AI Paper Atlas

Research map / Large language models / Reasoning and chain of thought

CellDuality: Unlocking Biological Reasoning in LLMs with Self-Supervised RLVR

Yuhang Chen, Zhen Tan, Ruichen Zhang, Mufan Qiu, Tianlong Chen

ICLR 2026 · Large language models › Reasoning and chain of thought · approach: RL with verifiable rewards for reasoning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes CellDuality, a self-supervised RLVR approach for unlocking open-ended biological reasoning in LLMs, presented at ICLR 2026.

Problem

Existing LLMs handle predictive biology tasks like cell type annotation but struggle with open-ended, mechanistic biological reasoning; RLVR typically needs verifiable answers that are scarce in biology.

Approach

Uses self-supervised signals to create verifiable rewards for RL training of LLMs on biological reasoning, avoiding the need for hand-labeled verifiable answers. Details beyond the title and truncated abstract are not available.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add self-supervised duality RL after SFT (ablation)CellDuality (Llama-3.2-3B) · sci-Plex3 perturbation response (OOD)scFID: 0.045 -> 0.038Improved
add self-supervised duality RL after SFT (ablation)CellDuality · Bastidas-Ponce-2019 annotation (OOD)accuracy: 72.87 -> 88.45Improved
add self-supervised duality RL after SFT (ablation)CellDuality · Segerstolpe-2016 annotationaccuracy: 98.76 -> 99.81Improved
add self-supervised duality RL after SFT (ablation)CellDuality · Drug sensitivity GSE110894 (OOD)accuracy: 96.45 -> 96.12Hurt
add self-supervised duality RL after SFT (ablation)CellDuality · Drug sensitivity GSE149383accuracy: 98.91 -> 99.12No clear effect
add self-supervised duality RL after SFT (ablation)CellDuality · Conditional cell generationGW distance: 63.87 -> 61.45Improved
use ground-truth reward RL instead of duality reward (ablation)CellDuality · sci-Plex3scFID: 0.038 -> 0.025Improved
use ground-truth reward RL instead of duality reward (ablation)CellDuality · He-2020-Livermacro F1: 95.41 -> 94.85Hurt
use ground-truth reward RL instead of duality reward (ablation)CellDuality · He-2020-Liveraccuracy: 96.34 -> 97.21Improved
use CellDuality instead of C2S-Scale SFTLlama-3.2-3B · sci-Plex3scFID: 0.02 -> 0.038Hurt

Experimental setup

Builds on: RLVR

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