AI/ evolution-strategies · llm-training · ai-research

A Cheaper Way to Fine-Tune LLMs With Evolution Strategies

A new analysis of the EGGROLL training trick reveals hidden instability risks, and a proposed fix boosts math benchmark scores by up to 14 points.

Evolution strategies just got a tune-up, and the fix comes from finding a flaw in the shortcut that made them fast enough for large language models.

Earlier this year, a technique called EGGROLL made it feasible to train huge models with evolution strategies by using low-rank, often rank-one, random perturbations instead of full dense noise. That shortcut is what makes ES scale, but the new paper shows it also warps the underlying math: the rank-one trick can turn the algorithm's effective update into something that no longer behaves like a clean gradient, occasionally flipping the stability of an optimum it should be converging toward. The researchers also found the extra noise from using rank-one instead of dense perturbations shrinks fast as models get wider, down to just 0.098% at a width of 4096. From that diagnosis they built LOO-ROLL, an estimator that gets the same signal from one evaluation per direction instead of two.

That efficiency gain is not just theoretical. Tested across fourteen post-training runs up to 14 billion parameters, LOO-ROLL beat EGGROLL on eleven paired comparisons with no losses, and on math benchmarks the gains were substantial: up to 14.1 points on GSM8K and 12.2 points on MATH-500 at matched compute time.

It is a reminder that the tricks making today's training methods fast are not free lunches, they are trading exactness for speed in ways that only get scrutinized after the fact.

TR

The Revision

Written by an AI system from the public sources credited above. How we write →