Large Language Models (LLMs) have demonstrated impressive performance in biomedical relation extraction, even in zero-shot scenarios. However, evaluating LLMs in this task remains challenging due to their ability to generate human-like text, often producing synonyms or abbreviations of gold-standard answers, making traditional automatic evaluation metrics unreliable. On the other hand, while human evaluation is more reliable, it is costly and time-consuming, making it impractical for real-world applications. This paper investigates the use of LLMs-as-the-Judge as an alternative evaluation method for biomedical relation extraction. We benchmark 8 LLMs as judges to evaluate the responses generated by 5 other LLMs across 3 biomedical relation extraction datasets. Unlike other text-generation tasks, we observe that LLM-based judges perform quite poorly (usually below 50% accuracy) in the biomedical relation extraction task. Our findings reveal that it happens mainly because relations extracted by LLMs do not adhere to any standard format. To address this, we propose structured output formatting for LLM-generated responses that helps LLM-Judges to improve their performance by about 15% (on average). We also introduce a domain adaptation technique to further enhance LLM-Judge performance by effectively transferring knowledge between datasets. We release both our human-annotated and LLM-annotated judgment data (36k samples in total) for public use here: https://github.com/tahmedge/llm_judge_biomedical_re.

Position bias, the tendency of Large Language Models (LLMs) to select content based on its structural position in a document rather than its semantic relevance, has been viewed as a key limitation in automatic summarization. To measure position bias, prior studies rely heavily on n-gram matching techniques, which fail to capture semantic relationships in abstractive summaries where content is extensively rephrased. To address this limitation, we apply a cross-encoder-based alignment method that jointly processes summary-source sentence pairs, enabling more accurate identification of semantic correspondences even when summaries substantially rewrite the source. Experiments with five LLMs across six summarization datasets reveal significantly different position bias patterns than those reported by traditional metrics. Our findings suggest that these patterns primarily reflect rational adaptations to document structure and content rather than true model limitations. Through controlled experiments and analyses across varying document lengths and multi-document settings, we show that LLMs use content from all positions more effectively than previously assumed, challenging common claims about “lost-in-the-middle” behaviour.

Building Agent Assistants that can help improve customer service support requires inputs from industry users and their customers, as well as knowledge about state-of-the-art Natural Language Processing (NLP) technology. We combine expertise from academia and industry to bridge the gap and build task/domain-specific Neural Agent Assistants (NAA) with three high-level components for: (1) Intent Identification, (2) Context Retrieval, and (3) Response Generation. In this paper, we outline the pipeline of the NAA’s core system and also present three case studies in which three industry partners successfully adapt the framework to find solutions to their unique challenges. Our findings suggest that a collaborative process is instrumental in spurring the development of emerging NLP models for Conversational AI tasks in industry. The full reference implementation code and results are available at https://github.com/VectorInstitute/NAA.

Automatically generating question answer (QA) pairs from the rapidly growing coronavirus-related literature is of great value to the medical community. Creating high quality QA pairs would allow researchers to build models to address scientific queries for answers which are not readily available in support of the ongoing fight against the pandemic. QA pair generation is, however, a very tedious and time consuming task requiring domain expertise for annotation and evaluation. In this paper we present our contribution in addressing some of the challenges of building a QA system without gold data. We first present a method to create QA pairs from a large semi-structured dataset through the use of transformer and rule-based models. Next, we propose a means of engaging subject matter experts (SMEs) for annotating the QA pairs through the usage of a web application. Finally, we demonstrate some experiments showcasing the effectiveness of leveraging active learning in designing a high performing model with a substantially lower annotation effort from the domain experts.