A reactive optimizer that uses a pretrained language-model agent to inspect training dynamics and rewrite executable optimizer code during a single training run is studied, evaluating this approach on language modeling and image generation when a recipe has not been tuned for the target setting.
Abstract
We study a reactive optimizer that uses a pretrained language-model agent to inspect training dynamics and rewrite executable optimizer code during a single training run. The agent can change update rules and schedules for individual parameter groups, without optimizer-specific meta-training or memory transferred from other runs. We evaluate this approach, OPAL, on language modeling and image generation when a recipe has not been tuned for the target setting. Across the evaluated settings, it matches or improves on selected fixed recipes. Analysis of its decisions shows changes to learning-rate schedules, optimizer structure, numerical precision, and momentum. Controlled experiments also show adaptation to repeated data corruption and reduced overfitting. We report these behaviors together with the agent's computational cost and implementation requirements.
RSI-Opt is introduced, a plug-and-play Recursive Self-Improvement Optimizer Agents which evolves the knowledge that guides this process: a prompt, a playbook of procedures, a library of tested optimizer code, and a decision cadence.
Language model agents increasingly automate post-training, but when given the full training stack they converge on supervised fine-tuning and spend most effort revising training data. Existing systems give the agent one kind of control, selecting rows from a fixed pool or calling a generator, whereas a human data engin...
Bing-Xiang He, Peng Sun, Shi-Hao Peng et al.· 0 citations
Results characterize when runtime-controlled composition of coding-agent products improves execution accuracy, as well as when runtime-controlled composition of coding-agent products improves execution accuracy.
Zheng-Yu Chen, Lin-Feng Liu, Hong Li et al.· 0 citations
Agent evolvers automate the design of the prompts, skills and workflows around language model agents, yet the optimization process they follow is still designed by hand: a fixed search loop decides how candidates are evaluated, which are kept and when the search stops. We propose FREEEVOLVE, which automates this proces...
Lecheng Kong, Li-Ke Hui, Nikos Kanakaris et al.· 0 citations
Repository-scale code optimization demands broad diagnosis, but unconstrained agent edits impede review and attribution. We present MACO-AI, a reusable multi-agent optimization skill that restricts source changes to annotations and intrinsics. Before delegation, initialization freezes the source state and measurement c...
Jun-Yu Liu, Zong-Yan Xian· Proceedings of the 41st IEEE...· 0 citations
Training a large language model can occupy thousands of GPUs for weeks, and its speed depends on its recipe: how the model is split across GPUs, how memory is traded for compute, and which kernels and communication overlaps it uses. Finding a good recipe is still manual work for experts. Automated tuners search fixed l...
Leo Fan, Hong-Yang Wang, Wen-Bo Chen et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
