When2Speak Dataset: Enhancing Turn-Taking in Multi-Party AI Chats
Discover When2Speak, a dataset boosting Large Language Models' timing and turn-taking in multi-party conversations for better AI interactions.
X-Voice: Zero-Shot Voice Cloning in 30 Languages
Discover X-Voice, a zero-shot cross-lingual voice cloning model enabling seamless speech in 30 languages with AI-powered multilingual synthesis.
Nearly Optimal Attention Coresets for AI Efficiency
Discover how nearly optimal attention coresets improve AI efficiency by reducing memory use and boosting model performance in deep learning.
Boost LMO Optimization Speed with Implicit Gradient Transport
Discover how implicit gradient transport accelerates LMO-based optimization, improving convergence with minimal computational overhead.
AstroAlertBench: Benchmarking Multimodal LLMs in Astronomy
Discover AstroAlertBench, a benchmark evaluating accuracy, reasoning, and honesty of multimodal LLMs in astronomical classification tasks.