claim
active
claim:failure-episodes-are-primarily-caused-by-hand-occluding-the-sticker-in-the-mirror-or-sticker-leaving-the-visual-field-due-to-head-rotation-plus-kinematic-reachability-limitsFailure episodes are primarily caused by hand occluding the sticker in the mirror or sticker leaving the visual field due to head rotation, plus kinematic reachability limits
Explains the ceiling on removal success as due to perceptual and kinematic constraints, not principled failures
Source paper
extracted_from(2026) · Dongmin Kim · Hoshinori Kanazawa · Yasuo Kuniyoshi
Neighborhood — ranked by edge-count
Findings (1)
finding
- Main behavioral result demonstrating the model's efficacy in the mirror-mark task
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Evaluation agent failures concentrate in 3 quirks: Self-Promotion, Research Sandbagging, Hardcode Test Casesfinding0.757Subtle or hard to elicit quirks.
- Karpathy case.
- Core mechanism claim linking mismatch detection to behavior through EFE minimization
- Current degeneracy mode.
- Acknowledges the confound of not explicitly instructing models to track wealth, yet points to reasoning gaps given code agents avoid errors without prompts.
- Concrete failure signatures extracted from traces.
- Causal effect: feature induces perception of bugs.
- Connects the model's behavior to Zaadnoordijk and Bayne's taxonomy of intentional agency