PhysNote: Enhancing Physical Reasoning in Vision-Language AI
Discover PhysNote, a new framework boosting Vision-Language Models' physical reasoning with improved consistency and knowledge retention.
AVES-DPO: Reducing Hallucinations in LVLMs with Self-Correction
Discover AVES-DPO, a new self-corrected preference learning method that effectively reduces hallucinations in Large Vision-Language Models (LVLMs).
Super-DeepG: Certified Geometric Robustness for AI Models
Discover Super-DeepG, a tool for certifying geometric robustness in neural networks, ensuring AI reliability in safety-critical applications.
AI-Driven Generative Design for Hydrogen Gas Turbine Combustors
Discover how invertible neural networks optimize hydrogen gas turbine combustor designs for efficiency and low emissions in clean energy applications.
Adaptive ToR: Efficient Multi-Intent NLU Retrieval System
Discover Adaptive ToR, a complexity-aware tree retrieval method improving accuracy and efficiency in multi-intent natural language understanding.