BestHub
Sign in
Tagged articles

self-evolving AI

12 articles · Page 1 of 1
Data Party THU
Data Party THU
Nov 2, 2026 · Artificial Intelligence

Astar: Self-Evolving Industrial AI That Learns from Its Own History to Beat Experts

Alibaba and Zhejiang University's Astar model enables industrial AI systems to autonomously evolve by learning from their own Git commit history and experiment results, achieving 54% single-generation success rate with a 0.6B model—surpassing GPT-5.5 and human experts—and delivering 23.6% offline HitRate and 4.86% GMV gains in Lazada's ad system.

AlibabaAstarGit history mining
0 likes · 15 min read
Astar: Self-Evolving Industrial AI That Learns from Its Own History to Beat Experts
Machine Heart
Machine Heart
Oct 31, 2026 · Artificial Intelligence

EvoSim: AI That Evolves Physics Models from Experimental Discrepancies

Tsinghua-led researchers introduce EvoSim, a self-evolving AI system that revises physics equations from experimental mismatches rather than just tuning parameters, achieving 1.79% SOC error in lithium-plating onset prediction and 7.62 mV RMSE in dynamic voltage forecasting across 109k samples, while cutting model revisions 10× and runtime 42%.

EvoSimbattery modelingdynamic voltage prediction
0 likes · 11 min read
EvoSim: AI That Evolves Physics Models from Experimental Discrepancies
Data Party THU
Data Party THU
Oct 11, 2026 · Artificial Intelligence

Evolutionary Computation for Trustworthy AI: From Attack-Defense Search to Self-Evolving Systems

This review examines how evolutionary computation supports trustworthy AI across three research lines: evolutionary attacks to find failures, evolutionary defenses to search protective measures, and trustworthy self-evolution to govern which changes persist long-term, highlighting the shift from static robustness to continuous governance of memory, skills, and safeguards.

AI safety benchmarksMulti-Objective OptimizationTrustworthy AI
0 likes · 31 min read
Evolutionary Computation for Trustworthy AI: From Attack-Defense Search to Self-Evolving Systems
Machine Learning Algorithms & Natural Language Processing
Machine Learning Algorithms & Natural Language Processing
Sep 16, 2026 · Artificial Intelligence

AI Solves Tests But Can't Self-Evolve: ByteDance Seed's Three RSI Benchmarks

ByteDance Seed and TokenWave introduce ASPIRE, S³Gym, and HarnessDev — three benchmarks that test whether AI agents can autonomously select learning goals, distill experience into improved decisions, and persistently upgrade their own execution systems without human-provided verification.

AI AgentsASPIREByteDance Seed
0 likes · 14 min read
AI Solves Tests But Can't Self-Evolve: ByteDance Seed's Three RSI Benchmarks
Machine Learning Algorithms & Natural Language Processing
Machine Learning Algorithms & Natural Language Processing
Aug 17, 2026 · Artificial Intelligence

Can AI Really Self‑Evolve? MLS‑Bench Reveals Limits of Kimi K3 and Qwen3.8‑Max

The MLS‑Bench benchmark evaluates 140 real research tasks across 12 domains, showing that while models like Kimi K3 and Qwen3.8‑Max can boost scores through multi‑round optimization, they rarely discover genuinely new methods or demonstrate reliable experimental planning under flexible compute budgets.

AI ResearchLarge Language ModelsMLS‑Bench
0 likes · 18 min read
Can AI Really Self‑Evolve? MLS‑Bench Reveals Limits of Kimi K3 and Qwen3.8‑Max
Machine Learning Algorithms & Natural Language Processing
Machine Learning Algorithms & Natural Language Processing
Aug 8, 2026 · Artificial Intelligence

What Is Self‑Evolving, Self‑Improving, and Recursive Self‑Improvement? A Comprehensive Guide

This article surveys recent AI research on self‑evolving and self‑improving systems, defines a three‑layer taxonomy (Artifacts, Harness, Model), reviews concrete implementations from OpenAI, Anthropic, Tencent, MiniMax, and others, and outlines open research directions and challenges.

