Papers › Pretrained Transformers as Universal Computation Engines

Pretrained Transformers as Universal Computation Engines

9 Mar 2021arXiv:2103.05247archive 2025-07-28

Kevin Lu, Aditya Grover, Pieter Abbeel, Igor Mordatch

We investigate the capability of a transformer pretrained on natural language to generalize to other modalities with minimal finetuning -- in particular, without finetuning of the self-attention and feedforward layers of the residual blocks. We consider such a model, which we call a Frozen Pretrained Transformer (FPT), and study finetuning it on a variety of sequence classification tasks spanning numerical computation, vision, and protein fold prediction. In contrast to prior works which investigate finetuning on the same modality as the pretraining dataset, we show that pretraining on natural language can improve performance and compute efficiency on non-language downstream tasks. Additionally, we perform an analysis of the architecture, comparing the performance of a random initialized transformer to a random LSTM. Combining the two insights, we find language-pretrained transformers can obtain strong performance on a variety of non-language tasks.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

kzl/universal-computation officialmentioned in papermentioned on GitHubpytorch report
bozeklab/lmagp mentioned on GitHubpytorch report
elipugh/tilt-transfer-transformers mentioned on GitHubpytorch report
sberbank-ai/fusion_brain_aij2021 mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutLSTMLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSigmoid ActivationSoftmaxTanh ActivationTransformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections