FLEX Dataset: Multimodal Fitness Action Quality Assessment
Discover FLEX, a large-scale multimodal dataset for fitness action quality assessment with expert annotations and multiview recordings.
StructEval: Benchmarking LLMs for Structured Output Quality
Discover StructEval, the benchmark assessing LLMs' ability to generate accurate structured outputs like JSON, HTML, and more across 18 formats.
LMask: Efficient Solutions for Constrained Routing Problems
Discover LMask, a novel framework using lazy masking to solve complex constrained routing problems with high-quality, feasible solutions.
NeuS-E: Boost Text-to-Video Generation with Neuro-Symbolic AI
Discover NeuS-E, a zero-training pipeline using neuro-symbolic feedback to enhance text-to-video generation with improved semantic and temporal consistency...
Zero-Shot Concept Bottleneck Models for AI Interpretability
Discover zero-shot concept bottleneck models that enable interpretable AI without training, using dynamic concept retrieval and large-scale concept banks.