mlsanityData to 7 Oct 2026

V-DyKnow: A Dynamic Benchmark for Time-Sensitive Knowledge in Vision Language Models

arXiv 2603.16581 ↗First posted 17 Mar 2026cs.AIcited by 0

On the map: Multimodal LLMs / vision-language models → VLM benchmarks and visual reasoning

Abstract (arXiv)

Vision-Language Models (VLMs) are trained on data snapshots of documents, including images and texts. Their training data and evaluation benchmarks are typically static, implicitly treating factual knowledge as time-invariant. However, real-world facts are intrinsically time-sensitive and subject to erratic and periodic changes, causing model predictions to become outdated. We present V-DyKnow, a Visual Dynamic Knowledge benchmark for evaluating time-sensitive factual knowledge in VLMs. Using V-DyKnow, we benchmark closed- and open-source VLMs and analyze a) the reliability (correctness and consistency) of model responses across modalities and input perturbations; b) the efficacy of knowledge editing and multi-modal RAG methods for knowledge updates across modalities; and c) the sources of outdated predictions, through data and mechanistic analysis. Our results show that VLMs frequently output outdated facts, reflecting outdated snapshots used in the (pre-)training phase. Factual reliability degrades from textual to visual stimuli, even when entities are correctly recognized. Besides, existing alignment approaches fail to consistently update the models' knowledge across modalities. Together, these findings highlight fundamental limitations in how current VLMs acquire and update time-sensitive knowledge across modalities. We release the benchmark, code, and evaluation data.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in VLM benchmarks and visual reasoning, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 19 of its 31 references that are in the map. First posted on arXiv 17 Mar 2026.

Builds on

19 of 31 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025CIVET: Systematic Evaluation of Understanding in VLMs5 Jun

Other references · 18

  1. 2023Improved Baselines with Visual Instruction Tuningcited by 4,362
  2. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,151
  3. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,669
  4. 2022Locating and Editing Factual Associations in GPTcited by 2,609
  5. 2025InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiencycited by 1,419
  6. 2023Dissecting Recall of Factual Associations in Auto-Regressive Language Modelscited by 512
  7. 2024The Platonic Representation Hypothesiscited by 362
  8. 2023Can We Edit Factual Knowledge by In-Context Learning?cited by 284
  9. 2022Aging with GRACE: Lifelong Model Editing with Discrete Key-Value Adaptorscited by 267
  10. 2023Can Pre-trained Vision and Language Models Answer Visual Information-Seeking Questions?cited by 203
  11. 2024ClashEval: Quantifying the tug-of-war between an LLM's internal prior and external evidencecited by 109
  12. 2024DyKnow: Dynamically Verifying Time-Sensitive Factual Knowledge in LLMscited by 23
6 more
  1. 2025How Do LLMs Acquire New Knowledge? A Knowledge Circuits Perspective on Continual Pre-Trainingcited by 22
  2. 2024Lifelong Knowledge Editing for Vision Language Models with Low-Rank Mixture-of-Expertscited by 14
  3. 2025MCIF: Multimodal Crosslingual Instruction-Following Benchmark from Scientific Talkscited by 14
  4. 2025LLMs as Repositories of Factual Knowledge: Limitations and Solutionscited by 10
  5. 2025Training-Free Personalization via Retrieval and Reasoning on Fingerprintscited by 10
  6. 2024Formality is Favored: Unraveling the Learning Preferences of Large Language Models on Data with Conflicting Knowledgecited by 8

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Test-Time Consistency in Vision Language Models
  2. 2025Dynamic Knowledge Integration for Enhanced Vision-Language Reasoning
  3. 2022Vision-Language Intelligence: Tasks, Representation Learning, and Large Models
  4. 2023Synthesize, Diagnose, and Optimize: Towards Fine-Grained Vision-Language Understanding
  5. 2026TGV-KV: Text-Grounded KV Eviction for Vision-Language Models
  6. 2025A Matter of Time: Revealing the Structure of Time in Vision-Language Models
  7. 2026On Test-Time Scaling for Vision-Language Models
  8. 2026DataComp-VLM: Improved Open Datasets for Vision-Language Models