SPARROW: Learning Spatial Precision and Temporal Referential Consistency in Pixel-Grounded Video MLLMs

• URL: http://arxiv.org/abs/2603.12382v1
• Date: Thu, 12 Mar 2026 18:59:57 GMT
• Title: SPARROW: Learning Spatial Precision and Temporal Referential Consistency in Pixel-Grounded Video MLLMs
• Abstract summary: We introduce SPARROW, a pixel-grounded video MLLM that unifies spatial accuracy and temporal stability through two key components.<n> SPARROW is supported by a curated referential video dataset of 30,646 videos and 45,231 Q&A pairs.<n>It delivers consistent gains across six benchmarks, improving up to +8.9 J&F on RVOS, +5 mIoU on visual grounding, and +5.4 CLAIR on GCG.
• Score: 39.73103140338364
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multimodal large language models (MLLMs) have advanced from image-level reasoning to pixel-level grounding, but extending these capabilities to videos remains challenging as models must achieve spatial precision and temporally consistent reference tracking. Existing video MLLMs often rely on a static segmentation token ([SEG]) for frame-wise grounding, which provides semantics but lacks temporal context, causing spatial drift, identity switches, and unstable initialization when objects move or reappear. We introduce SPARROW, a pixel-grounded video MLLM that unifies spatial accuracy and temporal stability through two key components: (i) Target-Specific Tracked Features (TSF), which inject temporally aligned referent cues during training, and (ii) a dual-prompt design that decodes box ([BOX]) and segmentation ([SEG]) tokens to fuse geometric priors with semantic grounding. SPARROW is supported by a curated referential video dataset of 30,646 videos and 45,231 Q&A pairs and operates end-to-end without external detectors via a class-agnostic SAM2-based proposer. Integrated into three recent open-source video MLLMs (UniPixel, GLUS, and VideoGLaMM), SPARROW delivers consistent gains across six benchmarks, improving up to +8.9 J&F on RVOS, +5 mIoU on visual grounding, and +5.4 CLAIR on GCG. These results demonstrate that SPARROW substantially improves referential stability, spatial precision, and temporal coherence in pixel-grounded video understanding. Project page: https://risys-lab.github.io/SPARROW

Related papers

• PhysMLLMs: Spatial Priors for Unified Referring Segmentation and Grounded Reasoning of Images and Videos [116.62982600141608]
  Video large language models support language guided video segmentation, but they often show temporal inconsistencies.<n>We present PhysMLLMs, a training-stage prior injection architecture that injects physics-inspired spatial continuity priors into Video MLLMs.<n>Across multiple video benchmarks, PhysMLLMs improves video segmentation mask quality and cross-frame consistency.
  arXiv  Detail & Related papers  (2026-08-25T14:01:59Z)
• GroundVTS: Visual Token Sampling in Multimodal Large Language Models for Video Temporal Grounding [8.196520511153368]
  Video temporal grounding (VTG) is a critical task in video and understanding a key capability for extending large language models (Vid-LLMs) to broader applications.<n>Existing Vid-LLMs rely on uniform frame sampling to extract video information, resulting in a sparse distribution of key frames and the loss of crucial temporal cues.<n>We propose Grounded Visual Token Sampling (GroundVTS), a Vid-LLM architecture that focuses on the most informative temporal segments.
  arXiv  Detail & Related papers  (2026-04-02T14:19:58Z)
• 1 + 1 > 2: Detector-Empowered Video Large Language Model for Spatio-Temporal Grounding and Reasoning [53.28271278708241]
  We present a Detector-Empowered Video LLM, short for DEViL.<n> DEViL couples a Video LLM with an open-vocabulary detector (OVD)<n>Unlike tokens that merely serve as spatial prompts or segmentor switches, the RST functions as both a control signal and a replacement for the OVD's text embedding.
  arXiv  Detail & Related papers  (2025-12-07T06:11:15Z)
• Unleashing the Potential of Multimodal LLMs for Zero-Shot Spatio-Temporal Video Grounding [47.400649582392255]
  We use large language models (MLLMs) to explore a zero-shot solution in STVG.<n>We propose a MLLM-based zero-shot framework for STVG, which includes novel temporal-augmented assembling strategies.
  arXiv  Detail & Related papers  (2025-09-18T17:35:50Z)
• VideoMolmo: Spatio-Temporal Grounding Meets Pointing [66.19964563104385]
  VideoMolmo is a model tailored for fine-grained pointing of video sequences.<n>A novel temporal mask fusion employs SAM2 for bidirectional point propagation.<n>To evaluate the generalization of VideoMolmo, we introduce VPoMolS-temporal, a challenging out-of-distribution benchmark spanning five real-world scenarios.
  arXiv  Detail & Related papers  (2025-06-05T17:59:29Z)
• Spatial-MLLM: Boosting MLLM Capabilities in Visual-based Spatial Intelligence [13.168559963356952]
  We present Spatial-MLLM, a novel framework for visual-based spatial reasoning from purely 2D observations.<n>Our key insight is to unleash the strong structure prior to the feed-forward visual geometry foundation model.<n>A connector then integrates both features into unified visual tokens for enhanced spatial understanding.
  arXiv  Detail & Related papers  (2025-05-29T17:59:04Z)
• ReferDINO: Referring Video Object Segmentation with Visual Grounding Foundations [33.74746234704817]
  Video object segmentation (RVOS) aims to segment target objects throughout a video based on a text description.<n>This is challenging as it involves deep vision-level understanding, pixel-level dense prediction andtemporal reasoning.<n>We propose bfReferDINO RVOS that inherits region-level vision-text alignment from foundational visual grounding models.
  arXiv  Detail & Related papers  (2025-01-24T16:24:15Z)
• VideoGLaMM: A Large Multimodal Model for Pixel-Level Visual Grounding in Videos [58.765796160750504]
  VideoGLaMM is a new model for fine-grained pixel-level grounding in videos based on user-provided textual inputs.<n>The architecture is trained to synchronize both spatial and temporal elements of video content with textual instructions.<n> Experimental results show that our model consistently outperforms existing approaches across all three tasks.
  arXiv  Detail & Related papers  (2024-11-07T17:59:27Z)
• SIGMA:Sinkhorn-Guided Masked Video Modeling [69.31715194419091]
  Sinkhorn-guided Masked Video Modelling ( SIGMA) is a novel video pretraining method. We distribute features of space-time tubes evenly across a limited number of learnable clusters. Experimental results on ten datasets validate the effectiveness of SIGMA in learning more performant, temporally-aware, and robust video representations.
  arXiv  Detail & Related papers  (2024-07-22T08:04:09Z)

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.