From Concepts to Actual Causes: Toward Closing the Concept-Malfunction Gap in Robot Foundation Models
cs.RO, cs.AI
Submitted: 2026-10-30
Updated: 2026-10-30
Comments: 4 pages, 1 figure, 1 table. Accepted at the Science of Physical AI Safety (SPAIS) Workshop at CoRL 2026
License: http://creativecommons.org/licenses/by/4.0/
The gist: Post-incident analysis of a robot foundation model poses a singular causal question: which internal components caused this failure, and to what degree? Mechanistic interpretability (MI)…
Terminology
Abstract
Post-incident analysis of a robot foundation model poses a singular causal question: which internal components caused this failure, and to what degree? Mechanistic interpretability (MI), however, establishes general causal claims by decoding the concepts a policy has learned, yet knowing which concepts are encoded is insufficient for identifying the actual cause of a specific incident. In this position paper, we analyze this concept-malfunction gap by posing fault isolation in the framework of actual causality. Specifically, we formalize a fault-isolation query as a tuple of intervened internal components, counterfactual values, contingencies, and read-outs, and lay out conditions for it to be well-specified. Through this lens, we identify shortcomings of current MI approaches: they assume fixed contingencies and read out only immediate effects, letting backups obscure actual causes. We demonstrate on a toy example that these shortcomings can lead to conflicting verdicts. Lastly, based on the well-specification conditions, we outline four open research tasks toward closing the gap.
Sources
- World Action Models are Zero-shot Policies
- Beyond Task Success: Behavioral and Representational Diagnostics for WAM and VLA
- The Hydra Effect: Emergent Self-repair in Language Model Computations
- Steering Language Models With Activation Engineering
Related papers
- Rethinking Actions: Continuity-Aware Priors for World-Model Planning
- MOSAIC: Modular Obstacle and Spatial-Aware Framework for Manipulation In Cluttered Scenes
- SAGAS: Semantic-Aware Graph-Assisted Stitching for Offline Temporal Logic Planning
- DexSeed: Expanding Human Hand-Object Interactions into Robot Demonstrations for Dexterous Policy Learning
- LaST0.5: A Generalist World Action Model with Interaction-Centric Latent Reasoning
- CIALLO: Learning Shared Intent Representations for Cross-Embodiment Manipulation