Oren Barkan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Oren Barkan has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Noam Koenigstein
Tel Aviv University, Tel Aviv, Israel
13 shared publications - Yehonatan Elisha
Tel Aviv University, Tel Aviv, Israel
5 shared publications - Ori Katz
Microsoft, Herzliya, Israel and Technion, Haifa, Israel
4 shared publications - Itzik Malkiel
Microsoft, Tel Aviv Yaffo, Israel
3 shared publications - Veronika Bogina
The University of Haifa, Haifa, Israel
3 shared publications - Avi Caciularu
Bar-Ilan University Ramat-Gan Israel
2 shared publications
Latest Publications
- Open Access
- 22 November 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Recommender Systems
- Vol. 4 (2) , 1-39
LXR: Learning to eXplain Recommendations
Recommender systems have become integral to many online services, leveraging user data to provide personalized recommendations. However, as these systems grow in complexity, understanding the rationale behind their recommendations becomes increasingly difficult. Explainable Artificial Intelligence (XAI) has emerged as a crucial field addressing this challenge, particularly in ensuring transparency and trustworthiness in automated decision-making processes. In this article, we introduce Learning to eXplain Recommendations (LXR) , a scalable, model-agnostic framework designed to generate counterfactually correct explanations for recommender systems. LXR generates explanations for recommendations produced by any differentiable recommender system. By leveraging both factual and counterfactual loss terms, LXR offers robust, accurate, and computationally efficient explanations that reflect the model’s internal decision-making process. A key feature of LXR is its focus on the factual correctness of explanations through counterfactual reasoning, bridging the gap between plausible and accurate explanations. Unlike traditional approaches that rely on exhaustive perturbations of user data, LXR uses a self-supervised learning method to generate explanations efficiently, without sacrificing accuracy. LXR operates in two stages: a pre-training step and a novel Inference-Time Fine-tuning (ITF) step that refines explanations at the individual recommendation level, significantly improving accuracy with minimal computational overhead. Additionally, LXR is applied to hybrid recommender models incorporating demographic data, demonstrating its versatility across real-world scenarios. Finally, we also showcase LXR’s ability to explain recommendations at various ranks within a user’s recommendation list. As a secondary contribution, we introduce several novel evaluation metrics, inspired by saliency maps from computer vision, to rigorously assess the counterfactual correctness of explanations in recommender systems. Our results demonstrate that LXR sets a new benchmark for explainability, providing accurate, transparent, and interpretable explanations. The code is available on our GitHub repository: https://github.com/DeltaLabTLV/LXR_ .
- 22 November 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Recommender Systems
- Vol. 4 (2) , 1-39
Open AccessLXR: Learning to eXplain Recommendations
Recommender systems have become integral to many online services, leveraging user data to provide personalized recommendations. However, as these systems grow in complexity, understanding the rationale behind their recommendations becomes increasingly difficult. Explainable Artificial Intelligence (XAI) has emerged as a crucial field addressing this challenge, particularly in ensuring transparency and trustworthiness in automated decision-making processes. In this article, we introduce Learning to eXplain Recommendations (LXR) , a scalable, model-agnostic framework designed to generate counterfactually correct explanations for recommender systems. LXR generates explanations for recommendations produced by any differentiable recommender system. By leveraging both factual and counterfactual loss terms, LXR offers robust, accurate, and computationally efficient explanations that reflect the model’s internal decision-making process. A key feature of LXR is its focus on the factual correctness of explanations through counterfactual reasoning, bridging the gap between plausible and accurate explanations. Unlike traditional approaches that rely on exhaustive perturbations of user data, LXR uses a self-supervised learning method to generate explanations efficiently, without sacrificing accuracy. LXR operates in two stages: a pre-training step and a novel Inference-Time Fine-tuning (ITF) step that refines explanations at the individual recommendation level, significantly improving accuracy with minimal computational overhead. Additionally, LXR is applied to hybrid recommender models incorporating demographic data, demonstrating its versatility across real-world scenarios. Finally, we also showcase LXR’s ability to explain recommendations at various ranks within a user’s recommendation list. As a secondary contribution, we introduce several novel evaluation metrics, inspired by saliency maps from computer vision, to rigorously assess the counterfactual correctness of explanations in recommender systems. Our results demonstrate that LXR sets a new benchmark for explainability, providing accurate, transparent, and interpretable explanations. The code is available on our GitHub repository: https://github.com/DeltaLabTLV/LXR_ .
