From Questions to Trust Reports: A LLM-IR Framework for the TREC 2025 DRAGUN Track
This work addresses the need for tools to help users assess online news trustworthiness, though it is incremental with moderate improvements in question generation.
The paper tackled the problem of evaluating online news trustworthiness by developing a system that uses LLM-based question generation and retrieval-augmented reporting, resulting in improved relevance and domain trust through Chain-of-Thought query expansion and re-ranking compared to baselines.
The DRAGUN Track at TREC 2025 targets the growing need for effective support tools that help users evaluate the trustworthiness of online news. We describe the UR_Trecking system submitted for both Task 1 (critical question generation) and Task 2 (retrieval-augmented trustworthiness reporting). Our approach combines LLM-based question generation with semantic filtering, diversity enforcement using clustering, and several query expansion strategies (including reasoning-based Chain-of-Thought expansion) to retrieve relevant evidence from the MS MARCO V2.1 segmented corpus. Retrieved documents are re-ranked using a monoT5 model and filtered using an LLM relevance judge together with a domain-level trustworthiness dataset. For Task 2, selected evidence is synthesized by an LLM into concise trustworthiness reports with citations. Results from the official evaluation indicate that Chain-of-Thought query expansion and re-ranking substantially improve both relevance and domain trust compared to baseline retrieval, while question-generation performance shows moderate quality with room for improvement. We conclude by outlining key challenges encountered and suggesting directions for enhancing robustness and trustworthiness assessment in future iterations of the system.