AI makes AI better.

7 picked from 88 candidates · 31 sources read · 71 loops indexed

1. 01
   Loop of the dayPaperarXiv23 Oct 2026Architecture and optimizer search

   RSI-Opt: Recursive Self-Improvement Optimizer Agents For Model Training

   The loop: A meta-agent rewrites the policy and code library of an optimizer agent based on evidence from earlier training runs. The improved optimizer agent then manages subsequent model training tasks more effectively.

   RSI-Opt demonstrates how a meta-agent can accumulate experience across training runs to evolve better optimizer code and policies, directly improving the training of subsequent models.

   loop fit 10/10Binghang Lu, Yige Sun et al.via arXiv
   
2. 02
   PaperarXiv23 Oct 2026Research automation

   Minimum Description Length Unsupervisedly Separates Law-Grade from Statistical-Grade Knowledge in a Self-Improving Science Agent - and Reveals 86% of Its Own Discoveries Are Statistical

   The loop: A self-improving science agent uses a Minimum Description Length compression ratio to unsupervisedly vet its generated scientific relations. This filters the agent's discovery ledger to retain true laws, which guides its future research.

   loop fit 9/10Wenbin Mao, Pei Zouvia arXiv
   
3. 03
   Forum postLessWrong25 Oct 2026Interpretability and oversight

   Is GPT 2XL already clean?

   The loop: Opus 5.5 acts as an automated interpretability researcher to analyze and identify the function of neurons in GPT-2-XL. These insights into model internals provide a pathway for researchers to better align and develop future models.

   loop fit 7/10Jasper Blankvia LessWrong
   
4. 04
   PaperarXiv26 Oct 2026Inference efficiency

   NgramRL: Accelerating RL Post-Training for Structured Generation via Rollout-Aware N-gram Speculative Decoding

   The loop: A model undergoing reinforcement learning post-training uses its own recent rollout history to construct an n-gram drafter. This drafter accelerates the autoregressive generation bottleneck, speeding up the model's own post-training process.

   loop fit 9/10Nilesh Prasad Pandey, Jangseon Park, Jason Kong, Onat Gungor, Tajana Rosingvia arXiv
   
5. 05
   PaperarXiv26 Oct 2026Self-modifying agents

   AlphaMemo: Structured Search-Process Memory for Self-Evolving Alpha Mining Agents

   The loop: A self-evolving alpha mining agent records reusable evidence of successful and failed edit motifs in a structured memory. The agent then uses this memory to suppress failure patterns and improve its own future discovery efficiency.

   loop fit 9/10Hang Yu, Zifan Zheng, Chun Gan, Jeff Z. Pan, Tongliang Liu, Zhiyong Wang,...via arXiv
   
6. 06
   PaperarXivTools and harness

   HarnessEvolve: Learning from Reference Trajectories for Reliable Agent Self-Evolution

   The loop: An agent learns from reference trajectories to evolve its own execution harness. The improved harness provides a more reliable environment, making the agent better at subsequent self-evolution tasks.

   loop fit 9/10via Awesome-AI4AI
   
7. 07
   PaperarXivTools and harness

   HarnessCompass: Guiding Automatic Harness Evolution toward Generalizable and Effective Agent Harnesses

   The loop: An automated system evolves agent harnesses based on performance feedback. These improved harnesses are then deployed to provide better environments and evaluation frameworks for future agents.

   loop fit 9/10via Awesome-AI4AI
   
8. 08
   PaperarXiv23 Oct 2026Self-modifying agents

   Agentic Knowledge-Model Co-Evolution via Consistency Regularization

   The loop: The SPLICE framework analyzes agent experience to produce knowledge candidates, which improves the internal model through self-distillation. This makes the agent better at analyzing future experiences and solving tasks.

   loop fit 9/10Sirui Chen, Zhining Liu, Chao Zhang, Jingrui He, Zheng Livia arXiv