- Open Access
- 13 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 2967-2971
Refining Fidelity Metrics for Explainable Recommendations
- 13 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 2967-2971
Open AccessRefining Fidelity Metrics for Explainable Recommendations
- Open Access
- 21 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 98-108
A Learning-based Approach for Explaining Language Models
We present Learning Attributions (LA), a novel method for explaining language models. The core idea behind LA is to train a dedicated attribution model that functions as a surrogate explainer for the language model. This attribution model is designed to identify which tokens are most influential in driving the model's predictions. By optimizing the attribution model to mask the minimal amount of information necessary to induce substantial changes in the language model's output, LA provides a mechanism to understand which tokens in the input are critical for the model's decisions. We demonstrate the effectiveness of LA across several language models, highlighting its superiority over multiple state-of-the-art explanation methods across various datasets and evaluation metrics.
- 21 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 98-108
Open AccessA Learning-based Approach for Explaining Language Models
We present Learning Attributions (LA), a novel method for explaining language models. The core idea behind LA is to train a dedicated attribution model that functions as a surrogate explainer for the language model. This attribution model is designed to identify which tokens are most influential in driving the model's predictions. By optimizing the attribution model to mask the minimal amount of information necessary to induce substantial changes in the language model's output, LA provides a mechanism to understand which tokens in the input are critical for the model's decisions. We demonstrate the effectiveness of LA across several language models, highlighting its superiority over multiple state-of-the-art explanation methods across various datasets and evaluation metrics.
- Open Access
- 21 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 570-580
Probabilistic Path Integration with Mixture of Baseline Distributions
Path integration methods generate attributions by integrating along a trajectory from a baseline to the input. These techniques have demonstrated considerable effectiveness in the field of explainability research. While multiple types of baselines for the path integration process have been explored in the literature, there is no consensus on the ultimate one. This work examines the performance of different baseline distributions on explainability metrics and proposes a probabilistic path integration approach where the baseline distribution is modeled as a mixture of distributions, learned for each combination of model architecture and explanation metric. Extensive evaluations on various model architectures show that our method outperforms state-of-the-art explanation methods across multiple metrics.
- 21 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 570-580
Open AccessProbabilistic Path Integration with Mixture of Baseline Distributions
Path integration methods generate attributions by integrating along a trajectory from a baseline to the input. These techniques have demonstrated considerable effectiveness in the field of explainability research. While multiple types of baselines for the path integration process have been explored in the literature, there is no consensus on the ultimate one. This work examines the performance of different baseline distributions on explainability metrics and proposes a probabilistic path integration approach where the baseline distribution is modeled as a mixture of distributions, learned for each combination of model architecture and explanation metric. Extensive evaluations on various model architectures show that our method outperforms state-of-the-art explanation methods across multiple metrics.
