6 papers
NavBind: Diagnosing Instruction-Action Binding at Route Forks in Navigation VLAs
Josh Qixuan Sun, Xiaoying Xing, Yurun Chen +3
Vision-language-action (VLA) navigation models translate visual observations and instructions into actions. Navigation success assesses whether an agent reaches the goal. However,…
HygieneRoboBench: Benchmarking Hygiene-Aware Planning for Household Robots
Yurun Chen, Josh Qixuan Sun, Jason Qin +4
Contact with contaminated objects can spread hazards through a household robot's grippers, tools, and shared surfaces, while new contacts can make an existing plan unsafe. Existing…
Training and Evaluating Ethical Reinforcement Learning Agents on Per-Episode Distributions
Prabhjyot Singh, Majid Ghasemi, Mark Crowley
Reinforcement Learning (RL) agents trained on a single reward signal exploit the gap between the designed reward and the intended behavior. This is particularly a problem when we a…
Preference-based Antibody Expression Ranking: Scaling with Large-scale Weak Supervision
Josh Qixuan Sun, Morteza Babaie, Wenyang Hou +2
Antibody expression ranking is a critical task in antibody design, yet its modelling is severely hindered by the scarcity of labeled experimental data. To address this, we propose…
Learning When to Trust in Contextual Social Bandits
Majid Ghasemi, Mark Crowley
Robust reinforcement learning typically assumes that feedback sources are either globally trustworthy or corrupted within a fixed global budget. We identify a more subtle failure m…
Objective Decoupling in Social Reinforcement Learning: Recovering Ground Truth from Sycophantic Majorities
Majid Ghasemi, Mark Crowley
Contemporary AI alignment strategies rely on a fragile premise: that human feedback, while noisy, remains a fundamentally truthful signal. In this paper, we identify this assumptio…