Fake-news classifier
According to the model card, this is an albert-base-v2 fine-tune that classifies titles as fake news or real news.
Open Source Model Profile · XSY
albert-base-v2-fakenews-discriminator is an ALBERT text-classification fine-tune from XSY. According to the model card, it distinguishes fake from real news titles.
albert-base-v2-fakenews-discriminator is published by XSY as a Transformers text-classification model. The captured configuration identifies AlbertForSequenceClassification with model type albert under Apache-2.0. According to the model card, it fine-tunes albert-base-v2 on news titles labeled fake or real.
According to the model card, this is an albert-base-v2 fine-tune that classifies titles as fake news or real news.
According to the model card, evaluation reports 0.9758 accuracy with 0.0910 validation loss after one epoch.
Captured configuration identifies AlbertForSequenceClassification with model type albert under Apache-2.0 licensing.
The model card documents learning rate 5e-05, batch size 16, seed 42, Adam, linear scheduling with 500 warmup steps, and one epoch.
Source: XSY/albert-base-v2-fakenews-discriminator
Captured: Unknown. Processed: 2026-09-07T19:34:38.719796+00:00.
albert-base-v2-fakenews-discriminator The dataset: Fake and real news dataset https://www.kaggle.com/clmentbisaillon/fake-and-real-news-dataset I use title and label to train the classifier label_0 : Fake news label_1 : Real news This model is a fine-tuned version of albert-base-v2 on an unknown dataset. It achieves the following results on the evaluation set: Loss: 0.0910 Accuracy: 0.9758 Model description More information needed Intended uses & limitations More information needed Training and evaluation data More information needed Training procedure Training hyperparameters The following hyperparameters were used during training:…
F001F002F003F004F005F006F009F010F011F012