- Open Access
- 2 August 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Recommender Systems
- Vol. 3 (1) , 1-23
Personalized Cadence Awareness for Next Basket Recommendation
This empirical study addresses the problem of Next Basket Repurchase Recommendation (NBRR), an often overlooked aspect of Next Basket Recommendation (NBR). While NBR aims to suggest items for a user’s next basket based on their prior basket history, NBRR focuses solely on recommending items previously purchased by the user. Despite the common ground between NBR and NBRR, the latter requires a distinct approach. In this paper, we survey recent developments in the fields of NBR and NBRR, emphasizing the different strategies employed for these closely related challenges. In addition, we review the common characteristics of users’ repurchase patterns, which characterize the NBRR problem. Building on these insights, we introduce a novel hyper-convolutional model tailored to capture behavioral patterns associated with repeated purchases. To evaluate its effectiveness, we conduct experiments on three publicly available datasets, offering a comprehensive analysis across three levels of granularity: user-level, order-level, and item-level. Our analysis illuminates the conditions under which the model excels and identifies scenarios where it may encounter challenges. This research contributes valuable insights into enhancing repurchase recommendation systems and advancing the understanding of user purchase behavior in general.
- 2 August 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Recommender Systems
- Vol. 3 (1) , 1-23
Open AccessPersonalized Cadence Awareness for Next Basket Recommendation
This empirical study addresses the problem of Next Basket Repurchase Recommendation (NBRR), an often overlooked aspect of Next Basket Recommendation (NBR). While NBR aims to suggest items for a user’s next basket based on their prior basket history, NBRR focuses solely on recommending items previously purchased by the user. Despite the common ground between NBR and NBRR, the latter requires a distinct approach. In this paper, we survey recent developments in the fields of NBR and NBRR, emphasizing the different strategies employed for these closely related challenges. In addition, we review the common characteristics of users’ repurchase patterns, which characterize the NBRR problem. Building on these insights, we introduce a novel hyper-convolutional model tailored to capture behavioral patterns associated with repeated purchases. To evaluate its effectiveness, we conduct experiments on three publicly available datasets, offering a comprehensive analysis across three levels of granularity: user-level, order-level, and item-level. Our analysis illuminates the conditions under which the model excels and identifies scenarios where it may encounter challenges. This research contributes valuable insights into enhancing repurchase recommendation systems and advancing the understanding of user purchase behavior in general.
- Open Access
- 13 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 3723-3733
A Counterfactual Framework for Learning and Evaluating Explanations for Recommender Systems
In the field of recommender systems, explainability remains a pivotal yet challenging aspect. To address this, we introduce the Learning to eXplain Recommendations (LXR) framework, a post-hoc, model-agnostic approach designed for providing counterfactual explanations. LXR is compatible with any differentiable recommender algorithm and scores the relevance of user data in relation to recommended items. A distinctive feature of LXR is its use of novel self-supervised counterfactual loss terms, which effectively highlight the most influential user data responsible for a specific recommended item. Additionally, we propose several innovative counterfactual evaluation metrics specifically tailored for assessing the quality of explanations in recommender systems. Our code is available on our GitHub repository: https://github.com/DeltaLabTLV/LXR.
- 13 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 3723-3733
Open AccessA Counterfactual Framework for Learning and Evaluating Explanations for Recommender Systems
In the field of recommender systems, explainability remains a pivotal yet challenging aspect. To address this, we introduce the Learning to eXplain Recommendations (LXR) framework, a post-hoc, model-agnostic approach designed for providing counterfactual explanations. LXR is compatible with any differentiable recommender algorithm and scores the relevance of user data in relation to recommended items. A distinctive feature of LXR is its use of novel self-supervised counterfactual loss terms, which effectively highlight the most influential user data responsible for a specific recommended item. Additionally, we propose several innovative counterfactual evaluation metrics specifically tailored for assessing the quality of explanations in recommender systems. Our code is available on our GitHub repository: https://github.com/DeltaLabTLV/LXR.
