From Questions to Trust Reports: A LLM-IR Framework for the TREC 2025 DRAGUN Track

TL;DR

This paper presents a system that uses large language models and advanced retrieval techniques to generate trustworthiness reports for online news, aiming to improve user evaluation of information credibility.

Contribution

The paper introduces a novel LLM-IR framework combining question generation, semantic filtering, clustering, and reasoning-based query expansion for trustworthiness assessment.

Findings

Chain-of-Thought expansion improves retrieval relevance.

Re-ranking with monoT5 enhances trustworthiness detection.

Moderate quality in question generation indicates need for further improvement.

Abstract

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.