- Rating
- 1
- Confidence
- 4
- Informativeness
- 1
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
Existing sequential recommendation methods typically compress user history into a single representation, failing to account for the multi-interest nature and multi-centric distribution of user preferences. The paper proposes OT-FlowRec, which (i) constructs a mixture of interest-specific transport flows, each representing a distinct user interest cluster; (ii) designs a geometric transport mechanism to align latent interest distributions with target preference prototypes; and (iii) integrates a progressive time-sampling warmup strategy to bridge the gap between continuous global distributions and ranking metrics. Experimental results show that the proposed OT-FlowRec achieves clearly performance gains over baselines on benchmark datasets.
Critical strengths
-
The paper addresses an important and practical problem in multimodal sequential recommendation. Since user interests are inherently multi-centric, modeling this characteristic can better capture user preferences and improve recommendation performance.
-
The proposed OT-FlowRec model has a sound architecture and reasonable component design. The paper is also well-written and easy to understand.
-
OT-FlowRec consistently outperforms the baselines across the selected datasets, demonstrating its effectiveness.
Critical weaknesses
-
Prototype-based representation learning for disentangling multiple user interests for recommendation has been extensively studied in prior work [a–f]. However, the paper provides insufficient evidence to distinguish its approach from existing methods, limiting its novelty.
-
Even if the novelty is considered from the optimal transport perspective, related work [g] has already explored this approach. The paper should discuss the differences and better highlight its key contributions.
-
The comparisons primarily involve single-interest modeling baselines, which are insufficient to convincingly demonstrate the proposed method's advantages. The improvements may largely stem from multi-interest modeling, whose effectiveness has already been established in prior work. Furthermore, despite leveraging multimodal content for interest modeling, the paper lacks comparisons with stronger multimodal sequential recommendation baselines, such as MISSRec [h]. Table 3 shows that each proposed component contributes only marginally to overall performance, suggesting that the main gains may come from multi-interest modeling.
-
The paper lacks an evaluation of interest disentanglement, leaving it unclear whether multiple user interests are effectively disentangled.
-
The paper does not discuss the computational costs of optimal transport and multi-interest modeling, making it difficult to assess the trade-off between effectiveness and efficiency.
-
The paper lacks discussion and comparisons with closely related VAE-based approaches [d, g, i, j].
[a] Controllable Multiinterest Framework for Recommendation. KDD 2020.
[b] Re4: Learning to Re-Contrast, Re-Attend, Re-Construct for Multi-Interest Recommendation. TheWebConf (WWW) 2022.
[c] User-Aware Multi-Interest Learning for Candidate Matching in Recommenders. SIGIR 2022.
[d] Aligning Dual Disentangled User Representations from Ratings and Textual Content. KDD 2022.
[e] Disentangled Multi-Interest Representation Learning for Sequential Recommendation. KDD 2024.
[f] Parameter-Efficient Variational AutoEncoder for Multimodal Multi-Interest Recommendation. ACM MM 2025.
[g] Optimal Transport Alignment of User Preferences from Ratings and Texts. UAI 2025.
[h] MISSRec: Pre-training and Transferring Multi-modal Interest-aware Sequence Representation for Recommendation. ACM MM 2022.
[i] TopicVAE: Topic-Aware Disentanglement Representation Learning for Enhanced Recommendation. ACM MM 2022.
[j] Align MacridVAE: Multimodal Alignment for Disentangled Recommendations. ECIR 2024.
Questions for authors
Please see the review.
4 comments
Thanks for your insightful and constructive comments. We responded them point-by-point and updated the relevant content in the manuscript.
1. Distinction from prototype-based and multi-interest methods
Thank you for the constructive suggestion. We distinguished between OT-FlowRec and seven prototype-based methods you mentioned in the table below. In our model, OT-FlowRec used parallel interest transport flows to prevent interest distribution collapse. The OT-MBFM module constructs minimum-cost transport paths toward preference prototypes, while PTW emphasizes terminal generative states to align flow learning with top- ranking.
