From Click Modeling to Offline and Off-Policy Evaluation in Carousel Recommendation
- URL: http://arxiv.org/abs/2608.22022v1
- Date: Sat, 22 Aug 2026 15:56:14 GMT
- Title: From Click Modeling to Offline and Off-Policy Evaluation in Carousel Recommendation
- Abstract summary: My PhD research aims to rethink how carousel clicks are modeled and how carousel recommendation policies can be evaluated.<n>My ongoing work includes a project using discrete choice models to represent clicks as choices, alongside a project that develops carousel-specific offline metrics.<n>The expected contribution of my thesis is a connected body of work that links carousel click modeling with offline and off-policy evaluation.
- Score: 1.6655489215063515
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Carousel interfaces are widely used in modern recommendation systems. Unlike traditional interfaces that present a single ranked list, carousels simultaneously present several ranked lists to the user, as horizontally swipeable rows stacked on top of each other. In this design, the rankings are closely tied to the two-dimensional layout. Consequently, user behavior is shaped not only by item preference, but also by row organization, viewport constraints, and item context. This tight coupling between ranking and presentation complicates the interpretation of user feedback, introducing new challenges for recommendation evaluation. My PhD research aims to address these challenges by rethinking how carousel clicks are modeled and how carousel recommendation policies can be evaluated from logged interaction data. So far, I have studied how users interact with carousel interfaces and developed a click model design framework that prioritizes mathematical relationships between observed variables over latent behavioral assumptions. Building on these results, my ongoing work includes a project using discrete choice models to represent clicks as choices, alongside a project that develops carousel-specific offline metrics. As a next step, I plan to develop off-policy evaluation methods that estimate the performance of recommendation policies from logged interactions. Taken together, the expected contribution of my thesis is a connected body of work that links carousel click modeling with offline and off-policy evaluation, so that carousel recommendation policies can be improved more reliably.
Related papers
- Revisiting N2DCG: An Empirically Grounded Reformulation of Carousel Recommendation Evaluation [56.85622784396052]
We propose a reformulation of N2DCG that normalizes appropriately by respecting constraints and uses an empirically grounded discount function.<n>We validate the proposed metric, showing that it better reflects users' empirical behavior on real-world eye-tracking data.
arXiv Detail & Related papers (2026-08-22T09:40:10Z) - Following the Eye-Tracking Evidence: Established Web-Search Assumptions Fail in Carousel Interfaces [60.155970399741214]
Carousel interfaces have been the de-facto standard for streaming media services for over a decade.<n>Previous work has assumed that behaviors established in single-list web-search interfaces, such as the F-pattern, also apply to carousel interfaces.<n>Our work shows that the field lacks a reliable foundation on which to build models of user behavior with these interfaces.
arXiv Detail & Related papers (2026-04-22T19:07:52Z) - From Latent to Observable Position-Based Click Models in Carousel Interfaces [3.441021278275805]
Click models are a central component of learning and evaluation in recommender systems.<n>Modern recommender platforms increasingly use complex interfaces such as carousels.<n>We study position-based click models in carousel interfaces and examine optimization methods.
arXiv Detail & Related papers (2026-02-18T15:40:30Z) - RecGaze: The First Eye Tracking and User Interaction Dataset for Carousel Interfaces [58.695627866883065]
We introduce the RecGaze dataset: the first comprehensive feedback dataset on carousels.<n>The dataset comprises of interactions from 3 movie selection tasks with 40 different carousel interfaces per user.<n>We provide results of a survey on carousel design and the first analysis of gaze data on carousels, which reveals a golden triangle or F-pattern browsing behavior.
arXiv Detail & Related papers (2025-04-29T14:09:20Z) - Multi-Behavior Sequential Recommendation with Temporal Graph Transformer [66.10169268762014]
We tackle the dynamic user-item relation learning with the awareness of multi-behavior interactive patterns.
We propose a new Temporal Graph Transformer (TGT) recommendation framework to jointly capture dynamic short-term and long-range user-item interactive patterns.
arXiv Detail & Related papers (2022-06-06T15:42:54Z) - Learning Dual Dynamic Representations on Time-Sliced User-Item
Interaction Graphs for Sequential Recommendation [62.30552176649873]
We devise a novel Dynamic Representation Learning model for Sequential Recommendation (DRL-SRe)
To better model the user-item interactions for characterizing the dynamics from both sides, the proposed model builds a global user-item interaction graph for each time slice.
To enable the model to capture fine-grained temporal information, we propose an auxiliary temporal prediction task over consecutive time slices.
arXiv Detail & Related papers (2021-09-24T07:44:27Z) - Measuring the User Satisfaction in a Recommendation Interface with
Multiple Carousels [7.8851236034886645]
It is common for video-on-demand and music streaming services to adopt a user interface composed of several recommendation lists.
We propose a two-dimensional evaluation protocol for a carousel setting that will measure the quality of a recommendation carousel.
arXiv Detail & Related papers (2021-05-14T20:33:51Z) - A Methodology for the Offline Evaluation of Recommender Systems in a
User Interface with Multiple Carousels [7.8851236034886645]
Video-on-demand and music streaming services provide the user with a page consisting of several recommendation lists.
Finding efficient strategies to select which carousels to display is an active research topic of great industrial interest.
We propose an offline evaluation protocol for a carousel setting in which the recommendation quality of a model is measured by how much it improves upon that of an already available set of carousels.
arXiv Detail & Related papers (2021-05-13T13:14:59Z) - Dynamic Graph Collaborative Filtering [64.87765663208927]
Dynamic recommendation is essential for recommender systems to provide real-time predictions based on sequential data.
Here we propose Dynamic Graph Collaborative Filtering (DGCF), a novel framework leveraging dynamic graphs to capture collaborative and sequential relations.
Our approach achieves higher performance when the dataset contains less action repetition, indicating the effectiveness of integrating dynamic collaborative information.
arXiv Detail & Related papers (2021-01-08T04:16:24Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.