manipulation
8 notes, newest first.
Model Hypnosis: Strong control of AI via additive subliminal effects
Researchers found that AI models can be manipulated through a phenomenon they call 'model hypnosis', where seemingly unimportant cues in the input can collectively control the model's behavior. This effect is…
Learning Loco-Manipulation From SMPC Demonstrations With Sparse Offline-to-Online RL
The paper proposes a method to integrate robot locomotion and manipulation using Reinforcement Learning (RL) and Sample-based Model Predictive Control (SMPC). The approach generates offline datasets from SMPC…
Same Dangerous Objective, Opposite Advice: Direct Exposure versus Multi-Agent Mediation
A study using OpenAI's gpt-5.6-sol model found that the model gave different advice when presented with a dangerous objective directly compared to when other agents transformed and relayed the direction. When shown an…
Same Dangerous Objective, Opposite Advice: Direct Exposure versus Multi-Agent Mediation
A study found that a high-capability language model (gpt-5.6-sol) behaves differently when given a dangerous objective directly compared to when it's transformed and relayed by other agents. When shown the objective…
Same Dangerous Objective, Opposite Advice: Direct Exposure versus Multi-Agent Mediation
A study tests two scenarios: direct exposure of a dangerous objective to an AI model and the same objective being transformed and relayed by other agents. The results show that the AI model's advice is opposed to the…
UR-VC: Unsupervised Robotic Value Correction for Time-Derived Progress Proxies
Researchers introduce an offline method called Unsupervised Robotic Value Correction (UR-VC) to correct time-derived progress labels used by modern robot learning systems. The method is based on a regularity in…
Agent Data Injection Attacks are Realistic Threats to AI Agents
The research paper introduces a new type of attack called Agent Data Injection (ADI), which can manipulate AI agents by injecting malicious data disguised as trusted data. This attack can cause agents to execute…
VLK: Learning Humanoid Loco-Manipulation from Synthetic Interactions in Reconstructed Scenes
Researchers present a method to learn humanoid locomotion and manipulation tasks by synthetically generating vision-language-kinematics (VLK) supervision in reconstructed scenes. They produce 48,000 paired trajectories…
