• Voicebox: Text-Guided Multilingual Universal Speech Generation at Scale
• Bigger, Better, Faster: Human-level Atari with human-level efficiency
• Seeing the World through Your Eyes
• Rerender A Video: Zero-Shot Text-Guided Video-to-Video Translation

Support the show

• Progressively Optimized Local Radiance Fields for Robust View Synthesis
• DISTAL: The Distributed Tensor Algebra Compiler
• Efficient 3D Semantic Segmentation with Superpoint Transformer
• Tune As You Scale: Hyperparameter Optimization For Compute Efficient Training
• Efficient Adapters for Giant Speech Models
• AssistGPT: A General Multi-modal Assistant that can Plan, Execute, Inspect, and Learn

Support the show

• Gorilla: Large Language Model Connected with Massive APIs
• Attention is Turing Complete
• Artificial Artificial Artificial Intelligence: Crowd Workers Widely Use Large Language Models for Text Production Tasks
• Faster sorting algorithms discovered using deep reinforcement learning

Support the show

• Galactic: Scaling End-to-End Reinforcement Learning for Rearrangement at 100k Steps-Per-Second
• One-for-All: Generalized LoRA for Parameter-Efficient Fine-tuning
• Image Captioners Are Scalable Vision Learners Too
• StyleTTS 2: Towards Human-Level Text-to-Speech through Style Diffusion and Adversarial Training with Large Speech Language Models
• Orca: Progressive Learning from Complex Explanation Traces of GPT-4
• The Curse of Recursion: Training on Generated Data Makes Models Forget

Support the show

• Benchmarking Neural Network Training Algorithms
• High-Fidelity Audio Compression with Improved RVQGAN
• Augmenting Language Models with Long-Term Memory
• FasterViT: Fast Vision Transformers with Hierarchical Attention
• Large Language Models as Tax Attorneys: A Case Study in Legal Capabilities Emergence

Support the show