AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Robust In-Context Reinforcement Learning Under Reward Poisoning Attacks

Paulius Sasnauskas, Yiğit Yalın, Goran Radanovic

ICML 2026 · Reinforcement learning › Visual and representation RL · approach: Scalable RL systems and environments · 0 citations (OpenAlex)

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
add adversarial training to DPTDPT · Bandit (eps=0.4, attacker targets AT-DPT)cumulative regret: 63.6 -> 24.2Improved
add adversarial training to DPTDPT · Bandit (clean environment)cumulative regret: 11.5 -> 13.0Hurt
add adversarial training to DPTDPT · Bandit (eps=0.4, uniform random attack)cumulative regret: 37.2 -> 38.7No clear effect
add adversarial training to DPTDPT · Linear bandit (eps=0.4, attacker targets AT-DPT)cumulative regret: 70.29 -> 2.49Improved
add adversarial training to DPTDPT · Darkroom2 (eps=0.4, attacker targets AT-DPT)episode reward: 216.1 -> 242.2Improved
add adversarial training to DPTDPT · Darkroom2 (clean environment)episode reward: 306.8 -> 267.4Hurt
add adversarial training to DPTDPT · Miniworld (eps=0.4, attacker targets AT-DPT)episode reward: 81.2 -> 104.8Improved
train against adaptive attacker instead of non-adaptive (ablation)AT-DPT · Bandit (eps=0.4, adaptive attacker targeting AT-DPT)cumulative regret: 88.1 -> 37.1Improved
train against adaptive attacker instead of non-adaptive (ablation)AT-DPT · Bandit (eps=0.4, non-adaptive attacker targeting AT-DPT)cumulative regret: 22.8 -> 38.0Hurt
train against a strong learned attacker rather than uniform random noise (ablation)AT-DPT · Bandit (eps=0.4, attacker targets AT-DPT)cumulative regret: 39.3 -> 24.2Improved
train with 30% suboptimal expert actions (ablation)AT-DPT · Bandit (eps=0.4, attacker targets AT-DPT)cumulative regret: 24.2 -> 41.2Hurt
train with 10% suboptimal expert actions (ablation)AT-DPT · Darkroom2 (eps=0.4, attacker targets AT-DPT)episode reward: 242.2 -> 225.2Hurt

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