AI makes AI better.

8 picked from 24 candidates · 30 sources read · 74 loops indexed

1. 01
   Loop of the dayPaperarXiv29 Oct 2026AI R&D evals

   InfraDev: Can Coding Agents Develop AI Infrastructure, and How Can We Tell?

   The loop: Coding agents develop production-grade AI infrastructure like Megatron-LM and vLLM within the InfraDev benchmark. This improves the evaluation of infrastructure engineering, guiding the development of agents that can build and optimize their own training and inference systems.

   InfraDev introduces a rigorous benchmark for evaluating coding agents on real-world AI infrastructure tasks, a crucial stepping stone toward models that can autonomously optimize their own training and inference stacks.

   loop fit 10/10Zhonghao Jiang, Xuyang Zheng et al.via arXiv
   
2. 02
   PaperarXiv29 Oct 2026Self-modifying agents

   EvoIn: Bridging Evolution and Internalization for Agent Fine-Tuning

   The loop: An agent temporarily instantiates evolved decision-making procedures in its harness to generate improved reasoning traces. The agent then fine-tunes on these traces, internalizing the logic to improve its own future decision-making without relying on the external harness.

   loop fit 9/10Yunhang Shen, Zhangquan Chen et al.via Hugging Face Papers
   
3. 03
   ProjectGitHub26 Oct 2026Self-modifying agents

   InfRec-Cardinal/The-Trinity

   The loop: Multimodal agents use capability-driven recursive self-improvement to enhance their own performance and capabilities.

   loop fit 7/10InfRec-Cardinalvia GitHub
   
4. 04
   PaperarXiv29 Oct 2026Tools and harness

   HEMO: Harness Self-Evolution for Multi-Objective Combinatorial Optimization

   The loop: A large language model treats its own automatic operator design harness as a self-evolving target. By updating its retrieval, selection, and orchestration logic based on downstream performance, the system improves its ability to discover optimization operators.

   loop fit 9/10Jianghan Zhu, Cong Zhang et al.via arXiv
   
5. 05
   PaperarXiv28 Oct 2026Self-reward and self-play

   Youtu-Parsing-Omni: One Encoder, One Schema, All Modalities

   The loop: The omni-modal parser generates its own rollouts against a reference-conditioned copy of itself to create token-level targets. The model is then trained on these rollouts and refreshes the reference copy with its updated weights, driving continuous self-improvement.

   loop fit 9/10Yecheng Wu, Song Han et al.via Hugging Face Papers
   
6. 06
   PaperarXiv28 Oct 2026Self-reward and self-play

   Weaken to Strengthen: Weakening-Induced On-Policy Self-Distillation for Large Reasoning Models

   The loop: A reasoning model generates a weaker counterpart of itself through quantization or pruning to create a contrastive training signal. This enables on-policy self-distillation without privileged information, improving the original model's accuracy on reasoning tasks.

   loop fit 9/10Miri Liu, Xiayu Zhao, Encheng Xie, ChengXiang Zhaivia arXiv
   
7. 07
   PaperarXiv28 Oct 2026AI R&D evals

   FeatureArc: Benchmarking Coding Agents on Evolving Features

   The loop: Coding agents extend software features over successive stages and build on their own earlier code within the FeatureArc benchmark. This improves the evaluation of agentic self-maintenance, guiding the development of agents capable of long-term iterative software engineering.

   loop fit 9/10Yunlong Wang, Alyssa Wu, Cuiyun Gao, Jia Feng, Songyuanyi Lu, Zhichao Huvia arXiv
   
8. 08
   PaperarXiv28 Oct 2026Self-modifying agents

   InfRec-Cardinal/The-Trinity

   The loop: Multimodal agents undertake capability-driven recursive self-improvement within The Trinity framework. By iteratively enhancing their own performance, the agents systematically expand their multimodal capabilities.

   loop fit 7/10Hehai Lin, Qi Yang et al.via arXiv