- Rating
- 2
- Confidence
- 4
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This manuscript proposes a memory-based framework for training-free zero-shot temporal action localization. It integrates cross-video memory, class-conditioned admission control, video-level reliability estimation, and adaptive decoding threshold to leverage test-time information without parameter updates. Extensive experiments on THUMOS14 and ActivityNet validate its effectiveness.
Critical strengths
- The experimental evaluation is comprehensive. Extensive ablation studies provide detailed insights into the contribution of each proposed component.
- The related work discussion is thorough and well-referenced. Moreover, the method demonstrates consistent performance gains over prior approaches across multiple benchmarks.
Critical weaknesses
-
The four components address different aspects of the localization pipeline, but their connections are not clearly established. As a result, the framework appears more like a combination of individual improvements than a cohesive methodological contribution.
-
The idea of leveraging test-time memory for zero-shot TAL has already been explored in [1]. Although this paper introduces parameter-free cross-video memory and additional mechanisms, its core motivation substantially overlaps with prior work. The authors should better clarify the fundamental differences from [1] beyond implementation details.
-
Key parameters, such as , , and , are mainly evaluated individually in the appendix. It remains unclear whether the framework is robust to their combined variations, particularly when generalized to other datasets.
-
In Table 3, class-conditioned admission provides only a marginal improvement on ActivityNet (+0.06 mAP). Similarly, the seven-prompt ensemble in Table 12 offers only a small gain over the best individual prompt, while several configurations in Table 16 achieve comparable performance. These observations suggest that the independent contributions of some components may be limited.
-
Table 22 shows that using mismatched or permuted class-to-direction assignments actually outperforms the correctly matched directions. This suggests that the improvements may primarily arise from a generic action–background contrast rather than class-specific modeling, weakening the central justification for the proposed admission mechanism.
[1] Jiang H, Tang H, Mu X, et al. Memory Matters: Boosting Training-Free Zero-Shot Temporal Action Localization with a Learnable Lookup Table. CVPR, 2026: 9762–9772.
Questions for authors
- Clarify the core methodological novelty compared with [1] and explain how the four components form a unified framework.
- Further justify the necessity of each component, particularly those providing only marginal improvements.
- Explain the substantial performance differences between THUMOS14 and ActivityNet, especially the limited effectiveness of class-conditioned admission on ActivityNet.
- Clarify why mismatched or permuted class-to-direction assignments outperform the correctly matched directions in Table 22.
- Provide further evidence of robustness to different hyperparameter combinations.