New Method Could Unlock AI's Power to See and Describe Images with Unmatched Detail
19 Dec 2024
Researchers at the Mohamed bin Zayed University developed an AI model that can create text-based conversations tied to specific objects or regions in an image.
The Preprocessing and Training That HierSpeech++ Went Through
17 Dec 2024
We trained HierSpeech++ with a batch size of 160 for 1,000k steps on eight NVIDIA A6000 GPUs.
The Backbone Speech Synthesizer for HierSpeech++
17 Dec 2024
We propose a hierarchical speech synthesizer as the backbone speech synthesizer for HierSpeech++
New Research Reveals Vulnerabilities in Popular Art Protection Tools Against AI Theft
10 Dec 2024
Study shows existing AI-based protection tools can’t stop style mimicry, leaving artists vulnerable. New protective solutions are urgently needed.
FlowVid: Taming Imperfect Optical Flows: Training With Joint Spatial-temporal Conditions
9 Oct 2024
This paper proposes a consistent V2V synthesis framework by jointly leveraging spatial conditions and temporal optical flow clues within the source video.
Knowledge-Distillation-Based Adversarial Training for Robust Multi-Exit Neural Networks
30 Sept 2024
Discover NEO-KD, a knowledge-distillation strategy that boosts adversarial robustness in multi-exit neural networks.
Game Theory and Blockchain: Why DAG-Protocol Fails Against Greedy Behavior
18 Sept 2024
Explore a game-theoretical analysis of DAG-Protocol, revealing why honest mining fails to create a stable equilibrium against greedy players in blockchains.
Controlling Automated Vehicles on Large Lane-free Roundabouts (Extended Version): References
3 Sept 2024
Explore a detailed list of references instrumental in guiding the paper on Controlling Automated Vehicles on Large Lane-free Roundabouts.
Handling Complex Vehicle Movements: Techniques for Finding Cross-Points and Obstacles
3 Sept 2024
This Appendix provides detailed methods for finding cross-points when an ego vehicle (EV) and an obstacle have different motion types. Learn more.