TME-PSR: Time-aware, Multi-interest, and Explanation Personalization for Sequential Recommendation

Qingzhuo Wang; Leilei Wen; Juntao Chen; Kunyu Peng; Ruiyang Qin; Zhihua Wei; Wen Shen

cs.IRApr 10, 2026

TME-PSR: Time-aware, Multi-interest, and Explanation Personalization for Sequential Recommendation

Authors: Qingzhuo Wang, Leilei Wen, Juntao Chen, Kunyu Peng, Ruiyang Qin, Zhihua Wei, Wen Shen

Organizations: Tongji University

Abstract

In this paper, we propose a sequential recommendation model that integrates Time-aware personalization, Multi-interest personalization, and Explanation personalization for Personalized Sequential Recommendation (TME-PSR). That is, we consider the differences across different users in temporal rhythm preference, multiple fine-grained latent interests, and the personalized semantic alignment between recommendations and explanations. Specifically, the proposed TME-PSR model employs a dual-view gated time encoder to capture personalized temporal rhythms, a lightweight multihead Linear Recurrent Unit architecture that enables fine-grained sub-interest modeling with improved efficiency, and a dynamic dual-branch mutual information weighting mechanism to achieve personalized alignment between recommendations and explanations. Extensive experiments on real-world datasets demonstrate that our method consistently improves recommendation accuracy and explanation quality, at a lower computational cost.

Figures & tables

Figure 1: Examples of movie interaction sequences of two different users, which show different preferences for time and interests. User 1 liked romance movies a long time ago, but recently the interest has shifted to superhero movies. User 2 has always liked superhero movies, but also follows both Marvel series and DC series two sub-interests.

Figure 2: Overall architecture of our TME-PSR. At the input layer, we propose a dual-view gated time encoder. Specifically, we use a time-aware encoder to extract short-term behavior patterns and long-term behavior patterns from input timestamps. Then, we adaptively weight the two patterns based on different user’s temporal rhythm preference, so as to extract personalized time-aware embeddings. At the encoding layer, we propose a multihead LRU architecture, each head captures a fine-grained sub-interest. At the prediction layer, we design a dynamic dual-branch MI weighting mechanism, so as to adaptively adjust the degree of semantic alignment between the recommendation branch and the explanation branch.

Model	AM-Electronics				AM-Movies				Yelp
	Recommendation		Explanation		Recommendation		Explanation		Recommendation		Explanation
	R@10	N@10	R@10	N@10	R@10	N@10	R@10	N@10	R@10	N@10	R@10	N@10
PITF	0.0286	0.0139	0.0880	0.0547	0.0235	0.0101	0.0984	0.0679	0.0204	0.0094	0.0333	0.0185
BPER	0.0307	0.0154	0.0901	0.0553	0.0281	0.0131	0.1028	0.0687	0.0212	0.0106	0.0393	0.0214
SCEMIM-GRU4Rec	0.0290	0.0152	0.0948	0.0629	0.0432	0.0216	0.1151	0.0771	0.0452	0.0228	0.0417	0.0233
SCEMIM-SASRec	0.0369	0.0195	0.1011	0.0681	0.0541	0.0257	0.1229	0.0809	0.0592	0.0364	0.0419	0.0231

Table 1: Recommendation and explanation performance, best results are marked in bold, second best results underlined. The improvement row is calculated as the percentage increase of the best result relative to the second best result.

Strategy			AM-Electronics
Strategy			Rec		Exp
Time-aware	Multi-interest	Explanation	R@10	N@10	R@10	N@10
			0.0327	0.0164	0.1087	0.0727
✓			0.0402	0.0197	0.1371	0.0912
	✓		0.0361	0.0184	0.1312	0.0849
		✓	0.0324	0.0160	0.1252	0.0844

Table 2: Ablation of key strategies the AM-Electronics dataset, best results are marked in bold, second best results underlined.

Figure 3: Comparison of four gating strategies. Results show that the model that adaptively focuses on both long-term and short-term features consistently achieves the best performance, which proves the effectiveness of the proposed method.

Figure 4: Left two columns show recommendation and explanation performance with different head numbers

H

under different embedding sizes

d

. Results show that the optimal head number increases with the increase of

d

. The third column shows the relationship between

H

and the encoder parameters. As

H

increases, the number of parameters decreases inversely.

Figure 5: Inference and training efficiency of GRU4Rec, SASRec, LRURec and the proposed multihead LRU. Results show that the proposed method outperforms the other three methods, and its efficiency increases with the head number

H

Figure 6: Visualization of multihead LRU output representation at $d=120$ , $H=6$ with t-SNE.

Figure 7: Comparison of three MI weighting strategies. Results show that the model with two dynamic weights consistently achieves the best performance, which proves the effectiveness of the proposed dynamic dual-branch MI weighting mechanism.

Figure 8: (a) Visualization of distributions of all interaction sequences average dynamic MI weights $\bar{\mu}^{\text{rec}}$ and $\bar{\mu}^{\text{exp}}$ . (b) K-means clustering based on $(\tilde{\mu}^{\text{rec}},\tilde{\mu}^{\text{exp}})$ .

Appendix figures & tables6 assets

Supplementary material from the paper’s appendix.

