The PerAnsSumm Shared Task at CL4Health@NAACL 2025 focused on Perspective-Aware Summarization of Healthcare Q/A forums, requiring participants to extract and summarize spans based on predefined perspective categories. Our approach leveraged LLM-based zero-shot prompting enhanced by semantically-similar In-Context Learning (ICL) examples. Using Qwen-Turbo with 20 exemplar samples retrieved through NV-Embed-v2 embeddings, we achieved a mean score of 0.58 on Task A (span identification) and Task B (summarization) mean scores of 0.36 in Relevance and 0.28 in Factuality, finishing 12th on the final leaderboard. Notably, our system achieved higher precision in strict matching (0.20) than the top-performing system, demonstrating the effectiveness of our post-processing techniques. In this paper, we detail our ICL approach for adapting Large Language Models to Perspective-Aware Medical Summarization, analyze the improvements across development iterations, and finally discuss both the limitations of the current evaluation framework and future challenges in modeling this task. We release our code for reproducibility.

This paper presents our approach to the CLPsych 2025 (Tseriotou et al., 2025) shared task, where our proposed system implements a comprehensive solution using In-Context Learning (ICL) with vector similarity to retrieve relevant examples that guide Large Language Models (LLMs) without specific fine-tuning. We leverage ICL to analyze self-states and mental health indicators across three tasks. We developed a pipeline architecture using Ollama, where we are running Llama 3.3 70B locally and specialized vector databases for post- and timeline-level examples. We experimented with different numbers of retrieved examples (k=5 and k=10) to optimize performance. Our results demonstrate the effectiveness of ICL for clinical assessment tasks, particularly when dealing with limited training data in sensitive domains. The system shows strong performance across all tasks, with particular strength in capturing self-state dynamics.

Healthcare providers spend a significant amount of time reading and synthesizing electronic health records (EHRs), negatively impacting patient outcomes and causing provider burnout. Traditional supervised machine learning approaches using large language models (LLMs) to summarize clinical text have struggled due to hallucinations and lack of relevant training data. Here, we present a novel, simplified solution for the “Discharge Me!” shared task. Our approach mimics human clinical workflow, using pre-trained LLMs to answer specific questions and summarize the answers obtained from discharge summaries and other EHR sections. This method (i) avoids hallucinations through hybrid-RAG/zero-shot contextualized prompting; (ii) requires no extensive training or fine-tuning; and (iii) is adaptable to various clinical tasks.