Research map / Large language models / Language, safety and interpretability
Towards Fine-Grained Reasoning for Fake News Detection
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use FinerFact instead of KAN | fake news detection · PolitiFact | F1: 0.8539 -> 0.9172 | Helped |
| use FinerFact instead of KAN | fake news detection · PolitiFact | Acc: 0.8586 -> 0.9092 | Helped |
| use FinerFact instead of KAN | fake news detection · GossipCop | F1: 0.7713 -> 0.8685 | Helped |
| use FinerFact instead of KAN | fake news detection · GossipCop | Acc: 0.7766 -> 0.8320 | Helped |
| remove the attention prior from node-importance learning (ablation) | FinerFact · PolitiFact and GossipCop | model performance | Hurt |
| remove the user-based reasoning channel (ablation) | FinerFact · PolitiFact and GossipCop | model performance | Hurt |
| replace kernel-based representations with GNN-based aggregation (ablation) | FinerFact · PolitiFact and GossipCop | model performance | Hurt |
| assign equal weight to every node instead of learning node importance (ablation) | FinerFact · PolitiFact and GossipCop | model performance | Hurt |
| replace mutual-reinforcement evidence ranking with random evidence sampling (ablation) | FinerFact · PolitiFact and GossipCop | model performance | Hurt |
| replace BERT with a 1-layer Transformer encoder (ablation) | FinerFact · PolitiFact | F1: 0.9172 -> 0.8810 | Hurt |
| replace BERT with a 1-layer Transformer encoder (ablation) | FinerFact · GossipCop | F1: 0.8685 -> 0.8602 | Hurt |
| use around 11 kernels for fine-grained matching (ablation) | FinerFact · PolitiFact and GossipCop | model performance | Helped |
Closest papers by meaning
- Text-Guided Fine-grained Counterfactual Inference for Short Video Fake News Detection (AAAI 2025 · similarity 0.87)
- Bad Actor, Good Advisor: Exploring the Role of Large Language Models in Fake News Detection (AAAI 2024 · similarity 0.87)
- Unveiling Implicit Deceptive Patterns in Multi-Modal Fake News via Neuro-Symbolic Reasoning (AAAI 2024 · similarity 0.87)
- Reinforced Adaptive Knowledge Learning for Multimodal Fake News Detection (AAAI 2024 · similarity 0.86)
- CoT is Not the Chain of Truth: An Empirical Internal Analysis of Reasoning LLMs for Fake News Generation (ICML 2026 · similarity 0.85)
- Is LLMs Hallucination Usable? LLM-based Negative Reasoning for Fake News Detection (AAAI 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.