Abstract
We present two LLM-based approaches to zero-shot source-and-target belief prediction on FactBank: a unified system that identifies events, sources, and belief labels in a single pass, and a hybrid approach that uses a fine-tuned DeBERTa tagger for event detection. We show that multiple open-sourced, closed-source, and reasoning-based LLMs struggle with the task. Using the hybrid approach, we achieve new state-of-the-art results on FactBank and offer a detailed error analysis. Our approach is then tested on the Italian belief corpus ModaFact.
Related Papers
Frequency Dynamic Convolutions for Sound Event Detection2025-06-15DiCoRe: Enhancing Zero-shot Event Detection via Divergent-Convergent LLM Reasoning2025-06-05Towards real-time assessment of infrasound event detection capability using deep learning-based transmission loss estimation2025-06-03DIAMOND: An LLM-Driven Agent for Context-Aware Baseball Highlight Summarization2025-06-03Hybrid Disagreement-Diversity Active Learning for Bioacoustic Sound Event Detection2025-05-27CosyVoice 3: Towards In-the-wild Speech Generation via Scaling-up and Post-training2025-05-23Exploring the Potential of SSL Models for Sound Event Detection2025-05-17Multimodal Event Detection: Current Approaches and Defining the New Playground through LLMs and VLMs2025-05-16