Live
•🔥 OpenBMB/ChatDevGitHub Trending•🔥 microsoft/agent-lightningGitHub Trending•🔥 apache/supersetGitHub Trending•🔥 shanraisshan/claude-code-best-practiceGitHub Trending•A-SelecT: Automatic Timestep Selection for Diffusion Transformer Representation LearningarXiv•GUIDE: Resolving Domain Bias in GUI Agents through Real-Time Web Video Retrieval and Plug-and-Play AnnotationarXiv•Sommelier: Scalable Open Multi-turn Audio Pre-processing for Full-duplex Speech Language ModelsarXiv•CANGuard: A Spatio-Temporal CNN-GRU-Attention Hybrid Architecture for Intrusion Detection in In-Vehicle CAN NetworksarXiv•DesignWeaver: Dimensional Scaffolding for Text-to-Image Product DesignarXiv•A Lightweight, Transferable, and Self-Adaptive Framework for Intelligent DC Arc-Fault Detection in Photovoltaic SystemsarXiv•Consistency Amplifies: How Behavioral Variance Shapes Agent AccuracyarXiv•Stabilizing Rubric Integration Training via Decoupled Advantage NormalizationarXiv•Semi-Automated Knowledge Engineering and Process Mapping for Total Airport ManagementarXiv•AIRA_2: Overcoming Bottlenecks in AI Research AgentsarXiv•BeSafe-Bench: Unveiling Behavioral Safety Risks of Situated Agents in Functional EnvironmentsarXiv•🔥 OpenBMB/ChatDevGitHub Trending•🔥 microsoft/agent-lightningGitHub Trending•🔥 apache/supersetGitHub Trending•🔥 shanraisshan/claude-code-best-practiceGitHub Trending•A-SelecT: Automatic Timestep Selection for Diffusion Transformer Representation LearningarXiv•GUIDE: Resolving Domain Bias in GUI Agents through Real-Time Web Video Retrieval and Plug-and-Play AnnotationarXiv•Sommelier: Scalable Open Multi-turn Audio Pre-processing for Full-duplex Speech Language ModelsarXiv•CANGuard: A Spatio-Temporal CNN-GRU-Attention Hybrid Architecture for Intrusion Detection in In-Vehicle CAN NetworksarXiv•DesignWeaver: Dimensional Scaffolding for Text-to-Image Product DesignarXiv•A Lightweight, Transferable, and Self-Adaptive Framework for Intelligent DC Arc-Fault Detection in Photovoltaic SystemsarXiv•Consistency Amplifies: How Behavioral Variance Shapes Agent AccuracyarXiv•Stabilizing Rubric Integration Training via Decoupled Advantage NormalizationarXiv•Semi-Automated Knowledge Engineering and Process Mapping for Total Airport ManagementarXiv•AIRA_2: Overcoming Bottlenecks in AI Research AgentsarXiv•BeSafe-Bench: Unveiling Behavioral Safety Risks of Situated Agents in Functional EnvironmentsarXiv
TechTonicShifts robot
AI NEWS
by techtonicshifts.blog
Self-Evolving AI

Self-Evolving AI

Autonomous AI systems, self-improvement, and meta-learning

2 articles

Tags in this category
AlphaCode 3 Improves Itself: Self-Play Training Achieves Competitive Programming Gold
Self-Evolving AIHot

AlphaCode 3 Improves Itself: Self-Play Training Achieves Competitive Programming Gold

DeepMind's AlphaCode 3 uses self-play and automated test generation to continuously improve its coding capabilities, reaching gold medal performance on Codeforces without human-labeled training data.

Google DeepMind
6m14.6k6 days ago
Meta's Self-Taught Reasoner (STaR) Scales to Complex Scientific Problems
Self-Evolving AI

Meta's Self-Taught Reasoner (STaR) Scales to Complex Scientific Problems

Meta Research extends the Self-Taught Reasoner framework to complex scientific domains, showing that models can bootstrap their own reasoning capabilities through iterative self-improvement on generated problems.

ArXiv
7m7.8k9 days ago