Research map / Large language models / Language, safety and interpretability
PHPFND: Detecting Fake News via Post-Hoc Processing of LLMs Hallucination
Jinke Ma, Jiachen Ma, Wei Zhang, Yong Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add hallucination detection module (ISHD) (ablation) | PHPFND · Weibo21 | macF1: 0.823 -> 0.942 | Improved |
| add hallucination detection module (ISHD) (ablation) | PHPFND · Pheme | macF1: 0.813 -> 0.925 | Improved |
| add hallucination correction (ablation) | PHPFND · Weibo21 | macF1: 0.910 -> 0.942 | Improved |
| add hallucination correction (ablation) | PHPFND · Pheme | macF1: 0.905 -> 0.925 | Improved |
| add feature enhancement (HDFE) (ablation) | PHPFND · Weibo21 | macF1: 0.871 -> 0.942 | Improved |
| add feature enhancement (HDFE) (ablation) | PHPFND · Pheme | macF1: 0.858 -> 0.925 | Improved |
| add hallucination-guided attention fusion (ablation) | PHPFND · Weibo21 | macF1: 0.901 -> 0.942 | Improved |
| add hallucination-guided attention fusion (ablation) | PHPFND · Pheme | macF1: 0.880 -> 0.925 | Improved |
| add feature separation (ablation) | PHPFND · Weibo21 | macF1: 0.906 -> 0.942 | Improved |
| add feature separation (ablation) | PHPFND · Pheme | macF1: 0.912 -> 0.925 | Improved |
| use hallucination signals (HDFE) vs plain encoder baseline on hallucination-inducing samples | PHPFND · Weibo21 hallucinated samples | error rate: 32.1% -> 18.7% | Improved |
| use hallucination signals (HDFE) vs plain encoder baseline on hallucination-inducing samples | PHPFND · Pheme hallucinated samples | error rate: 35.6% -> 19.2% | Improved |
Closest papers by meaning
- LLM-Check: Investigating Detection of Hallucinations in Large Language Models (NeurIPS 2024 · similarity 0.91)
- Is LLMs Hallucination Usable? LLM-based Negative Reasoning for Fake News Detection (AAAI 2025 · similarity 0.90)
- Scalable Token-Level Hallucination Detection in Large Language Models (NeurIPS 2026 · similarity 0.89)
- ANAH-v2: Scaling Analytical Hallucination Annotation of Large Language Models (NeurIPS 2024 · similarity 0.88)
- Bad Actor, Good Advisor: Exploring the Role of Large Language Models in Fake News Detection (AAAI 2024 · similarity 0.88)
- HaDeMiF: Hallucination Detection and Mitigation in Large Language Models (ICLR 2025 · similarity 0.88)
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.
