Adaptive Evidential Learning for Temporal-Semantic Robustness in Moment Retrieval

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Huang, Haojian, Ma, Kaijing, Chen, Jin, Chen, Haodong, Wu, Zhou, Zang, Xianghao, Fang, Han, Ban, Chao, Sun, Hao, Chen, Mulin, He, Zhongjiang
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908683596726272
author Huang, Haojian
Ma, Kaijing
Chen, Jin
Chen, Haodong
Wu, Zhou
Zang, Xianghao
Fang, Han
Ban, Chao
Sun, Hao
Chen, Mulin
He, Zhongjiang
author_facet Huang, Haojian
Ma, Kaijing
Chen, Jin
Chen, Haodong
Wu, Zhou
Zang, Xianghao
Fang, Han
Ban, Chao
Sun, Hao
Chen, Mulin
He, Zhongjiang
contents In the domain of moment retrieval, accurately identifying temporal segments within videos based on natural language queries remains challenging. Traditional methods often employ pre-trained models that struggle with fine-grained information and deterministic reasoning, leading to difficulties in aligning with complex or ambiguous moments. To overcome these limitations, we explore Deep Evidential Regression (DER) to construct a vanilla Evidential baseline. However, this approach encounters two major issues: the inability to effectively handle modality imbalance and the structural differences in DER's heuristic uncertainty regularizer, which adversely affect uncertainty estimation. This misalignment results in high uncertainty being incorrectly associated with accurate samples rather than challenging ones. Our observations indicate that existing methods lack the adaptability required for complex video scenarios. In response, we propose Debiased Evidential Learning for Moment Retrieval (DEMR), a novel framework that incorporates a Reflective Flipped Fusion (RFF) block for cross-modal alignment and a query reconstruction task to enhance text sensitivity, thereby reducing bias in uncertainty estimation. Additionally, we introduce a Geom-regularizer to refine uncertainty predictions, enabling adaptive alignment with difficult moments and improving retrieval accuracy. Extensive testing on standard datasets and debiased datasets ActivityNet-CD and Charades-CD demonstrates significant enhancements in effectiveness, robustness, and interpretability, positioning our approach as a promising solution for temporal-semantic robustness in moment retrieval. The code is publicly available at https://github.com/KaijingOfficial/DEMR.
format Preprint
id arxiv_https___arxiv_org_abs_2512_00953
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Adaptive Evidential Learning for Temporal-Semantic Robustness in Moment Retrieval
Huang, Haojian
Ma, Kaijing
Chen, Jin
Chen, Haodong
Wu, Zhou
Zang, Xianghao
Fang, Han
Ban, Chao
Sun, Hao
Chen, Mulin
He, Zhongjiang
Computer Vision and Pattern Recognition
In the domain of moment retrieval, accurately identifying temporal segments within videos based on natural language queries remains challenging. Traditional methods often employ pre-trained models that struggle with fine-grained information and deterministic reasoning, leading to difficulties in aligning with complex or ambiguous moments. To overcome these limitations, we explore Deep Evidential Regression (DER) to construct a vanilla Evidential baseline. However, this approach encounters two major issues: the inability to effectively handle modality imbalance and the structural differences in DER's heuristic uncertainty regularizer, which adversely affect uncertainty estimation. This misalignment results in high uncertainty being incorrectly associated with accurate samples rather than challenging ones. Our observations indicate that existing methods lack the adaptability required for complex video scenarios. In response, we propose Debiased Evidential Learning for Moment Retrieval (DEMR), a novel framework that incorporates a Reflective Flipped Fusion (RFF) block for cross-modal alignment and a query reconstruction task to enhance text sensitivity, thereby reducing bias in uncertainty estimation. Additionally, we introduce a Geom-regularizer to refine uncertainty predictions, enabling adaptive alignment with difficult moments and improving retrieval accuracy. Extensive testing on standard datasets and debiased datasets ActivityNet-CD and Charades-CD demonstrates significant enhancements in effectiveness, robustness, and interpretability, positioning our approach as a promising solution for temporal-semantic robustness in moment retrieval. The code is publicly available at https://github.com/KaijingOfficial/DEMR.
title Adaptive Evidential Learning for Temporal-Semantic Robustness in Moment Retrieval
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.00953