- Rating
- 2(+1)
- Confidence
- 3
- Informativeness
- 1
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper proposes FaME, a lightweight classifier for multimodal tasks with limited labeled data. It fits a Gaussian discriminant model separately to each frozen modality representation and combines their class evidence through a prior-corrected product rule. Missing modalities are handled by omitting their evidence. The authors evaluate the method on six image–text datasets under few-shot supervision and simulated modality missingness.
Critical strengths
- The method is simple, inexpensive to adapt, and works with independently pretrained encoders whose representation spaces differ.
- The evaluation covers multiple datasets, encoder pairs, and data-scarcity settings. The few-shot results are strong relative to both learned fusion and encoder-adaptation baselines.
- The ablations include useful controls, particularly unimodal classifiers, alternative fusion rules, and covariance estimators.
Critical weaknesses
- The methodological contribution is limited. FaME combines established modality-specific GDA classifiers with a classical prior-corrected product rule. Under the shared-covariance assumption, its prediction reduces to summing modality-specific discriminant scores with a class-prior term. It introduces neither a learned cross-modal interaction nor a mechanism for estimating the relative reliability of different modalities. The contribution therefore appears primarily empirical, and it is unclear whether this combination constitutes a sufficiently substantial technical advance.
- The role of the shared coefficient α is not supported by the stated formulation. With a uniform multiclass prior, any positive shared α leaves the predicted class unchanged; for multilabel tasks, it also preserves the per-label rankings used to compute average precision. Nevertheless, Table 13 reports different results for the shared-α and α=1 variants. This suggests a mismatch between the mathematical formulation, experimental protocol, and implementation, and weakens the claimed role of α as a fusion calibration parameter.
- The CUB few-shot protocol cannot be implemented as currently described. CUB has 200 classes but only 5,000 training and 994 validation examples. Constructing 64-shot training or even 8-shot validation sets would require more examples than are available unless an unstated sampling or truncation rule is used. This ambiguity affects both the interpretation and reproducibility of the few-shot results.
Questions for authors
- What is the specific technical contribution of FaME beyond applying existing GDA estimators and a classical prior-corrected product rule to frozen multimodal representations?
- What causes the shared-α and α=1 variants to differ in Table 13 under the stated equations and evaluation metrics? Were they evaluated using exactly the same support sets and missingness masks?
- How are the CUB training and validation sets constructed when k times the number of classes exceeds the available examples? Please report the sampling rule and the number of distinct examples used at each budget.
- Could the authors reconcile the Appendix F.3 discussion with Table 24? The table entries give an average of approximately 0.484 for mean late fusion at r=0.75, rather than the stated 0.522, and FaME is higher on six rather than four encoder pairs.
- How robust is FaME to imbalanced few-shot support sets or label-prior shift between the support and test sets? The balanced few-shot protocol makes the empirical prior uniform and does not evaluate this dependence.