Following your suggestion, we have added a comparison result with prototype-based methods you mentioned in following table(full results on page 8 in table 2 and page 17 in table 6). Our method achieved the best performance on all four metrics across three datasets.
| Method | How the baseline works | OT-FlowRec (ours) | Main difference |
|---|---|---|---|
| ComiRec | Used dynamic routing to extract multiple interest vectors from past interactions. It retrieved items for each vector and merged the candidates. | Updated the current sequence state separately for each selected prototype and scored items with the resulting vectors. | ComiRec used extracted interest vectors directly for retrieval. OT-FlowRec used prototypes to guide how the user state is updated before scoring. |
| Re4 | Used re-contrast to distinguish interests, re-attention to align attention weights with matching scores, and reconstruction to recover representative historical items for each interest. | Used OT to define targets for state updates and trained the prediction network with the ranking loss for the observed next item. | Re4’s auxiliary tasks refined the interest vectors extracted from the history. OT-FlowRec’s flow-matching loss supervised the generation of prediction vectors. |
| ADDVAE | Learned interest factors separately from user interactions and related text, then aligned the two sets through attention so that text supplements the interaction signal. | Encoded interactions in chronological order and used the current user state to select the prototypes that guide next-item prediction. | ADDVAE learned user preferences from a set of interactions and related text. OT-FlowRec modeled next-item preferences based on the ordered interaction history.. |
| TopicVAE | Learned topics and their associations with items from item content, decomposed user interactions by topic, and combined the predictions across topics. | Selected relevant prototypes based on the current sequence and generated one prediction vector conditioned on each selected prototype. | TopicVAE used topics to decompose observed interactions. OT-FlowRec used prototypes to guide predictions along distinct interest directions. |
| Align MacridVAE | Aligned image and text representations of the same item, groups items into categories in a shared space, and used a VAE to learn user preferences within each category. | Fused item ID, image, and text features before sequence encoding. The current user state determined which prototypes guide prediction. | The prototypes defined semantic item categories in Align MacridVAE and distinct prediction directions for a user’s current sequence in OT-FlowRec. |
| BANDVAE | Used OT to align interest factors from ratings and text, then combined the preferences from both sources through bidirectional barycentric mapping. | Used OT to compute soft matches between recent interaction representations and interest prototypes. These matched define the flow-matching targets. | BANDVAE’s transport plan combined preferences from two sources. OT-FlowRec’s transport plan provided supervision for the state updated used in prediction. |
| MISSRec | Clusters image and text features to obtain interest prototypes. It learned transferable sequence representations through interest-aware encoding and decoding and contrastive pretraining. At recommendation time, it scored candidates with the vector from the final decoder position. | Retained the prediction vectors from the selected interest branches and assigned each candidate its highest matching score across branches. | MISSRec scored all candidates with one sequence vector. In OT-FlowRec, each candidate received its score from the branch that best matched it. |
2. Distinction from BANDVAE’s use of OT
Thank you for the constructive suggestion. OT-FlowRec and BANDVAE (related work g) differed in two points, (i) the learning mechanism, i.e. how they used OT during training; (ii) Input modality differences.
(i) Learning mechanism. In OT-FlowRec, we used flow-matching targets from the OT coupling between the source and target distributions. These targets directly supervised the direction of the state updates that generated next-item prediction vectors. In contrast,BANDVAE used OT to align interest factors learned from ratings and text, then combined preference information from the two sources through bidirectional barycentric mapping.
(ii) Input modality differences. OT-FlowRec input item IDs, images, and text. Item ID embeddings captured collaborative information. Images provided visual cues such as color and shape, and text describes product attributes such as specifications. The sequence encoder incorporated these complementary modality for each item. In contrast,BANDVAE only used ratings and text as the same modality, failing to fully leverage multimodal information.
3. Stronger baselines and the sources of improvement
Thank you for your constructive suggestion. Following your suggestion, we added ComiRec(KDD 2020) and Re4(WWW 2022) as multi-interest baselines and MISSRec(MM 2023) as a stronger multimodal sequential recommendation baseline. OT-FlowRec outperformed all three baselines in the table below (full results on page 8 in table 8). Its relatived NDCG@20 gains over MISSRec were 15.98%, 7.44%, and 2.05% across the three datasets.
