Pith. sign in

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

arxiv 2411.18676 v2 pith:WAGF7VW6 submitted 2024-11-27 cs.RO cs.AIcs.LG

classification cs.ROcs.AIcs.LG
keywords instructionsmodelschallengingsafetyteamingassessingbenchmarkscurrent
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
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.

Discussion (0). Sign in to comment.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Prompt-Driven Exploration

    cs.LG 2026-07 conditional novelty 7.0 of 10

    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.

  2. RL$^2$-VLA: Adaptive RL Latent Compositional Steering with Test-Time Scaling for Vision-Language-Action Models

    cs.RO 2026-07 conditional novelty 6.0 of 10

    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.

  3. Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses

    cs.CR 2026-03 unverdicted novelty 6.0 of 10

    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.

  4. State Backdoor: Towards Stealthy Real-world Poisoning Attack on Vision-Language-Action Model in State Space

    cs.CR 2026-01 conditional novelty 6.0 of 10

    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.

  5. Code Monitor Red Teaming for Public-Test-Passing Code

    cs.AI 2026-07 conditional novelty 5.0 of 10

    Weak LLM verifiers miss most hidden bugs in public-test-passing code, and adversarial pressure makes that monitoring worse.

  6. Imitation Learning Based on Disentangled Representation Learning of Behavioral Characteristics

    cs.RO 2025-09 conditional novelty 4.0 of 10

    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.

  7. Steering Robots with Inference-Time Interactions

    cs.RO 2025-06 conditional novelty 4.0 of 10

    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.

Pith tools