Combining Hierachical VAEs with LLMs for clinically meaningful timeline summarisation in social media
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914681097027584 |
|---|---|
| author | Song, Jiayu Chim, Jenny Tsakalidis, Adam Ive, Julia Atzil-Slonim, Dana Liakata, Maria |
| author_facet | Song, Jiayu Chim, Jenny Tsakalidis, Adam Ive, Julia Atzil-Slonim, Dana Liakata, Maria |
| contents | We introduce a hybrid abstractive summarisation approach combining hierarchical VAE with LLMs (LlaMA-2) to produce clinically meaningful summaries from social media user timelines, appropriate for mental health monitoring. The summaries combine two different narrative points of view: clinical insights in third person useful for a clinician are generated by feeding into an LLM specialised clinical prompts, and importantly, a temporally sensitive abstractive summary of the user's timeline in first person, generated by a novel hierarchical variational autoencoder, TH-VAE. We assess the generated summaries via automatic evaluation against expert summaries and via human evaluation with clinical experts, showing that timeline summarisation by TH-VAE results in more factual and logically coherent summaries rich in clinical utility and superior to LLM-only approaches in capturing changes over time. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_16240 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Combining Hierachical VAEs with LLMs for clinically meaningful timeline summarisation in social media Song, Jiayu Chim, Jenny Tsakalidis, Adam Ive, Julia Atzil-Slonim, Dana Liakata, Maria Computation and Language Artificial Intelligence We introduce a hybrid abstractive summarisation approach combining hierarchical VAE with LLMs (LlaMA-2) to produce clinically meaningful summaries from social media user timelines, appropriate for mental health monitoring. The summaries combine two different narrative points of view: clinical insights in third person useful for a clinician are generated by feeding into an LLM specialised clinical prompts, and importantly, a temporally sensitive abstractive summary of the user's timeline in first person, generated by a novel hierarchical variational autoencoder, TH-VAE. We assess the generated summaries via automatic evaluation against expert summaries and via human evaluation with clinical experts, showing that timeline summarisation by TH-VAE results in more factual and logically coherent summaries rich in clinical utility and superior to LLM-only approaches in capturing changes over time. |
| title | Combining Hierachical VAEs with LLMs for clinically meaningful timeline summarisation in social media |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2401.16240 |