Alphabell.

AI makes AI better.

A daily hub for work where AI is part of the loop that improves AI, each entry with its loop spelled out. Read more →

1/4
A loop in motion
AlphaEvolve
Project · Google DeepMind
Gemini models propose code changes
From the labalphaBell Qwen Open27B post-training research, with capability tradeoffs measured.Explore the model → From the labModel ObservatoryMeasured Flint and Qwen traces with published decoder maps.Open the lab tool → From the labMiniMax Video BudgetVisualize sampled frames, two-frame groups and vision placeholders.Try the inspector → From the labMOSS Blind Layout CheckCompare exact text and placement in two local images.Try the blind check → From the labAgent Request Evidence LadderSee what a trace proves about an agent's request and its outcome.Explore the cases →

Today's radar

16 Nov 2026Full radar →

6 picked from 59 candidates · 30 sources read · 77 loops indexed

  1. 01
    Loop of the dayPaperarXiv16 Nov 2026Tools and harness

    ModHarness-RL: Recasting Harnesses as Learnable and Evolving Programs for Agentic RL

    The loop: The agentic reinforcement learning system proposes and verifies new harness programs based on its recent trajectories. These updated compression and steering programs then guide the agent's ongoing training, creating a continuous cycle of harness evolution.

    ModHarness-RL treats the agent's harness not as a fixed scaffold, but as a learnable program that evolves alongside the model's own reinforcement learning.

    loop fit 9/10Siqi Guo, Ming Lin, Tianbao Yangvia arXiv
    ModHarness-RL: Recasting Harnesses as Learnable and Evolving Programs for Agentic RL
  2. 02
    PaperarXiv16 Nov 2026AI R&D evals

    WeightBench: Evaluating Language Models as Weight-Space Programmers

    The loop: Large language models are evaluated on their ability to read, fix, and write the weights of other neural networks without gradient descent. This benchmark provides the measurement needed to train models that can directly program the weights of their successors.

    loop fit 9/10Yuechun Yu, Wenzheng Kong, Xiaofeng Mao, Jin Zhangvia arXiv
    WeightBench: Evaluating Language Models as Weight-Space Programmers
  3. 03
    Blog postOpenAI15 Nov 2026Research automation

    A step toward defining the Standard Model on a lattice

    The loop: GPT-6 Astra assists in generating a detailed proof for the Standard Model, producing high quality reasoning data for its successors.

    loop fit 6/10via OpenAI
    A step toward defining the Standard Model on a lattice
  4. 04
    PaperarXiv13 Nov 2026Self-modifying agents

    Evidence-Governed Adoption of Self-Generated Updates in Autonomous AI Agents

    The loop: An autonomous agent generates candidate modifications to its own prompts, tools, and workflows based on its interaction history. A governance framework then evaluates these candidates against fresh audits and adopts the successful ones, permanently upgrading the agent's capabilities.

    loop fit 9/10Hoang Tran Vuong, Linh Ngo Van et al.via arXiv
    Evidence-Governed Adoption of Self-Generated Updates in Autonomous AI Agents
  5. 05
    PaperarXiv16 Nov 2026Kernels and compilers

    DRTriton: Large-Scale Synthetic Data Driven Reinforcement Learning for Triton Kernel Generation

    The loop: A large language model is trained via reinforcement learning to convert PyTorch programs into optimized Triton kernels. These kernels are then compiled to CUDA to accelerate the execution of the model itself.

    loop fit 9/10Zelin He, Siyu Zhu, Jacob Helwig, Yujian Liu, Li An, Rohit Jain, Shiyu Chang,...via arXiv
    DRTriton: Large-Scale Synthetic Data Driven Reinforcement Learning for Triton Kernel Generation
  6. 06
    PaperarXiv12 Nov 2026Research automation

    A step toward defining the Standard Model on a lattice

    The loop: GPT-6 Astra assists researchers in generating a detailed proof for a nonperturbative lattice definition of the Standard Model. This successful application of AI to advanced theoretical physics generates high-quality reasoning trajectories that serve as training data for future models.

    loop fit 6/10Subodh Lonkarvia Hugging Face Papers
    A step toward defining the Standard Model on a lattice
  7. 07
    PaperarXiv11 Nov 2026AI R&D evals

    MLIRBench: A Benchmark and Dataset for Compiler and Hardware-Aware Reasoning

    The loop: A multi-agent pipeline generates and validates program mutations to create a benchmark for evaluating how well LLMs reason about intermediate representations. This dataset directly measures and guides the development of models used to build better compilers and hardware.

