AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

In-Context Algorithm Emulation in Fixed-Weight Transformers

Jerry Yao-Chieh Hu, Hude Liu, Jennifer Yuntong Zhang, Han Liu

ICLR 2026 · Architectures and efficiency › Attention and transformers · approach: Transformer expressivity and learnability

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase number of attention heads from 1 to 12 (ablation)multi-head attention emulating a softmax head · Attention-head emulationMSE: 3.469 -> 0.686Improved
use one frozen prompt-programmed attention instead of per-algorithm trained attentionfrozen attention · Synthetic LassoMSE: 0.068 -> 0.059Improved
use one frozen prompt-programmed attention instead of per-algorithm trained attentionfrozen attention · Synthetic linear regressionMSE: 0.147 -> 0.120Improved
use one frozen prompt-programmed attention instead of per-algorithm trained attentionfrozen attention · Synthetic ridge regressionMSE: 0.004 -> 0.071Hurt
use one frozen prompt-programmed attention instead of per-algorithm trained attentionfrozen attention · Ames Housing ridge regressionMSE: 0.0132 -> 0.0252Hurt
use one frozen prompt-programmed attention instead of per-algorithm trained attentionfrozen attention · Ames Housing LassoMSE: 0.0354 -> 0.0322Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy