EVIDENT: Routing MLLM Adaptation through Entity-Grounded Visual Evidence for Cross-Domain Video Temporal Grounding

• URL: http://arxiv.org/abs/2605.26104v1
• Date: Mon, 25 May 2026 17:58:11 GMT
• Title: EVIDENT: Routing MLLM Adaptation through Entity-Grounded Visual Evidence for Cross-Domain Video Temporal Grounding
• Abstract summary: Fine-tuning MLLMs for Video Temporal Grounding (VTG) often improves in-domain performance but degrades sharply under domain shift.<n>We introduce EVIDENT, a framework that anchors temporal grounding in the inherent entity-attention of pre-trained MLLMs.
• Score: 19.421119820746913
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Fine-tuning MLLMs for Video Temporal Grounding (VTG) often improves in-domain performance but degrades sharply under domain shift. In this work, we find that this failure is primarily driven not just by unseen query concepts, but by visual domain shift, which prevents the model from coupling its learned temporal localization knowledge with its inherent entity-attention capability. To address this, we introduce EVIDENT, a parameter-efficient adaptation framework that anchors temporal grounding in the inherent entity-attention of pre-trained MLLMs by routing VTG adaptation through explicit visual entity evidence. EVIDENT consists of three components: (i) an Entity Bottleneck Adapter that transforms dense visual tokens into compact entity-level slots, (ii) an Entity-Binding Distillation loss that instills objectness priors into the semantically unstructured MLLM visual space, guiding each slot to bind to a coherent entity, and (iii) an Entity-to-eVidence gating mechanism that leverages the captured entities as evidence, steering the model to localize moments containing query-relevant entities. Together, these components enable VTG fine-tuning to rely on entity-grounded evidence rather than brittle dataset shortcuts. Experiments on cross-domain VTG benchmarks show that EVIDENT consistently improves out-of-domain robustness while preserving competitive in-domain performance with modest parameter overhead. These results suggest that entity-level grounding is an effective inductive bias for generalizable temporal localization.

