AI RESEARCH
Long Live Fine-Tuning: Task-Specific Transformers Outperform Zero-Shot LLMs for Misinformation Response Classification on Reddit
arXiv CS.CL
•
ArXi:2606.04274v1 Announce Type: new As large language models (LLMs) become default tools for online information verification, an implicit assumption follows them: that scale and general capability are sufficient for nuanced classification of misinformation dis