I am not affiliated with the authors, mainly posting this to get some technical commentary on it. Full arXiv paper here.

Large language models (LLMs) are powerful but static; they lack mechanisms to adapt their weights in response to new tasks, knowledge, or examples. We introduce Self-Adapting LLMs (SEAL), a framework that enables LLMs to self-adapt by generating their own finetuning data and update directives. Given a new input, the model produces a self-edit—a generation that may restructure the information in different ways, specify optimization hyperparameters, or invoke tools for data augmentation and gradient-based updates. Through supervised finetuning (SFT), these self-edits result in persistent weight updates, enabling lasting adaptation. To train the model to produce effective self-edits, we use a reinforcement learning loop, using the downstream performance of the updated model as the reward signal. Unlike prior approaches that rely on separate adaptation modules or auxiliary networks, SEAL directly uses the model’s generation to parameterize and control its own adaptation process. Experiments on knowledge incorporation and fewshot generalization show that SEAL is a promising step toward language models capable of self-directed adaptation in response to new data. Our website and code is available at https://jyopari.github.io/posts/seal.

They use Llama 3.2 1B - Instruct and claim massive improvements on self-edit policy for selected ARC 1 tasks (20% to 72% jump, with a 100% upper bound for hand-crafted self-edit solutions). I see the paper as a clear demonstration of a LLM selecting its own weight updates to better answer questions and I feel the implications are massive (if it can be scaled up in the big labs). I don't have the technical experience however for a full dive.