Appendix

Dataset	Movies	Electronics	Yelp
# Users	22,248	18,129	90,742
# Items	15,458	10,435	46,372
# Explanations	31,451	23,541	119,874
# Actions	305,801	144,772	1,204,515
Avg. actions/users	13.75	7.95	13.27
Avg. actions/items	19.78	13.87	25.98

Appendix

Table 3: The statistics of the preprocessed datasets.

Model	AM-Electronics		AM-Movies		Yelp
	Recommendation		Recommendation		Recommendation
	R@10	N@10	R@10	N@10	R@10	N@10
GRU4Rec	0.0259	0.0134	0.0431	0.0208	0.0451	0.0228
SASRec	0.0342	0.0182	0.0536	0.0256	0.0604	0.0364
FMLPRec	0.0357	0.0186	0.0558	0.0253	0.0620	0.0403
LRURec	0.0345	0.0159	0 .0560	0.0259	0.0625	0.0397

Appendix

Table 4: overall with non-SCEMIM models

Strategy			AM-Electronics				AM-Movies				Yelp
Strategy			Rec		Exp		Rec		Exp		Rec		Exp
Time-aware	Multi-interest	Explanation	R@10	N@10	R@10	N@10	R@10	N@10	R@10	N@10	R@10	N@10	R@10	N@10
			0.0327	0.0164	0.1087	0.0727	0.0549	0.0269	0.1171	0.0830	0.0650	0.0397	0.0422	0.0230
✓			0.0402	0.0197	0.1371	0.0912	0.0563	0.0266	0.1287	0.0841	0.0672	0.0410	0.0454	0.0250
	✓		0.0361	0.0184	0.1312	0.0849	0.0550	0.0274	0.1300	0.0897	0.0652	0.0399	0.0428	0.0234
		✓	0.0324	0.0160	0.1252	0.0844	0.0567	0.0272	0.1242	0.0868	0.0654	0.0399	0.0437	0.0238

Appendix

Table 5: Ablation of key strategies on three datasets, best results are marked in bold, second best results underlined.

Figure 9: Impact of hyperparameters across three datasets. Results show that when set

\alpha=0.9

and set

\beta=0.1

, the proposed model achieve the best performance.

Figure 10: Relationship between the gating parameters

\gamma^{\text{rec}}

and the average time interval

\bar{t}

(left) and relationship between the gating parameters

\gamma^{\text{exp}}

and

\bar{t}

(right). Each point represents a single user sequence. Results show that as

\bar{t}

increases, the two gating parameters decrease linearly.

Figure 11: Examples of three representative user types, correspond to the three clusters in Figure 7(b).

Sequential Recommendation

Sequential recommendation predicts what a user will choose next from ordered interactions, adapting to changing interests and context.

See all 136 papers

Dual-Interest Sequential Product Recommendation With Multi-Granular SSM
Sep 18, 2026Shuiying Liao, P. Y. MokSequential Recommendation User Preference Modeling
The Hong Kong University of Science and Technology Clear Water Bay, Hong Kong
T-RoPE: Time-Aware Rotary Position Embedding for Sequential Recommendation
Sep 24, 2026Yang Liu, Noel Loo, Ali Khanafer +4Rotary Positional Embeddings Sequential Recommendation
Shopify · Massachusetts Institute of Technology · Liquid AI
Debiasing Sequential Recommendation with Time-aware Inverse Propensity Scoring
Mar 5, 2026Sirui Huang, Jing Long, Qian Li +2Exposure Bias Sequential Recommendation
Hong Kong Polytechinic University Hong Kong SAR, China · University of Technology Sydney Sydney, Australia · Curtin University Perth, Australia +1

Sep 18, 2026J/K move · Enter open · S save

Dual-Interest Sequential Product Recommendation With Multi-Granular SSM

Shuiying Liao, P. Y. Mok

The Hong Kong University of Science and Technology Clear Water Bay, Hong Kong

Sequential Recommendation User Preference Modeling Selective SSMs

Open paper AskPDF arXiv

Sequential recommendation aims to predict the next item a user will interact with based on their historical behavior. Advances in Transformers have significantly improved sequential recommendation but are still limited by cost efficiency. Although State Space Models (SSMs) have recently enabled efficient long-range modeling, most existing methods encode each item with a single static contextual role, overlooking the phenomenon of item polysemy. In fact, the same item often plays different semantic roles depending on user context, and existing methods are limited in capturing dynamic behavior across different temporal granularities. In this work, we propose DSRec, a novel dual-interest cross-SSM model that explicitly disentangles item roles across long-term and short-term semantic context. Sequential items are encoded into long-term interest embeddings that capture stable preferences via historical aggregation, and a short-term interest branch that emphasizes local session intent modulated by inter-click time intervals. These interest embeddings are processed through distinct SSM encoders: a full-sequence Mamba for long-term modeling, and a time-modulated SSM that dynamically adjusts state evolution based on temporal gaps. To enable effective cross-granularity alignment, we adopt a residual cross-fusion mechanism that exchanges contextual information between the two branches while preserving semantic independence. Experiments on public benchmarks demonstrate that DSRec outperforms other state-of-the-art methods.

Abstract

Sequential Recommendation

Explore similar work

Dual-Interest Sequential Product Recommendation With Multi-Granular SSM

T-RoPE: Time-Aware Rotary Position Embedding for Sequential Recommendation

Debiasing Sequential Recommendation with Time-aware Inverse Propensity Scoring