KI-Forschungspapiere Täglich

NaturalSpeech 3: Zero-Shot Speech Synthesis with Factorized Codec and Diffusion Models

Finetuned Multimodal Language Models Are High-Quality Image-Text Data Filters

MathScale: Scaling Instruction Tuning for Mathematical Reasoning

Wukong: Towards a Scaling Law for Large-Scale Recommendation

MAGID: An Automated Pipeline for Generating Synthetic Multi-modal Datasets

EasyQuant: An Efficient Data-free Quantization Algorithm for LLMs

Feast Your Eyes: Mixture-of-Resolution Adaptation for Multimodal Large Language Models

Modeling Collaborator: Enabling Subjective Vision Classification With Minimal Human Effort via LLM Tool-Use

RT-Sketch: Goal-Conditioned Imitation Learning from Hand-Drawn Sketches

MagicClay: Sculpting Meshes With Generative Neural Fields