- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
This paper proposes GHiML for hierarchical multi-label classification. It uses a VLM to identify ecological attributes and generate structured image descriptions, then combines frozen image and text features through three classification branches and a learned router. The paper also introduces a hierarchy-aware evaluation metric. Experiments cover three benchmarks and show improvements over an image-only baseline.
Critical strengths
-
The application is well motivated. Ecological descriptions could provide useful context for recognizing rare or visually ambiguous categories in benthic imagery.
-
The evaluation includes multiple datasets, in-domain and out-of-domain benchmarks. The authors also provide useful analyses of text complementarity and the proposed metric's behavior.
Critical weaknesses
-
Replacing image-specific text with a constant embedding does not hurt either F1 or $d_{\mathrm{inv}}$ (Tab 3) on BN-Su. On FathomNet, training with noise instead of text also gives similar results (Table 10). These results leave unclear how much the method benefits from the content of the ecological descriptions.
-
Generic captions are compared only through k-NN analysis.
-
The ecological descriptions are not validated for accuracy or visual support. Attributes such as predation strategy may come from prior knowledge rather than visible evidence, and the impact of incorrect descriptions is unclear.
-
Figures 2 and 3 are difficult to read and lack a complete example showing how the descriptions improve a prediction.
Questions for authors
-
How does the method compare with generic captions and constant text when trained using the same architecture and procedure?
-
Have the authors checked the accuracy of the generated ecological descriptions, particularly for properties that cannot be directly observed in a single image?