- Rating
- 2
- Confidence
- 4
- Informativeness
- 1
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper proposes a new method for zero-shot skeleton action recognition. Specifically, it first employs an EMA-based prototype updating strategy to obtain class-center features for each action category. Then, it designs an energy formulation to address the cross-modal consistency issue. Moreover, it amplifies the discrepancies between hard neighboring categories to increase inter-class distances. Experimental results demonstrate competitive performance compared to prior works.
Critical strengths
- The paper is well written, with a clear structure and detailed presentation. The figures are intuitive and informative, and the experiments are comprehensive and thorough.
- The proposed method is technically sound, particularly the positive cross-modal neighborhood energy in CPNM. Meanwhile, the motivation is clear and provides an effective solution to address the cross-modal geometric consistency issue.
Critical weaknesses
- I believe that the cross-modal geometry issue has already been discussed in Flora [1] (CVPR 2026 Findings), particularly for generative-based architectures. Flora directly aligns the distributions of two modalities in a shared latent space, which shares similar underlying principles with the proposed method. Moreover, Flora adopts a simpler and more straightforward approach, which undermines the novelty of the proposed module. Additionally, EMA-based prototype updating has been widely explored in image-based zero-shot learning. Thus, the overall novelty and technical contributions are limited.
- Regarding HDNA, I also believe that its motivation and technical design are similar to those of FS-VAE and TDSM. Both methods aim to increase inter-class distances through similar loss function designs. However, the paper does not discuss these works and does not even cite TDSM.
- The insights regarding cross-modal consistency and inter-class separation have already been extensively discussed and explored in prior works, which diminishes the significance of the claimed contributions to the community.
- In the experimental section, the proposed method is not compared with several recent methods published in 2025 and 2026, such as Neuron (CVPR 2025), TDSM (ICCV 2025), and Flora (CVPR 2026 Findings), raising concerns about the effectiveness and competitiveness of the proposed method.
Questions for authors
- Compared to Flora, what are the advantages of Eq. (6)? Meanwhile, how does HDNA differ from the loss function designs in TDSM and FS-VAE?
- If the cross-modal geometry issue has been effectively addressed, can the proposed method also achieve competitive performance in low-shot training settings, as demonstrated in Flora? Alternatively, can it outperform prior methods in more challenging settings, such as the 60/60 split on NTU120?