# Structured Claim-Level Discourse Representations for Dense Health Narratives

arXiv: https://arxiv.org/abs/2609.18905 · HTML: https://mlsanity.com/paper/2609.18905/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 55 references that are in the map.
- First posted on arXiv 16 Sep 2026.

## Builds on (most cited first, 22 of 22)

- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [End-to-end Neural Coreference Resolution](https://mlsanity.com/paper/1707.07045/) (2017)
- [Trust or Escalate: LLM Judges with Provable Guarantees for Human Agreement](https://mlsanity.com/paper/2407.18370/) (2024)
- [CoAnnotating: Uncertainty-Guided Work Allocation between Human and Large Language Models for Data Annotation](https://mlsanity.com/paper/2310.15638/) (2023)
- [Stance Detection with Bidirectional Conditional Encoding](https://mlsanity.com/paper/1606.05464/) (2016)
- [Argumentation Mining in User-Generated Web Discourse](https://mlsanity.com/paper/1601.02403/) (2016)
- [LibriSpeech-PC: Benchmark for Evaluation of Punctuation and Capitalization Capabilities of end-to-end ASR Models](https://mlsanity.com/paper/2310.02943/) (2023)
- [MEGAnno+: A Human-LLM Collaborative Annotation System](https://mlsanity.com/paper/2402.18050/) (2024)
- [Multicalibration for Confidence Scoring in LLMs](https://mlsanity.com/paper/2404.04689/) (2024)
- [What is the Essence of a Claim? Cross-Domain Claim Identification](https://mlsanity.com/paper/1704.07203/) (2017)
- [Fine-Grained Argument Unit Recognition and Classification](https://mlsanity.com/paper/1904.09688/) (2019)
- [Keeping Humans in the Loop: Human-Centered Automated Annotation with Generative AI](https://mlsanity.com/paper/2409.09467/) (2024)
- [Can External Validation Tools Improve Annotation Quality for LLM-as-a-Judge?](https://mlsanity.com/paper/2507.17015/) (2025)
- [Clinical BERTScore: An Improved Measure of Automatic Speech Recognition Performance in Clinical Settings](https://mlsanity.com/paper/2303.05737/) (2023)
- [Stance Prediction for Contemporary Issues: Data and Experiments](https://mlsanity.com/paper/2006.00052/) (2020)
- [Disordered-DABS: A Benchmark for Dynamic Aspect-Based Summarization in Disordered Texts](https://mlsanity.com/paper/2402.10554/) (2024)
- [MODABS: Multi-Objective Learning for Dynamic Aspect-Based Summarization](https://mlsanity.com/paper/2406.03479/) (2024)
- [Reliable Annotations with Less Effort: Evaluating LLM-Human Collaboration in Search Clarifications](https://mlsanity.com/paper/2507.00543/) (2025)
- [From Argumentation to Deliberation: Perspectivized Stance Vectors for Fine-grained (Dis)agreement Analysis](https://mlsanity.com/paper/2502.09644/) (2025)
- [Crowdsourcing-Based Knowledge Graph Construction for Drug Side Effects Using Large Language Models with an Application on Semaglutide](https://mlsanity.com/paper/2504.04346/) (2025)
- [The Overlooked Repetitive Lengthening Form in Sentiment Analysis](https://mlsanity.com/paper/2604.01268/) (2026)

## Abstract (arXiv)

Health discourse in social media videos often contains densely entangled claims spanning multiple thematic aspects, stances, evidential frames, and rhetorical functions within short conversational spans. Existing approaches largely rely on coarse topic-level, sentiment-based, or stance-oriented representations that do not adequately capture this structure. Our analysis identifies an average of 13.22 atomic claims per minute, motivating richer claim-level discourse representations. We introduce a structured framework for claim-level discourse analysis in dense health narratives. Our framework models discourse through tuples linking atomic claims with thematic aspects, stance, and multidimensional pragmatic discourse attributes. To support this setting, we construct a benchmark spanning four health domains with 1,191 manually annotated claims from 60 videos. Using this framework, we evaluate automated structured discourse analysis under different discourse context settings. Results show that current LLMs achieve strong performance on thematic categorization and stance prediction, but struggle with high-dimensional pragmatic profiling. We also find that different discourse tasks benefit from different forms of contextual reasoning, suggesting that future systems may require task decomposition and specialized inference strategies.