Cheers: Decoupling Patch Details from Semantic Representations Enables Unified Multimodal Comprehension and Generation Paper • 2603.12793 • Published Mar 13 • 38
In-Context Reinforcement Learning for Tool Use in Large Language Models Paper • 2603.08068 • Published Mar 9 • 43
Linear representations in language models can change dramatically over a conversation Paper • 2601.20834 • Published Jan 28 • 21
cisco-ai/SecureBERT2.0-code-vuln-detection Text Classification • 0.1B • Updated Oct 30, 2025 • 1.23k • 3