In plain words: The model writes its own practice text and update instructions, then fine-tunes its weights on them so new knowledge sticks. Trained by rewarding edits that improve the updated model, it did better on knowledge and few-shot tasks than setups using a separate helper network.
Abstract
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 with 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 own generation to control its adaptation process. Experiments on knowledge incorporation and few-shot generalization show that SEAL is a promising step toward language models capable of self-directed adaptation. Our website and code is available at https://jyopari.github.io/posts/seal.
Adam Zweiger, Jyothish Pari, Han Guo, Ekin Akyürek, Yoon Kim, Pulkit Agrawal
arXiv:2506.10943 · cs.LG · submitted Jun 12, 2025 · updated Sep 18, 2025
abstract · pdf · html
Two things that stand out:
- The knowledge incorporation results (47% vs 46.3% with GPT-4.1 data, both much higher than the small-model baseline) show the model does discover better training formats, not just more data. Though the catastrophic forgetting problem remains unsolved, and it's not completely clear whether data diversity is improved.
- The computational overhead is brutal - 30-45 seconds per reward evaluation makes this impractical for most use cases. But for high-value document processing where you really need optimal retention, it could be worth it.
The restriction to tasks with explicit evaluation metrics is the main limitation. You need ground truth Q&A pairs or test cases to compute rewards. Still, for domains like technical documentation or educational content where you can generate evaluations, this could significantly improve how we process new information.
Feels like an important step toward models that can adapt their own learning strategies, even if we're not quite at the "continuously self-improving agent" stage yet.