paper-with-me

Papers

GLUS: Global-Local Reasoning Unified into A Single Large Language Model for Video Segmentation

2025-04-10 · CVPR 2025 1 · Lang Lin, Xueyang Yu, Ziqi Pang, Yu-Xiong Wang

This paper proposes a novel framework utilizing multi-modal large language models (MLLMs) for referring video object segmentation (RefVOS). Previous MLLM-based methods commonly struggle with the dilemma between "Ref" and "VOS": they either specialize in understanding a few key frames (global reasoning) or tracking objects on continuous frames (local reasoning), and rely on external VOS or frame selectors to mitigate the other end of the challenge. However, our framework GLUS shows that global and local consistency can be unified into a single video segmentation MLLM: a set of sparse "context frames" provides global information, while a stream of continuous "query frames" conducts local object tracking. This is further supported by jointly training the MLLM with a pre-trained VOS memory bank to simultaneously digest short-range and long-range temporal information. To improve the information efficiency within the limited context window of MLLMs, we introduce object contrastive learning to distinguish hard false-positive objects and a self-refined framework to identify crucial frames and perform propagation. By collectively integrating these insights, our GLUS delivers a simple yet effective baseline, achieving new state-of-the-art for MLLMs on the MeViS and Ref-Youtube-VOS benchmark. Our project page is at https://glus-video.github.io/.

📄 PDF Abstract BibTeX arXiv:2504.07962

Code (1)

GLUS-video/GLUS pytorch

Tasks

Contrastive LearningLanguage ModelingLanguage ModellingLarge Language ModelObject TrackingReferring Video Object SegmentationSemantic SegmentationVideo Object SegmentationVideo SegmentationVideo Semantic Segmentation

Methods 이 논문이 사용한 방법론

VOS VOS is a type of video object segmentation model consisting of two network components. The target appearance model consists of a light-weight module, which is learned during…
Contrastive Learning 설명 없음
SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

GLUSE: Enhanced Channel-Wise Adaptive Gated Linear Units SE for Onboard Satellite Earth Observation Image Classification

2025-04-16 · Thanh-Dung Le, Vu Nguyen Ha, Ti Ti Nguyen, Geoffrey Eappen 외

This study introduces ResNet-GLUSE, a lightweight ResNet variant enhanced with Gated Linear Unit-enhanced Squeeze-and-Excitation (GLUSE), an adaptive channel-wise attention mechanism. By integrating dynamic gating into t…

Computational EfficiencyEarth Observationimage-classificationImage Classification

Sparsity Moves Computation: How FFN Architecture Reshapes Attention in Small Transformers

2026-05-10 · Gabriel Smithline, Chris Mascioli arxiv

Architectural choices inside the Transformer feedforward network (FFN) block do not merely affect the block itself; they reshape the computations learned by the rest of the model. We study this effect in one-layer Transf…

GLUScope: A Tool for Analyzing GLU Neurons in Transformer Language Models

2026-02-27 · Sebastian Gerstner, Hinrich Schütze arxiv

We present GLUScope, an open-source tool for analyzing neurons in Transformer-based language models, intended for interpretability researchers. We focus on more recent models than previous tools do; specifically we consi…

Weight-based Decomposition: A Case for Bilinear MLPs

2024-06-06 · Michael T. Pearce, Thomas Dooms, Alice Rigg

Gated Linear Units (GLUs) have become a common building block in modern foundation models. Bilinear layers drop the non-linearity in the "gate" but still have comparable performance to other GLUs. An attractive quality o…

Divine Benevolence is an $x^2$: GLUs scale asymptotically faster than MLPs

2026-02-16 · Alejandro Francisco Queiruga arxiv

Scaling laws can be understood from ground-up numerical analysis, where traditional function approximation theory can explain shifts in model architecture choices. GLU variants now dominate frontier LLMs and similar oute…