Open research commons for machine learning

Closing the Sim-to-Real Gap in Robotics Experience Replay for Continual Learning Low Rank Tensor Network Decompositions Learning Adaptive Graph Topologies

38,150Research Directions

Featured Paper

Fully adaptive density-based clustering

The Annals of Statistics 2015 37 citations

Previously: Contrast adaptive binarization of low quality docu... · Semi-supervised deep learning by metric embedding · Path Planning for Multiple Heterogeneous Unmanned ...

Today in ML History — September 7

2020

An End-to-end Architecture of Online Multi-channel Speech Separation

Ed Lin, Jian Wu, Jian Wu 0027, Jinyu Li, Jinyu Li 0001, Lei Xie, Lei Xie 0001, Takuya Yoshioka, Yi Luo, Yi Luo 0004

The paper proposes an end-to-end neural architecture for real-time, multi-channel speech separation that processes streaming audio signals with minimal latency.
2021

Besov Function Approximation and Binary Classification on Low-Dimensional Manifolds Using Convolutional Residual Networks

Hao Liu, Hao Liu 0028, Minshuo Chen, Tuo Zhao, Wenjing Liao

This study demonstrates that convolutional residual networks can effectively approximate Besov functions and perform binary classification on data lying on low-dimensional manifolds.
2023

Senate hearing on AI and U.S. competitiveness

Joe Manchin, John Barrasso, David M. Turk +1

The U.S. Senate Committee on Energy and Natural Resources held a full committee hearing to examine recent advances in artificial intelligence and the Department of Energy's role in ensuring U.S. security.
2023

Google DeepMind publishes study on AI prompting techniques

Google DeepMind researchers published a study demonstrating that adding the phrase 'Take a deep breath and work on this problem step by step' significantly improved the accuracy of the PaLM 2 AI model on math problems.
2025

OpenAI publishes research on AI hallucinations

OpenAI published a research paper analyzing the causes of hallucinations in large language models, attributing them to pretraining incentives that prioritize word prediction over truthfulness.

Featured Directions

Dynamic and Procedural Evaluation of LLM Reasoning1,735 papers
Prompt Sensitivity and Robustness in Large Language Models1,176 papers
Automated Discovery of Task Specific Neural Architectures700 papers
Automated Discovery of State and Temporal Abstractions422 papers
Automated Peer Review for High-Volume Scientific Literature275 papers
Synchronizing Peer Review with Rapid Innovation Cycles138 papers

Author Spotlight

Akshit Achara

King's College London

1 papers 1 directions

Multi-Way Representation Alignment2026