# TRUST: Item-Calibrated Interval Evidence for Temporal Session-Based Recommendation

Source: [TRUST: Item-Calibrated Interval Evidence for Temporal Session-Based Recommendation](<https://www.sota2.com/research/paper/6a3e07d8206542251d7cb7d3>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-27T05:00:11.922Z

Authors: Linjiang Guo, Nitin Bisht, Shiqing Wu, Yifan Yin, Guandong Xu

Published: 2026-06-25T16:05:36.000Z

## Abstract

Temporal signals have been widely used in session-based recommendation to infer user interest. Existing temporal session-based recommenders primarily rely on absolute interval values, implicitly assuming that the same interval carries similar interest signals across items. However, we empirically find that this assumption does not hold: each item has its own interval distribution, so an interval should be interpreted relative to the item it belongs to. Based on this observation, we propose TRUST, a framework that evaluates each observed interval relative to the empirical interval distribution of the corresponding item. Specifically, we propose a score function to guide global neighbor sampling, session graph encoding, and final interest aggregation. Experiments on public datasets show that TRUST consistently improves over representative temporal and non-temporal baselines, and plug-in experiments further show that the proposed scoring function can improve existing temporal session recommenders as a model-agnostic method. Component-wise ablations further show that calibrating the temporal signals within each module, rather than removing the module itself, consistently improves neighbor sampling, session graph encoding, and interest aggregation.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Session-based Recommendation on RetailRocket (HR/MRR@5, @10)](<https://www.sota2.com/research/sota/session-based-recommendation-on-retailrocket-hr-mrr-at-5-at-10.md>): Session-based recommendation; RetailRocket; HR@5: 52.3.

- [Session-based Recommendation on Nowplaying](<https://www.sota2.com/research/sota/session-based-recommendation-on-nowplaying.md>): Session-based recommendation; Nowplaying; HR@5: 12.68.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2606.27214>)

[PDF](<https://arxiv.org/pdf/2606.27214.pdf>)
