Frontier LLMs are effective batch optimizers: Assessing reasoning models in continuous and discrete settings

A new arXiv preprint (2609.03177v1) finds that frontier large language models (LLMs) are competitive zero-shot batch optimizers for numerical test functions but their performance is brittle compared to classical non-LLM optimization approaches. However, the study shows LLMs are significantly better in semantically rich settings, indicating their batch optimization behavior is highly effective when navigating discrete spaces similar to their pretraining data.

arXiv:2609.03177v1 Announce Type: new Abstract: Frontier large language models LLMs have become attractive priors for optimization due to their large-scale pretraining that enables them to navigate a variety of optimization settings. However, the effectiveness of modern reasoning LLMs in batch optimization settings remains underexplored. Here we investigate the performance of the current generation of frontier LLMs as batch optimizers in both continuous and discrete settings. We find that while LLMs are competitive zero-shot batch optimizers for numerical test functions, their performance is brittle compared to classical non-LLM optimization approaches. However, LLM priors are significantly better in semantically rich settings, indicating that their batch optimization behavior is highly effective when navigating and reasoning over the discrete spaces most similar in structure to their pretraining data.