We also evaluated contributions of the individual components. We removed OT-MBFM and PTW one at a time from the single-interest variant of OT-FlowRec. According to result,removing either component lowered every metric on all datasets (page19 table 9). These results indicate that each component also provided a certain performance gain in single-interest modeling. Hence, all components contribute to the performance gains, rather than the improvements arising from multi-interest modeling.
| Baseline | Beauty | |||
|---|---|---|---|---|
| H@10 | H@20 | N@10 | N@20 | |
| ComiRec | 0.0760 | 0.1057 | 0.0428 | 0.0503 |
| Re4 | 0.0557 | 0.0792 | 0.0320 | 0.0379 |
| MISSRec | 0.0758 | 0.1109 | 0.0376 | 0.0463 |
| ours | 0.0891 | 0.1267 | 0.0442 | 0.0537 |
| Variant | Beauty | |||
|---|---|---|---|---|
| H@10 | H@20 | N@10 | N@20 | |
| Single-interest | 0.0880 | 0.1262 | 0.0437 | 0.0532 |
| Single-interest & w/o OT-MBFM | 0.0833 | 0.1216 | 0.0414 | 0.0510 |
| Single-interest & w/o PTW | 0.0865 | 0.1243 | 0.0428 | 0.0523 |
| OT-FlowRec | 0.0891 | 0.1267 | 0.0442 | 0.0537 |
4. Interest Disentanglement Evaluation
Thank you for raising this point. To evaluate the interest disentanglement, we used Jensen–Shannon (JS) divergence[1] to evaluate the level of interest disentanglement. Higher JS divergence indicates greater differences between the interest distributions. We computed the average pairwise JS divergence between different interest prototypes and showed results in below table. The experimental results indicate that our method got a well separation among the interest representations.
[1] Si Z, Sun Z, Zhang X, et al. When search meets recommendation: Learning disentangled search representation for recommendation[C]//Proceedings of the 46th international ACM SIGIR conference on research and development in information retrieval. 2023: 1313-1323.
| Dataset | JS Divergence ↑ |
|---|---|
| Beauty | 0.5877 |
| Toys | 0.6423 |
5. Efficiency Analysis
Thank you for the suggestion. We added a complexity analysis of OT and multi-interest modeling and measurements of inference time. Based on our analysis, the time complexity are for OT and for multi-interest modeling. Here, is the maximum history length, is the number of interest prototypes, is the embedding dimension, and is the number of Sinkhorn iterations. is the number of bridges, is the number of Euler steps per bridge, and is the item set.
We ran inference on an RTX 3090 and the inference time over five runs. The table below shows sizmilar inference times across the component ablations. Together with the recommendation results, these measurements showed that OT-FlowRec balances ranking performance and inference cost.
| Dataset | Method | Full-test time (ms) |
|---|---|---|
| Toys | Full | 385.6 |
| w/o MPSD | 371.7 | |
| w/o OT-MBFM | 375.4 | |
| w/o PTW | 372.7 |
6. Comparison with related VAE methods
Thank you for your constructive suggestion. Following your suggestion, we added a discussion of VAE baselines and compared them with OT-FlowRec.
ADDVAE, TopicVAE, and Align MacridVAE learned user preferences from sets of interactions and used content in different ways. ADDVAE used attention to align interest factors learned from interactions with those learned from text, so text could complement interaction signals. TopicVAE learned topics from item content, groups user interactions by topic, and aggregated the topic-level predictions. Align MacridVAE first aligned the image and text representations of each item, then groups items into semantic categories and learned user preferences within each category. OT-FlowRec preserved interaction order and selects interest prototypes based on the current sequence state. Different orderings of the same items can therefore lead to different prototype selections and next-item predictions. OT-FlowRec used OT coupling to construct flow-matching targets that supervise state updates for next-item prediction, whereas BANDVAE used OT to align rating- and text-based interests through bidirectional barycentric mapping. For input modalities, OT-FlowRec integrated item IDs, images, and text, while BANDVAE only used ratings and text, without fully leveraging multimodal information.
As shown in the table below, OT-FlowRec outperformed all four VAE baselines on all metrics across the three datasets(page 17 table 6). Relative to BANDVAE, the strongest VAE baseline on NDCG@20, OT-FlowRec improved that metric by 14.74% on Beauty, 21.31% on Video Games, and 10.48% on Toys.
| Baseline | Beauty | |||
|---|---|---|---|---|
| H@10 | H@20 | N@10 | N@20 | |
| ADDVAE | 0.0631 | 0.0933 | 0.0330 | 0.0407 |
| TopicVAE | 0.0634 | 0.1003 | 0.0328 | 0.0420 |
| Align MacridVAE | 0.0557 | 0.0855 | 0.0287 | 0.0362 |
| BANDVAE | 0.0709 | 0.1020 | 0.0390 | 0.0468 |
| ours | 0.0891 | 0.1267 | 0.0442 | 0.0537 |
As the concerns raised in the review have not been addressed in the rebuttal phase, I would like to maintain my original score, as it continues to reflect my assessment of the paper.