- Rating
- 3
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 3
Read review
Paper summary
This paper addresses Long-Tailed Domain Incremental Learning (LDIL), where class imbalance and domain shifts occur simultaneously. The proposed DC-Comp introduces Intra-class Balanced Discriminativity Mining to augment local features through discriminative token sampling, particularly for tail classes, and Inter-domain Discriminativity Reserving and Complementing to preserve historical features and generate feature-level pseudo-samples for training on subsequent domains. Extensive experiments across multiple benchmarks and settings demonstrate improvements over existing methods.
Critical strengths
- The paper addresses a practical and underexplored problem by considering long-tailed distributions in domain incremental learning.
- The proposed discriminativity mining and inter-domain knowledge complement mechanisms are clever and intuitive, effectively addressing the key challenges of LDIL.
Critical weaknesses
- Despite the extensive baseline comparisons, the experimental analysis is somewhat limited. It is unclear how DC-Comp differs from the strongest baselines and what leads to the performance gaps. Also, some baselines were originally designed for different learning settings. It would be useful to clarify how they were adapted to LDIL, including hyperparameter choices, and categorize them by their original learning settings.
- Although the proposed mechanisms are intuitive and interesting, the paper provides limited discussion of the insights underlying their design. The paper mainly focuses on describing the procedures and reporting empirical improvements, without sufficiently explaining the reasoning behind the design choices and their implications for learning under class imbalance and domain shifts. A more insightful discussion would strengthen the methodological contribution.
Questions for authors
- Are samples excluded during long-tail subsampling also excluded from evaluation? How many trials were conducted, and what are the standard deviations of the reported results?
- In Section 3.2, what exactly are the keys used for cosine-similarity-based prompt matching? Are they learnable parameters or class prototypes?
- How does performance change with different Pareto distribution parameters or imbalance ratios?
- Why are the improvements on ImageNet-C substantially larger than those on other datasets?
- Among the newly introduced components of the proposed method, which hyperparameters are predefined and which are dynamically determined during training?
5 comments
Response to the questions
Thank you for the positive assessment and for the constructive suggestions. Our responses to the questions follow, point by point.
Q1: Long-tail subsampling, trials, and standard deviations
(1) As specified in Eq.(11) and (12) of our paper, no sampling is performed during evaluation. This is because the learned attention layer can be directly generalized to the full set of visual tokens, eliminating the need for sampling at inference time.
(2) The inference process can be formulated as: $ f(x)=\theta_c(\theta_a(\theta_b(x))), $ where denotes the input image, and , , and represent the prompt-conditioned backbone, attention-based token aggregation module, and classifier, respectively.
(3) Accordingly, all test images are processed using the same deterministic inference pipeline, without multiple-trial sampling or random evaluation procedures. Therefore, the evaluation results are deterministic under a fixed model and test configuration, and reporting standard deviations across repeated sampling trials is not applicable.
(4) We also conducted a supplementary token-selection experiment on ImageNet-C using the classifier saved in the available checkpoints. For each test image, we compared inference using all 197 visual tokens with inference using seven visual tokens sampled via Gumbel Top-k. We repeated the sampled inference five times. Accuracy is averaged over 15 stages, and the standard deviations reflect the five sampling trials.
| Setting | All tokens | Sampled tokens |
|---|---|---|
| Unif | 65.06% | 63.44% ± 0.06% |
| Rand | 65.87% | 64.12% ± 0.07% |
| Origin | 87.33% | 86.26% ± 0.02% |
The experimental results show that utilizing all tokens consistently outperforms token sampling under both long-tailed and original data settings. This is because retaining all tokens preserves more comprehensive image information, allowing the model to jointly exploit complementary semantic cues from different tokens and thereby improve inference performance.
Q2: Keys used for prompt matching
(1) The keys in Section 3.2 are learnable embeddings associated with the prompts, as in the KA-Prompt approach. They are not class prototypes. (2) Specifically, each prompt is paired with a key ; the prompt is selected by comparing that key with the image query using cosine similarity.
