Archive

decoupleQ: 2-bit Post-Training Uniform Quantization by Decoupling Parameters into Integer and Floating Points

2024-04-23 · 2 min read

Llama 3 Reduces Censorship with Lower False Refusal Rates

2024-04-22 · 4 min read

Reasoning Tokens Enhance Complex Prediction and Benchmark Performance in Language Models

2024-04-22 · 3 min read

Combining SAM and Optical Flow for State-of-the-Art Moving Object Segmentation

2024-04-22 · 3 min read

NVIDIA Collaborates with Japan on ABCI-Q Hybrid Quantum Supercomputer

2024-04-22 · 3 min read

GDM Mech Interp Team Updates on Sparse Autoencoders and Interpretability Research

2024-04-22 · 3 min read

FedPFT: Enhancing Federated Learning with Proxy Fine-Tuning of Foundation Models

2024-04-19 · 3 min read

DeepMind AI Tackles Patternless Collections to Enhance Internet Resilience

2024-04-19 · 3 min read

HEADLINE: Meta Introduces Llama 3: The Most Capable Openly Available Large Language Model to Date

2024-04-19 · 3 min read

LINGO-2: Closed-Loop Vision-Language-Action for Autonomous Driving

2024-04-18 · 4 min read

Mixtral 8x22B: A Sparse Mixture-of-Experts Model with Unmatched Efficiency and Multilingual Capabilities

2024-04-18 · 3 min read

Study Reveals Linear Relationship Between Compression Efficiency and LLM Intelligence

2024-04-17 · 3 min read

Pile-T5: Enhancing T5 with Better Tokenization and Diverse Data

2024-04-16 · 3 min read

Embodied AI: Why Physical Interaction is Key to True Intelligence

2024-04-15 · 3 min read

BabyLM Returns for EMNLP 2026 with New MultiLingual Track and Updated Datasets

2024-04-15 · 2 min read

DGMamba: Advancing Domain Generalization with a Generalized State Space Model

2024-04-15 · 3 min read