representational collapse

4 papers

• Don't Just Chase “Highlighted Tokens” in MLLMs: Revisiting Visual Holistic Context Retention
• On Vanishing Gradients, Over-Smoothing, and Over-Squashing in GNNs: Bridging Recurrent and Graph Learning
• Tracing the Representation Geometry of Language Models from Pretraining to Post-training
• What Happens During the Loss Plateau? Understanding Abrupt Learning in Transformers