Cover Image

Does Acceleration Cause Hidden Instability in Vision Language Models? Uncovering Instance-Level Divergence Through a Large-Scale Empirical Study

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Sun, Yizheng, Li, Hao, Xu, Chang, Zhou, Hongpeng, Lin, Chenghua, Batista-Navarro, Riza, Sun, Jingyuan
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

LVPruning: An Effective yet Simple Language-Guided Vision Token Pruning Approach for Multi-modal Large Language Models
by: Sun, Yizheng, et al.
Published: (2025)

Large Language Models in Argument Mining: A Survey
by: Li, Hao, et al.
Published: (2025)

Arg-LLaDA: Argument Summarization via Large Language Diffusion Models and Sufficiency-Aware Refinement
by: Li, Hao, et al.
Published: (2025)

Aspect-based Sentiment Evaluation of Chess Moves (ASSESS): an NLP-based Method for Evaluating Chess Strategies from Textbooks
by: Alrdahi, Haifa, et al.
Published: (2024)

Large-Scale Terminal Agentic Trajectory Generation from Dockerized Environments
by: Wu, Siwei, et al.
Published: (2026)

Natural Context Drift Undermines the Natural Language Understanding of Large Language Models
by: Wu, Yulong, et al.
Published: (2025)

LongEval: A Comprehensive Analysis of Long-Text Generation Through a Plan-based Paradigm
by: Wu, Siwei, et al.
Published: (2025)

Investigating a Benchmark for Training-set free Evaluation of Linguistic Capabilities in Machine Reading Comprehension
by: Schlegel, Viktor, et al.
Published: (2024)

Induced Numerical Instability: Hidden Costs in Multimodal Large Language Models
by: Wong, Wai Tuck, et al.
Published: (2026)

Computational Lesions in Multilingual Language Models Separate Shared and Language-specific Brain Alignment
by: Cui, Yang, et al.
Published: (2026)

Learning to Generate and Evaluate Fact-checking Explanations with Transformers
by: Feher, Darius, et al.
Published: (2024)

Pay Attention to Real World Perturbations! Natural Robustness Evaluation in Machine Reading Comprehension
by: Wu, Yulong, et al.
Published: (2025)

Hidden States Know Where Reasoning Diverges: Credit Assignment via Span-Level Wasserstein Distance
by: Chen, Xinzhu, et al.
Published: (2026)

Natural Language Satisfiability: Exploring the Problem Distribution and Evaluating Transformer-based Language Models
by: Madusanka, Tharindu, et al.
Published: (2025)

ADAS-TO: A Large-Scale Multimodal Naturalistic Dataset and Empirical Characterization of Human Takeovers during ADAS Engagement
by: Wang, Yuhang, et al.
Published: (2026)

Uncovering Bias in Large Vision-Language Models at Scale with Counterfactuals
by: Howard, Phillip, et al.
Published: (2024)

Exploring Causes and Mitigation of Hallucinations in Large Vision Language Models
by: Sun, Yaqi, et al.
Published: (2025)

CE-CoLLM: Efficient and Adaptive Large Language Models Through Cloud-Edge Collaboration
by: Jin, Hongpeng, et al.
Published: (2024)

UIS-Mamba: Exploring Mamba for Underwater Instance Segmentation via Dynamic Tree Scan and Hidden State Weaken
by: Cong, Runmin, et al.
Published: (2025)

CANTONMT: Investigating Back-Translation and Model-Switch Mechanisms for Cantonese-English Neural Machine Translation
by: Hong, Kung Yin, et al.
Published: (2024)

CantonMT: Cantonese to English NMT Platform with Fine-Tuned Models Using Synthetic Back-Translation Data
by: Hong, Kung Yin, et al.
Published: (2024)

ILIAS: Instance-Level Image retrieval At Scale
by: Kordopatis-Zilos, Giorgos, et al.
Published: (2025)

A Self-Evolving Framework for Efficient Terminal Agents via Observational Context Compression
by: Ren, Jincheng, et al.
Published: (2026)

Instance Consistency Regularization for Semi-Supervised 3D Instance Segmentation
by: Wu, Yizheng, et al.
Published: (2024)

BRIDGE: Bootstrapping Text to Control Time-Series Generation via Multi-Agent Iterative Optimization and Diffusion Modeling
by: Li, Hao, et al.
Published: (2025)

Accelerating Multimodal Large Language Models via Dynamic Visual-Token Exit and the Empirical Findings
by: Wu, Qiong, et al.
Published: (2024)

Uncovering the Internet's Hidden Values: An Empirical Study of Desirable Behavior Using Highly-Upvoted Content on Reddit
by: Goyal, Agam, et al.
Published: (2024)

Evaluating the Generation Capabilities of Large Chinese Language Models
by: Zeng, Hui, et al.
Published: (2023)

Uncovering the Hidden Cost of Model Compression
by: Misra, Diganta, et al.
Published: (2023)

Does Vision Accelerate Hierarchical Generalization in Neural Language Learners?
by: Kuribayashi, Tatsuki, et al.
Published: (2023)

Comparative Study of Vision-Based Metric Measurement for Large-Scale Planar Scenes
by: Sun, ZhiXin
Published: (2026)

Seeing Through Deception: Uncovering Misleading Creator Intent in Multimodal News with Vision-Language Models
by: Wu, Jiaying, et al.
Published: (2025)

CAPC-CG: A Large-Scale, Expert-Directed LLM-Annotated Corpus of Adaptive Policy Communication in China
by: Sun, Bolun, et al.
Published: (2025)

InstanceV: Instance-Level Video Generation
by: Chen, Yuheng, et al.
Published: (2025)

Seeing isn't Hearing: Benchmarking Vision Language Models at Interpreting Spectrograms
by: Loakman, Tyler, et al.
Published: (2025)

Propaganda AI: An Analysis of Semantic Divergence in Large Language Models
by: Min, Nay Myat, et al.
Published: (2025)

CraterBench-R: Instance-Level Crater Retrieval for Planetary Scale
by: Fang, Jichao, et al.
Published: (2026)

Hidden Tail: Adversarial Image Causing Stealthy Resource Consumption in Vision-Language Models
by: Zhang, Rui, et al.
Published: (2025)

Separator Injection Attack: Uncovering Dialogue Biases in Large Language Models Caused by Role Separators
by: Li, Xitao, et al.
Published: (2025)

Vision-Based Water Level and Flow Estimation
by: Sun, ZhiXin
Published: (2026)