AI Paper Atlas

Research map / Large language models / Agents, code and math

Early Signs of Steganographic Capabilities in Frontier LLMs

Artur Zolkowski, Kei Nishimura-Gasparian, Robert McCarthy, Roland S. Zimmermann, David Lindner

ICLR 2026 · Large language models › Agents, code and math · approach: Mathematical and structural reasoning benchmarks

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
Allow unmonitored CoT for encoder and decoder (ablation)Frontier LLMs, message passing without coordination · Message passing past GPT-4.1-mini monitortransmission successImproved
Let encoder and decoder share the encoding scheme, with CoT (ablation)Reasoning models and GPT-4.5 · Message passing with monitormonitor bypass rateImproved
Remove CoT in coordinated message passing (ablation)Frontier LLMs · Message passing with monitorsuccessHurt
Use uncommon rather than common encoding schemes (no cover task) (ablation)Frontier LLMs · Encoding and decoding messagesimplementation accuracyHurt
Increase state-tracking sequence length (first-letter, no cover task) (ablation)GPT-4.5 · State-tracking 1Daccuracy: 100% -> 90% -> 67%Hurt
Embed blatant model-generated scheme within a cover task (ablation)GPT-4.5 · State-tracking 1D, 12 movesaccuracy: 97% -> 73%Hurt
Use subtle model-generated encoding prompts (ablation)GPT-4.5 · State-tracking 1DaccuracyNo clear effect
Use common encodings (Base64/Hex) instead of unrestricted reasoning on AQuA-RAT (ablation)GPT-4.1 · AQuA-RATaccuracy: 88% -> 47-78%Hurt
Give a specific steganography scheme (strong hint)GPT-4.1 · Case-study exfiltration and reward hackinggoal success: 90% and 59%Improved

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