[TMP-107] SciNoBo: A Collaborative AI Assistant in Science Communication
SciNoBo enhances science communication in health and climate by using AI to verify, simplify, and ground complex scientific statements.
SciNoBo is a microproject aimed at improving science communication, particularly in health and climate change, by integrating AI with science journalism. It supports science communicators, including journalists and policymakers, by using AI to identify, verify, and simplify complex scientific statements in mass media. The AI ensures accurate, evidence-based communication to non-expert audiences. Building on previous work with neuro-symbolic question-answering systems, the project leverages advanced language models, argumentation mining, claim verification, and text simplification. It explores technologies like instruction fine-tuning of large language models and retrieval augmentation to enhance natural language understanding. Aligned with the “Collaborative AI” theme, the system collaborates with science communicators to ground statements in scientific evidence (interactive grounding) and provide simplified explanations. This innovative AI solution aims to make complex topics more accessible and accurate for the public. The solution will be tested in a real-world scenario with OpenAIRE, utilizing OpenAIRE Research Graph (ORG) services in open science publications.
The project is divided into two phases that ran in parallel. The main focus in phase I is the construction of the data collections and the adaptations and improvements needed in PDF processing tools. Phase II deals with the development of the two subsystems: claim analysis and text simplification as well as their evaluation.
- Phase I: Two collections with News and scientific publications will be compiled in the areas of Health and Climate. The News collection will be built based on an existing dataset with News stories and ARC automated classification system in the areas of interest. The second collection with publications will be provided by OpenAIRE ORG service and further processed, managed and properly indexed by ARC SciNoBo toolkit. A small-scale annotation is foreseen by DFKI in support of the simplification subsystem.
- Phase II: We developed, fine tuned and evaluated the two subsystems. Concretely, the “claim analysis” subsystem encompasses (i) ARC previous work on “claim identification”, (ii) a retrieval engine fetching relevant scientific publications (based on our previous miniProject), and (iii) an evidence-synthesis module indicating whether the publications fetched and the scientists’ claims therein, support or refute the News claim under examination.
Tangible Outcomes
- Kotitsas, S., Kounoudis, P., Koutli, E., & Papageorgiou, H. (2024, March). Leveraging fine-tuned Large Language Models with LoRA for Effective Claim, Claimer, and Claim Object Detection. In Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 2540-2554).
https://aclanthology.org/2024.eacl-long.156/ - HCN dataset: news articles in the domain of Health and Climate Change. The dataset contains news articles, annotated with the major claim, claimer(s) and claim object(s).
https://github.com/iNoBo/news_claim_analysis - Website demo:
http://scinobo.ilsp.gr:1997/services - Services for claim identification and the retrieval engine
http://scinobo.ilsp.gr:1997/live-demo?HFSpace=inobo-scinobo-claim-verification.hf.space - Service for the text simplification
http://scinobo.ilsp.gr:1997/text-simplification
Partners
- ATHENA RC, Haris Papageorgiou
- German Research Centre for Artificial Intelligence (DFKI), Julián Moreno Schneider
- OpenAIRE, Natalia Manola