• Reinforced Self-Training (ReST) for Language Modeling
• Large Language Models as General Pattern Machines
• Anaphoric Structure Emerges Between Neural Networks
• Consciousness in Artificial Intelligence: Insights from the Science of Consciousness

Support the show

• Dynamic 3D Gaussians: Tracking by Persistent Dynamic View Synthesis
• Self Expanding Neural Networks
• Fast as CHITA: Neural Network Pruning with Combinatorial Optimization

Support the show

• Estimating the Carbon Footprint of BLOOM, a 176B Parameter Language Model
• Teach LLMs to Personalize -- An Approach inspired by Writing Education
• DragNUWA: Fine-grained Control in Video Generation by Integrating Text, Image, and Trajectory
• Dual-Stream Diffusion Net for Text-to-Video Generation

Support the show

• Bayesian Flow Networks
• Solving Challenging Math Word Problems Using GPT-4 Code Interpreter with Code-based Self-Verification
• RAVEN: In-Context Learning with Retrieval Augmented Encoder-Decoder Language Models

Support the show

• SpeechX: Neural Codec Language Model as a Versatile Speech Transformer
• Platypus: Quick, Cheap, and Powerful Refinement of LLMs
• RestoreFormer++: Towards Real-World Blind Face Restoration from Undegraded Key-Value Pairs
• OctoPack: Instruction Tuning Code Large Language Models

Support the show

• PIPPA: A Partially Synthetic Conversational Dataset
• Composable Function-preserving Expansions for Transformer Architectures
• Self-Alignment with Instruction Backtranslation

Support the show

• Follow Anything: Open-set detection, tracking, and following in real-time
• AudioLDM 2: Learning Holistic Audio Generation with Self-supervised Pretraining
• Alexa, play with robot: Introducing the First Alexa Prize SimBot Challenge on Embodied AI

Support the show

• MetaGPT: Meta Programming for Multi-Agent Collaborative Framework
• FocalFormer3D : Focusing on Hard Instance for 3D Object Detection
• Shepherd: A Critic for Language Model Generation
• JEN-1: Text-Guided Universal Music Generation with Omnidirectional Diffusion Models

Support the show

• Separate Anything You Describe
• Pre-Trained Large Language Models for Industrial Control
• ReCLIP: Refine Contrastive Language Image Pre-Training with Source Free Domain Adaptation
• Simple synthetic data reduces sycophancy in large language models
• 3D Gaussian Splatting for Real-Time Radiance Field Rendering

Support the show

• A Practical Deep Learning-Based Acoustic Side Channel Attack on Keyboards
• AlphaStar Unplugged: Large-Scale Offline Reinforcement Learning
• From Discrete Tokens to High-Fidelity Audio Using Multi-Band Diffusion
• FLIQS: One-Shot Mixed-Precision Floating-Point and Integer Quantization Search
• SynJax: Structured Probability Distributions for JAX

Support the show