detection process as a token-level task and the classification process as a sentence-level task enables us utilize the structural correspondences between the two levels inorder to successfully simulataneously achieve detection and classification of this biomedical entity. Ultimately, we erase the need of multiple models for each task because we don't compromise the standalone performance and matter of fact outperform standalone performance. We also propose a scalable and flexible approach to align comparable datasets, which we adopt to align arbitrary classifications in prior dataset to standardardised classification for health outcomes.

Micheal Abaho

2

3

SHORT BIO