Q3: Sensitivity Analysis of the Pareto parameter
(1) We reported only in the paper because it is a commonly used setting in long-tailed learning. (2) We also evaluated different values of . As shown below, our method surpasses the state-of-the-art Dual-Prompt+CSA by 10.6%-14.9% across varying .
| Methods | =2 | =4 | =6 | =8 | =10 |
|---|---|---|---|---|---|
| Dual-Prompt+CSA | 77.8 | 74.2 | 68.5 | 56.3 | 40.2 |
| Ours | 88.4 | 76.0 | 80.3 | 69.7 | 55.1 |
Q4: Larger gains on ImageNet-C
Thank you for your careful reading. (1) The more pronounced performance gains of our method on ImageNet-C can be attributed to the construction of its domains, where different corruptions are applied to the same source images. Consequently, object content exhibits stronger semantic consistency across domains than in other benchmarks. (2) This cross-domain semantic consistency makes discriminative knowledge preserved from historical domains particularly valuable for learning the same classes in subsequent domains, thereby explaining the greater performance gains achieved by the proposed Inter-domain Discriminativity Reserving and Complementing (IDRC) module.
Q5: Predefined and dynamically determined hyperparameters
(1) The token group size is predefined and fixed at 7 in our experiments. (2) In contrast, the class balance ratio is computed from the class frequencies in the current domain. The number of sampling rounds and the number of preserved tokens are then determined dynamically from that distribution, with calculated according to Eq. (10).
We hope these responses clarify the design and evaluation of DC-Comp. Thank you again for your constructive suggestions.
Response to Authors
Thank you for the detailed responses and additional experiments. I appreciate the authors' efforts to address my concerns.
-
Q1: My question concerned dataset-level long-tail subsampling and variance across independent training runs, rather than token sampling during inference. Thus, the response does not directly address my original question.
-
Cross-domain complementarity: The explanation for the larger gains on ImageNet-C is reasonable and helps clarify why the improvements are relatively smaller on other benchmarks. This suggests that the effectiveness of the proposed method may depend on the degree of semantic consistency between domains, as class-discriminative features learned in one domain are not necessarily transferable to another.
Overall, I find the proposed approach interesting and the results promising. I maintain my original rating (Weak Accept).
Response to the weaknesses
Thank you for your positive review and constructive feedback. Below is our point-by-point response.
W1: Experimental analysis
(a) Difference from baselines: (1) Our DC-Comp is built on the KA-Prompt baseline by exploiting its prompt learning mechanism. (2) Compared to KA-Prompt, we introduce three original designs: (i) Discriminativity Estimation Module (DEM) identifies class relevant patch tokens, (ii) Intra-class Balanced Discriminativity Mining (IBDM) adjusts token sampling according to class frequency for information-balanced learning, (iii) Inter-domain Discriminativity Reserving and Complementing (IDRC) then uses complementary information from the same classes in previous domains.
(b) Performance gaps: (1) Existing domain incremental learning (DIL) methods, including the KA-Prompt baseline, are primarily designed for balanced data distributions. Consequently, their learning capacity is limited when facing long-tailed data streams. (2) DIL+LTL methods integrate long-tailed learning (LTL) mechanisms into DIL frameworks to improve intra-domain learning under imbalanced data distributions. However, existing LTL methods are primarily designed for static learning scenarios and overlook the complementary information across domains, restricting knowledge acquisition in LDIL. (3) In contrast, our method explicitly mines and reuses discriminative semantics across domains to facilitate learning from long-tailed data streams. This capability is enabled by the complementary designs of DEM, IBDM, and IDRC, which jointly promote reliable semantic acquisition and cross-domain information utilization.
(c) Adaptation to LDIL: (1) All compared DIL and DIL+LTL methods are implemented using the same ViT-B backbone and evaluated on identical LDIL benchmarks under consistent training settings, ensuring a fair comparison. (2) For each approach, we conduct an independent hyperparameter grid search to identify its optimal configuration.
W2: Discussion of the insights
The three proposed components address successive sources of insufficient information in LDIL through complementary designs:
(1) DEM estimates the reliability of individual tokens, enabling the model to identify and prioritize informative features while suppressing unreliable information.
(2) Building upon the estimated token-level reliability, IBDM adaptively allocates more augmentation opportunities to underrepresented classes, thereby alleviating the insufficient intra-class information caused by long-tailed distributions.
(3) IDRC preserves selected historical tokens and generates pseudo samples by combining tokens from historical instances, enabling complementary same-class knowledge from previous domains to facilitate current-domain learning.