REVIEW 7 cited by
Embodied Red Teaming for Auditing Robotic Foundation Models
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Language-conditioned robot models have the potential to enable robots to perform a wide range of tasks based on natural language instructions. However, assessing their safety and effectiveness remains challenging because it is difficult to test all the different ways a single task can be phrased. Current benchmarks have two key limitations: they rely on a limited set of human-generated instructions, missing many challenging cases, and focus only on task performance without assessing safety, such as avoiding damage. To address these gaps, we introduce Embodied Red Teaming (ERT), a new evaluation method that generates diverse and challenging instructions to test these models. ERT uses automated red teaming techniques with Vision Language Models (VLMs) to create contextually grounded, difficult instructions. Experimental results show that state-of-the-art language-conditioned robot models fail or behave unsafely on ERT-generated instructions, underscoring the shortcomings of current benchmarks in evaluating real-world performance and safety. Code and videos are available at: https://s-karnik.github.io/embodied-red-team-project-page.
Forward citations
Cited by 7 Pith papers
-
Prompt-Driven Exploration
Prompt-Driven Exploration refines language prompts from rollout videos via a VLM, enabling RL to escape zero-reward VLA and LLM policies where action-space noise fails.
-
RL$^2$-VLA: Adaptive RL Latent Compositional Steering with Test-Time Scaling for Vision-Language-Action Models
RL² improves VLA robot success rates by conditionally composing an offline RL policy's actions with the frozen VLA only when a failure detector flags impending failure.
-
Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses
The survey organizes over 400 papers on embodied AI safety into a multi-level taxonomy and flags overlooked issues such as fragile multimodal fusion and unstable planning under jailbreaks.
-
State Backdoor: Towards Stealthy Real-world Poisoning Attack on Vision-Language-Action Model in State Space
A backdoor attack on vision-language-action robot policies uses the arm's initial joint configuration as the trigger, achieving >90% triggered failure with only small clean-task degradation.
-
Code Monitor Red Teaming for Public-Test-Passing Code
Weak LLM verifiers miss most hidden bugs in public-test-passing code, and adversarial pressure makes that monitoring worse.
-
Imitation Learning Based on Disentangled Representation Learning of Behavioral Characteristics
A weakly-supervised CVAE with action chunking lets a robot change wiping speed online from instruction labels, but the same mechanism fails to disentangle wiping force and fails on spatial pick-and-place directives.
-
Steering Robots with Inference-Time Interactions
Frozen imitation policies can be steered at inference time via user interactions, with a diffusion-sampling method and a constraint-enforcing framework that provides formal task guarantees.
Discussion (0). Sign in to comment.