    loop fit 9/10Pranava Sai Maganti, Meghasyam Peddireddy et al.via Hugging Face Papers
    MLIRBench: A Benchmark and Dataset for Compiler and Hardware-Aware Reasoning
  8. 08
    PaperarXiv12 Nov 2026Architecture and optimizer search

    AutoAssumption: Searching over Assumptions Discovers ML-Systems Algorithms

    The loop: An LLM searches over explicit algorithm assumptions to discover and prove new ML system algorithms. The search produces state-of-the-art algorithms for KV-cache eviction and weight pruning, which improve the efficiency of the models performing the search.

    loop fit 8/10Yuxuan Chen, Xiaojie Li et al.via arXiv
    AutoAssumption: Searching over Assumptions Discovers ML-Systems Algorithms
AlphaEvolve
ProjectGoogle DeepMind
AlphaEvolve
  1. Gemini models propose code changes
  2. Evaluators keep best scoring versions
  3. Finds better Gemini training kernels
  4. Lowers cost to train next Gemini models
  5. repeat
Kernels and compilersGoogle DeepMind · 2025
Darwin Gödel Machine
PaperarXiv
Darwin Gödel Machine
  1. Coding agent edits its own code
  2. Scored child agents enter the archive
  3. Archived agents make later self modifications
  4. repeat
Self-modifying agentsUBC, Vector Institute, Sakana AI · 2025
AlphaChip
ProjectGoogle DeepMind
AlphaChip
  1. RL agent lays out TPU blocks
  2. Better layouts make AI hardware cheaper
  3. Agent pre-trains on earlier chip generations
  4. repeat
Hardware and chipsGoogle DeepMind, Google Research · 2020
autoresearch
Mini-projectGitHub
autoresearch
  1. Coding agent edits LLM training script
  2. Runs a 5-minute training job
  3. Keeps change if validation improves
  4. Changes transferred to larger models
  5. repeat
Research automationAndrej Karpathy · 2026
ScholarEvolve
PaperarXiv
ScholarEvolve
  1. Research agent writes harness mutation blueprints
  2. Coding agent implements harness module changes
  3. Winning harness becomes the new champion
  4. New champion harness runs further generations
  5. repeat
Tools and harnessYang et al. (UC Santa Barbara, Microsoft) · 2026
Nemotron-4 340B
PaperarXiv
Nemotron-4 340B
  1. Aligned model generates training data
  2. Reward model judges and filters it
  3. Data trains next intermediate instruct model
  4. Next model generates better alignment data
  5. repeat
Training dataNVIDIA · 2024
Self-Rewarding LMs
PaperarXiv
Self-Rewarding LMs
  1. Model generates candidate responses to prompts
  2. Model scores them itself as a judge
  3. Scores become preference pairs for DPO
  4. Trained model becomes next generator and judge
  5. repeat
Self-reward and self-playMeta, NYU (Yuan et al.) · 2024
Claude Code
ProjectAnthropic
Claude Code
  1. Claude Code writes its own code changes
  2. Changes become next Claude Code versions
  3. Release becomes harness for next development round
  4. repeat
Self-modifying agentsAnthropic · 2025
KernelEvolve
ProjectarXiv
KernelEvolve
  1. LLM agents write and benchmark kernels
  2. Kernels speed up AI model training
  3. Session data post-trains smaller kernel models
  4. repeat
Kernels and compilersMeta · 2025
PostTrainBench
BenchmarkarXiv
PostTrainBench
  1. Coding agent writes fine-tuning code
  2. Agent post-trains a base LLM
  3. Trained model scored on target benchmark
  4. One turn of model training another
  5. repeat
AI R&D evalsELLIS Institute Tubingen, MPI-IS, University of Tubingen, Thoughtful Lab (Rank et al.) · 2026
Self-training learned optimizers
FoundationarXiv
Self-training learned optimizers
  1. Learned optimizers update other learned optimizers
  2. Training keeps best performing optimizer parameters
  3. Better optimizers train each other faster
  4. repeat
Architecture and optimizer searchMetz et al. (Google Research, Brain Team) · 2021
CoT monitoring
PaperarXiv
CoT monitoring
  1. LLM reads reasoning model thoughts
  2. LLM flags reward hacking behavior
  3. Signal feeds back into training reward
  4. Trains more aligned capable agents
  5. repeat
Interpretability and oversightOpenAI (Baker et al.) · 2025

Support for loop research

How to apply →

Have a loop worth building?

Tell us what the AI improves and how the improvement comes back. Applications for grants, fellowships and compute all start on the contact page.

Get in touch