Related papers

• VFM$^{4}$SDG: Unveiling the Power of VFMs for Single-Domain Generalized Object Detection [33.39250067795076]
  In real-world scenarios, continual changes in weather, illumination, and imaging conditions cause significant domain shifts.<n>We propose a dual-prior learning framework for generalized object detection (SDGOD)<n>In the encoding stage, we propose Cross-domain Stable Prior Distillation to enhance the robustness of object-background and inter-instance modeling.<n>In the decoding stage, we propose Semantic-Contextual Prior-based Query Enhancement.
  arXiv  Detail & Related papers  (2026-04-23T10:04:36Z)
• SlotVTG: Object-Centric Adapter for Generalizable Video Temporal Grounding [5.9095311593289575]
  We propose SlotVTG, a framework that steers MLLMs toward object-centric, input-grounded visual reasoning at minimal cost.<n>Our approach significantly improves OOD robustness while maintaining competitive In-Domain (ID) performance with minimal overhead.
  arXiv  Detail & Related papers  (2026-03-26T17:59:31Z)
• MEMTS: Internalizing Domain Knowledge via Parameterized Memory for Retrieval-Free Domain Adaptation of Time Series Foundation Models [51.506429027626005]
  Memory for Time Series (MEMTS) is a lightweight and plug-and-play method for retrieval-free domain adaptation in time series forecasting.<n>Key component of MEMTS is a Knowledge Persistence Module (KPM), which internalizes domain-specific temporal dynamics.<n>This paradigm shift enables MEMTS to achieve accurate domain adaptation with constant-time inference and near-zero latency.
  arXiv  Detail & Related papers  (2026-02-14T14:00:06Z)
• RB-FT: Rationale-Bootstrapped Fine-Tuning for Video Classification [14.224783616912783]
  Vision Language Models (VLMs) are becoming increasingly integral to multimedia understanding.<n>They often struggle with domain-specific video classification tasks, particularly with limited data.<n>We propose two-stage self-improvement paradigm to bridge this gap without new annotations.
  arXiv  Detail & Related papers  (2025-11-19T23:12:18Z)
• Style-Adaptive Detection Transformer for Single-Source Domain Generalized Object Detection [7.768332621617199]
  Single-source domain generalization aims to develop a detector using only source domain data that generalizes well to unseen target domains.<n>Existing methods are primarily CNN-based and improve robustness through data augmentation combined with feature alignment.<n>We propose Style-Adaptive DEtection TRansformer (SA-DETR), a DETR-based detector tailored for single-source domain generalization.
  arXiv  Detail & Related papers  (2025-04-29T07:38:37Z)
• RT-DATR: Real-time Unsupervised Domain Adaptive Detection Transformer with Adversarial Feature Alignment [6.951423778505503]
  Domain-adaptive object detectors based on CNN and transformers have made significant progress in cross-domain detection tasks.<n>We propose RT-DATR, a simple and efficient real-time domain adaptive detection transformer.<n> Experimental results on various cross-domian benchmarks demonstrate that our method outperforms current state-of-the-art approaches.
  arXiv  Detail & Related papers  (2025-04-12T12:36:01Z)
• Exploiting Aggregation and Segregation of Representations for Domain Adaptive Human Pose Estimation [50.31351006532924]
  Human pose estimation (HPE) has received increasing attention recently due to its wide application in motion analysis, virtual reality, healthcare, etc.<n>It suffers from the lack of labeled diverse real-world datasets due to the time- and labor-intensive annotation.<n>We introduce a novel framework that capitalizes on both representation aggregation and segregation for domain adaptive human pose estimation.
  arXiv  Detail & Related papers  (2024-12-29T17:59:45Z)
• Unsupervised Video Domain Adaptation for Action Recognition: A Disentanglement Perspective [37.45565756522847]
  We consider the generation of cross-domain videos from two sets of latent factors. TranSVAE framework is then developed to model such generation. Experiments on the UCF-HMDB, Jester, and Epic-Kitchens datasets verify the effectiveness and superiority of TranSVAE.
  arXiv  Detail & Related papers  (2022-08-15T17:59:31Z)
• Domain-Agnostic Prior for Transfer Semantic Segmentation [197.9378107222422]
  Unsupervised domain adaptation (UDA) is an important topic in the computer vision community. We present a mechanism that regularizes cross-domain representation learning with a domain-agnostic prior (DAP) Our research reveals that UDA benefits much from better proxies, possibly from other data modalities.
  arXiv  Detail & Related papers  (2022-04-06T09:13:25Z)
• Decompose to Adapt: Cross-domain Object Detection via Feature Disentanglement [79.2994130944482]
  We design a Domain Disentanglement Faster-RCNN (DDF) to eliminate the source-specific information in the features for detection task learning. Our DDF method facilitates the feature disentanglement at the global and local stages, with a Global Triplet Disentanglement (GTD) module and an Instance Similarity Disentanglement (ISD) module. By outperforming state-of-the-art methods on four benchmark UDA object detection tasks, our DDF method is demonstrated to be effective with wide applicability.
  arXiv  Detail & Related papers  (2022-01-06T05:43:01Z)
• Unsupervised Domain Adaptation for Spatio-Temporal Action Localization [69.12982544509427]
  S-temporal action localization is an important problem in computer vision. We propose an end-to-end unsupervised domain adaptation algorithm. We show that significant performance gain can be achieved when spatial and temporal features are adapted separately or jointly.
  arXiv  Detail & Related papers  (2020-10-19T04:25:10Z)
• Domain Adversarial Fine-Tuning as an Effective Regularizer [80.14528207465412]
  In Natural Language Processing (NLP), pretrained language models (LMs) that are transferred to downstream tasks have been recently shown to achieve state-of-the-art results. Standard fine-tuning can degrade the general-domain representations captured during pretraining. We introduce a new regularization technique, AFTER; domain Adversarial Fine-Tuning as an Effective Regularizer.
  arXiv  Detail & Related papers  (2020-09-28T14:35:06Z)

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.