Order-Agnostic Decoding for Sample-Efficient RNA Inverse Folding

Antonia Panescu, Shujun He, Yixuan He, Rex Ying

Keywords: RNA Inverse Folding, RNA Secondary Structure Inverse Design, Motif In-Painting, Order-Agnostic Decoding, AI4RNA, Reinforcement Learning

TL;DR: We replace the autoregressive decoder for 2D RNA inverse folding with random-order decoding, resulting in 5x more perfect designs per sample and motif in-painting capability.

Abstract: RNA inverse folding, the task of designing a sequence that folds into a target structure, is a core computational task in RNA synthetic biology. However, current methods rely on best-of-$N$ screening to compensate for low per-sample reliability and left-to-right autoregressive sequence generation, which locks in upstream nucleotides before downstream pairing partners are observed and cannot natively support in-painting tasks. Here, we introduce an order-agnostic decoder for 2D secondary-structure inverse design, trained under a uniform-permutation distribution over decoding orders. On the OpenKnot Round 7b 240-mer pseudoknot benchmark, our method produces five times as many perfect-structure designs per sample as the strongest autoregressive baseline, while being more sample-efficient, and recovers diversity through order randomization rather than token sampling. To showcase in-painting capability, we perform both motif preservation and motif redesign, with the latter strictly impossible under autoregressive decoding. Together, these capabilities could enable functional motif redesign for aptamer and riboswitch engineering, and reframe sample-efficient generation as a principled alternative to the dominant best-of-$N$ regime in RNA design.

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 244