- Open Access
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 5246-5247
Harnessing GPT for Topic-Based Call Segmentation in Microsoft Dynamics 365 Sales
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 5246-5247
Open AccessHarnessing GPT for Topic-Based Call Segmentation in Microsoft Dynamics 365 Sales
- Open Access
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 57-67
Deep Integrated Explanations
This paper presents Deep Integrated Explanations (DIX) - a universal method for explaining vision models. DIX generates explanation maps by integrating information from the intermediate representations of the model, coupled with their corresponding gradients. Through an extensive array of both objective and subjective evaluations spanning diverse tasks, datasets, and model configurations, we showcase the efficacy of DIX in generating faithful and accurate explanation maps, while surpassing current state-of-the-art methods. Our code is available at: https://github.com/dix-cikm23/dix
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 57-67
Open AccessDeep Integrated Explanations
This paper presents Deep Integrated Explanations (DIX) - a universal method for explaining vision models. DIX generates explanation maps by integrating information from the intermediate representations of the model, coupled with their corresponding gradients. Through an extensive array of both objective and subjective evaluations spanning diverse tasks, datasets, and model configurations, we showcase the efficacy of DIX in generating faithful and accurate explanation maps, while surpassing current state-of-the-art methods. Our code is available at: https://github.com/dix-cikm23/dix
- Open Access
- 3 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 34746-34763
Not All Memories Created Equal: Dynamic User Representations for Collaborative Filtering
Collaborative filtering methods for recommender systems tend to represent users as a single static latent vector. However, user behavior and interests may dynamically change in the context of the recommended item being presented to the user. For example, in the case of movie recommendations, it is usually true that movies that the user watched more recently are more informative than movies that were watched a long time ago. However, it is possible that a particular movie from the past may become suddenly more relevant for prediction in the presence of a recommendation for its sequel movie. In response to this issue, we introduce the Attentive Item2Vec++ (AI2V++) model, a neural attentive collaborative filtering approach in which the user representation adapts dynamically in the presence of the recommended item. AI2V++ employs a novel context-target attention mechanism in order to learn and capture different characteristics of the user’s historical behavior with respect to a potential recommended item. Furthermore, analysis of the neural-attentive scores allows for improved interpretability and explainability of the model. We evaluate our proposed approach on five publicly available datasets and demonstrate its superior performance in comparison to state-of-the-art baselines across multiple accuracy metrics.
- 3 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 34746-34763
Open AccessNot All Memories Created Equal: Dynamic User Representations for Collaborative Filtering
Collaborative filtering methods for recommender systems tend to represent users as a single static latent vector. However, user behavior and interests may dynamically change in the context of the recommended item being presented to the user. For example, in the case of movie recommendations, it is usually true that movies that the user watched more recently are more informative than movies that were watched a long time ago. However, it is possible that a particular movie from the past may become suddenly more relevant for prediction in the presence of a recommendation for its sequel movie. In response to this issue, we introduce the Attentive Item2Vec++ (AI2V++) model, a neural attentive collaborative filtering approach in which the user representation adapts dynamically in the presence of the recommended item. AI2V++ employs a novel context-target attention mechanism in order to learn and capture different characteristics of the user’s historical behavior with respect to a potential recommended item. Furthermore, analysis of the neural-attentive scores allows for improved interpretability and explainability of the model. We evaluate our proposed approach on five publicly available datasets and demonstrate its superior performance in comparison to state-of-the-art baselines across multiple accuracy metrics.
- Open Access
GAM: Explainable Visual Similarity and Classification via Gradient Activation Maps
We present Gradient Activation Maps (GAM) - a machinery for explaining predictions made by visual similarity and classification models. By gleaning localized gradient and activation information from multiple network layers, GAM offers improved visual explanations, when compared to existing alternatives. The algorithmic advantages of GAM are explained in detail, and validated empirically, where it is shown that GAM outperforms its alternatives across various tasks and datasets.
Open AccessGAM: Explainable Visual Similarity and Classification via Gradient Activation Maps
We present Gradient Activation Maps (GAM) - a machinery for explaining predictions made by visual similarity and classification models. By gleaning localized gradient and activation information from multiple network layers, GAM offers improved visual explanations, when compared to existing alternatives. The algorithmic advantages of GAM are explained in detail, and validated empirically, where it is shown that GAM outperforms its alternatives across various tasks and datasets.