AI agent harnessAI benchmarkingAutonomous AI Research
0 likes · 35 min read
What Is Self‑Evolving, Self‑Improving, and Recursive Self‑Improvement? A Comprehensive Guide
DataFunSummit
DataFunSummit
Jul 22, 2026 · Artificial Intelligence

Why Knowledge Bases Alone Can’t Empower AI Agents: The Need for Actionable Experience

Large language models may know a great deal, yet they still stumble on concrete tasks because knowledge must be transformed into actionable, context‑aware skills; this article analyses how skill representation, model‑specific cognition, and continuous practice reshape knowledge engineering for self‑evolving AI agents.

AI AgentsExperience LearningLarge Language Models
0 likes · 16 min read
Why Knowledge Bases Alone Can’t Empower AI Agents: The Need for Actionable Experience
DataFunTalk
DataFunTalk
Jul 20, 2026 · Artificial Intelligence

Why Knowledge Bases Alone Won’t Make AI Agents Effective: The Need for Actionable Experience

Large language models may know a lot, but they still fail at concrete tasks because knowledge must be transformed into actionable, experience‑based skills; the article analyzes how self‑evolving AI agents require skill libraries, context‑aware representations, and continuous practice‑driven knowledge production rather than static documentation.

AI agentActionable KnowledgeAutonomous Systems
0 likes · 16 min read
Why Knowledge Bases Alone Won’t Make AI Agents Effective: The Need for Actionable Experience
Smart Sea Tide
Smart Sea Tide
Apr 28, 2026 · Artificial Intelligence

How Autoresearch Achieves 5‑Minute Self‑Evolving AI Research Loops

Autoresearch, an open‑source 630‑line framework by Andrej Karpathy, lets an AI autonomously run 5‑minute training experiments, modify code, and iteratively improve a 0.8 B model by 19 % in eight hours, using the val_bpb metric and a minimal three‑file architecture.

AI automationAutoResearchKarpathy
0 likes · 5 min read
How Autoresearch Achieves 5‑Minute Self‑Evolving AI Research Loops
AI Large-Model Wave and Transformation Guide
AI Large-Model Wave and Transformation Guide
Apr 16, 2026 · Artificial Intelligence

How MiniMax M2.7 Is Pioneering Self‑Evolving AI Models

MiniMax’s open‑source M2.7 model, released in April 2026, demonstrates the first self‑evolving AI agent that autonomously updates its memory, learns new skills, and optimizes its own training loop, achieving up to 30% performance gains and leading benchmark scores across programming, ML automation, and productivity tasks.

Agentic AIbenchmarkcost efficiency
0 likes · 9 min read
How MiniMax M2.7 Is Pioneering Self‑Evolving AI Models
Su San Talks Tech
Su San Talks Tech
Apr 11, 2026 · Artificial Intelligence

Why Hermes Agent Is the Next‑Gen Self‑Evolving AI Assistant

Hermes Agent, an open‑source AI framework from Nous Research, combines a self‑evolving skill loop, a five‑layer memory system, and a universal message‑gateway to deliver a continuously improving personal assistant that works across Linux, macOS, and major IM platforms, with simple CLI installation and extensive customization.

CLI toolHermes Agentlayered memory
0 likes · 12 min read
Why Hermes Agent Is the Next‑Gen Self‑Evolving AI Assistant
AI Engineering
AI Engineering
Jan 19, 2026 · Artificial Intelligence

How We Built a Self‑Evolving AI System Without Reward Functions

The Oxford study demonstrates that large language models can self‑evolve through a four‑step deploy‑validate‑filter‑inherit loop, eliminating handcrafted reward functions, and achieves dramatic performance gains on Blocksworld, Rovers, and Sokoban while providing theoretical proof of equivalence to REINFORCE.

AI safetyLLM planningQwen3
0 likes · 8 min read
How We Built a Self‑Evolving AI System Without Reward Functions