Learned Optimizer Generalization Across Tasks

via “zero-shot task generalization across domains”

Olmo 3.1 32B Instruct is a large-scale, 32-billion-parameter instruction-tuned language model engineered for high-performance conversational AI, multi-turn dialogue, and practical instruction following. As part of the Olmo 3.1 family, this...

Unique: Instruction-tuning approach enables zero-shot task transfer by training on diverse task families with explicit instruction signals, rather than relying solely on pretraining patterns — this explicit task-instruction pairing during training improves generalization to novel task phrasings compared to base models

vs others: Outperforms base language models on zero-shot task diversity due to instruction-tuning, while maintaining faster inference than larger 70B+ models that may have marginal performance gains on specialized domains

via “learned-optimizer-generalization-across-tasks”

* ⭐ 07/2023: [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control (RT-2)](https://arxiv.org/abs/2307.15818)

Unique: Systematically evaluates optimizer generalization across diverse task distributions rather than reporting single-benchmark performance, using multi-domain evaluation to expose overfitting and identify robust algorithmic patterns.

vs others: Provides empirical generalization evidence that discovered optimizers work beyond their discovery tasks, unlike single-benchmark optimizer papers which may report inflated performance on cherry-picked problems.