- Rating
- 1
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
ErrataWM augments frozen world models with a non-parametric memory of prediction errors. Residuals are stored as context-indexed prototypes and retrieved through applicability gates and reliability estimates to correct future predictions. Small correction modules are trained offline, while deployment adaptation uses gradient-free memory updates. Experiments evaluate prediction accuracy across navigation, manipulation, and driving, alongside navigation and manipulation success rates.
Critical strengths
- Context-indexed prototypes, positive/negative applicability scopes, and reliability updates provide a concrete mechanism for reusing historical errors while suppressing harmful corrections.
- Table 2 includes kNN, random-context, TENT, and TTT controls, with ErrataWM achieving the best prediction improvement in seven of eight settings.
Critical weaknesses
- The prior-work comparison is unreliable. Table 4 incorrectly labels Feedback-WM as updating model weights, attributes 81.6% LIBERO success to AdaPower despite its original Table 1 reporting 41.5%, and lists τ₀-WM results that I could not locate in the cited version. These issues, together with cost comparisons limited to TENT/TTT, undermine the broader competitiveness and efficiency claims.
- The control pipeline and mechanism lack sufficient validation. Multi-step prediction using corrected DINO features, corrected-image generation, and the RECON/CALVIN control protocols are insufficiently specified. Removing the context-contrastive loss barely changes prediction gains, yet no downstream ablation establishes the benefit over simpler correction. No physical closed-loop robot evaluation supports the deployment claims.
Questions for authors
- Please provide the exact source versions, tables, and evaluation settings for Table 4, and correct the weight-update classifications. Which competitiveness conclusions remain valid?
- Can you specify the implemented multi-step correction/MPC pipeline, corrected-image generation, RECON execution environment, and LIBERO/CALVIN evaluation protocols.
- What is the downstream control performance without Lcontrol, and how much does context-specific memory improve over kNN or context-independent correction under the same control settings?