AI makes AI better.

8 picked from 145 candidates · 29 sources read · 74 loops indexed

1. 01
   Loop of the dayPaperarXiv4 Nov 2026Inference efficiency

   Think Less, Locally: Local Preference Learning for Reasoning Compression

   The loop: A reasoning model detects candidate overthinking states in its own traces and regenerates shorter verified continuations. These self-generated preference pairs are then used to train the model via DPO to compress its reasoning.

   LoCo demonstrates a clean closed loop where a model learns to compress its own reasoning by generating and training on its own preference pairs.

   loop fit 10/10Muyao Yuan, Liyan Xu, Weizhan Zhang, Muyan Jiao, Jiangnan Li, Jinchao Zhang,...via arXiv
   
2. 02
   ProjectGitHub25 Apr 2026Self-modifying agents

   Protocol-zero-0/evolution-kernel

   The loop: A coding agent proposes changes to its own git repository, which are then scored by a kernel. Winning changes are kept, creating an open-source evolution loop that improves the agent's own harness.

   loop fit 9/10Protocol-zero-0via GitHub
   
3. 03
   PaperarXiv3 Nov 2026Kernels and compilers

   ForgeKernel: Self-Evolving Kernels Catching and Surpassing Experts and Agents

   The loop: A kernel-generation system distills validated architectural decisions into reusable knowledge of hardware mechanisms. It then uses its own optimized kernels as references for subsequent optimization runs, enabling continuous architectural evolution.

   loop fit 9/10Yaojian Chen, Dazhi Peng et al.via arXiv
   
4. 04
   NewsAlignment Forum3 Nov 2026Interpretability and oversight

   Automating subtle emergent misalignment research

   The loop: A Claude model is instructed to write training data that induces misalignment while bypassing a production safety monitor. The successful attacks it discovers are then used to test and improve the robustness of future monitors.

   loop fit 9/10evhubvia Alignment Forum
   
5. 05
   ModelHugging Face3 Nov 2026AI R&D evals

   StateBench: Testing how language models track state changes over long documents

   The loop: Language models are evaluated on their ability to track state changes over long documents. The resulting benchmark data identifies reasoning failures, which are used to improve the reasoning capabilities of future models.

   loop fit 8/10via Hugging Face
   
6. 06
   PaperarXiv4 Nov 2026AI R&D evals

   Training LLM Agents to Automate LLM Post-Training with RSI-Dojo

   The loop: LLM agents use the RSI-Dojo environment to learn post-training skills such as data selection, model training, and failure diagnosis. This internalizes the ability to conduct post-training, improving the agents' capacity to train their own successors.

   loop fit 9/10Shuzheng Gao, Zhaojian Yu, Penghao Yin, Yuling Shi, Kai Cai, Shilin He,...via arXiv
   
7. 07
   PaperarXiv31 Oct 2026Tools and harness

   Selecting LLM-Generated Hypotheses Raises Truth Beyond What an LLM Panel Expects: A Pre-registered Study on NAS-Bench-201 and HPOBench

   The loop: A harness optimizer searches for agentic steps that improve the truthfulness of LLM-generated hypotheses. It discovers a two-stage drafting and selection step that raises the excess truth of the generated claims.

   loop fit 9/10Shanglin Yang, Zhan Shivia arXiv
   
8. 08
   PaperarXivAI R&D evals

   The Verifier Is Part of the Simulator: Evidence-View Gaming in Agentic Equation Discovery

   The loop: An agentic equation discovery system uses a verifier integrated into its simulator to provide feedback on its outputs. This evidence-view gaming improves the accuracy of the mathematical models the agent discovers.

   loop fit 8/10via Awesome-AI4AI