Pseudo-labeling¶
Pseudo-labeling is a semi-supervised learning technique that automatically generates labels for unlabeled data, typically using model predictions, external systems (such as large language models), or heuristics. In fake news detection, pseudo-labels can be generated by LLMs and used to augment training data when human annotations are scarce.
Key observations¶
LLM-generated pseudo-labels have varying quality: While LLMs can generate pseudo-labels at scale without human effort, their individual reliability is often limited. Aggregation, confidence filtering, and integration with other signals (e.g., graph structure) are necessary to leverage pseudo-labels effectively.
Confidence-based filtering improves quality: Selecting only high-confidence pseudo-labels (e.g., top 5% confidence scores) significantly improves downstream detection performance by reducing noise in the training signal.
Structural integration enhances reliability: Pseudo-labels propagated through graph structures can achieve higher quality than isolated predictions by leveraging consensus across related samples. Masking mechanisms prevent over-reliance on individual noisy labels.
Related topics¶
- Large Language Models — systems that generate pseudo-labels
- Semi Supervised Learning — learning paradigm leveraging unlabeled data
- Weak supervision — learning from noisy or imperfect labels
- Label propagation — propagating pseudo-labels through networks or graphs
Key papers¶
- Hu, Hu & Zhang (2025) — Synergizing LLMs with Global Label Propagation for Multimodal Fake News Detection — integrates LLM-generated pseudo labels with graph-based label propagation for multimodal detection
- Su et al. (2023) — Adapting Fake News Detection to the Era of Large Language Models — discusses LLM-based pseudo-labeling for fake news detection