The Shape of Time: Video-Token Contrast for Temporal Understanding in VideoLMs

Yumeng Shi; Quanyu Long; Yin Wu; Wenya Wang

Home/Archive/arXiv

arXiv · 2609.04110

The Shape of Time: Video-Token Contrast for Temporal Understanding in VideoLMs

Yumeng Shi·Quanyu Long·Yin Wu·Wenya Wang

View original source ↗Read source document ↗Export metadata ↓

Abstract

Seeing frames in order does not mean representing time. Modern VideoLMs receive ordered video streams, yet their main supervision acts on generated text rather than video-token representations where event dynamics should first emerge. This mismatch allows models to learn temporal answers from shortcuts such as objects, scenes, and language priors, without requiring internal video representations to capture event progression. To address this, we propose VT-Contrast, a representation-level temporal counterfactual objective for VideoLMs. Its design asks where temporal supervision should act and what temporal differences it should expose. VT-Contrast supervises selected late-layer last-frame video tokens, where temporal information is expected to be integrated before language generation, and contrasts order-preserving views with same-video reordered counterfactuals graded by Kendall tau distance. It requires no architectural changes, is compatible with diverse VideoLM training tasks, and improves overall performance across temporal understanding benchmarks. Our code is available at https://github.com/ANDgate99/VT-Contrast.

Explore related subjects

cs.CV

Keep this discovery

BibTeX RIS

Yumeng Shi, Quanyu Long, Yin Wu, Wenya Wang. 2026-09-03. The Shape of Time: Video-Token Contrast for Temporal Understanding in VideoLMs. https://arxiv.org/abs/2609.04110

Cite the original work for its findings. Save a collection to share your selection of sources.

Discover connections

cs.CV subject

Connections use source metadata and explicit phrase matches, not verified experimental comparisons.

KEEP EXPLORING

Related papers

arXiv2026

Stochastic Optimization of Tree Tensor Networks

Tensor networks, originally developed for quantum many-body physics, are promising models for machine learning. We derive stochastic Riemannian optimizers for tree tensor networks (TTNs) on both their parameter and quotient manifolds, including adaptive and learning-rate-free schemes suitable for minibatch training. Using a hybrid CNN-TTN architecture, we evaluate the methods on Fashion-MNIST, CIFAR10, and Imagenette. The proposed optimizers achieve predictive performance comparable to unconstrained optimization while enabling numerically stable downstream compression.

math.OC↗

arXiv2026

Can We Change the Stroke Size for Easier Diffusion?

Diffusion models can be challenged in the low signal-to-noise regime, where they have to make pixel-level predictions despite the presence of high noise. The geometric intuition is akin to using the finest stroke for oil painting throughout, which may be ineffective. We therefore study \emph{stroke-size control} as a controlled intervention that changes the roughness of the supervised target, predictions and perturbations across timesteps, in an attempt to ease the low signal-to-noise challenge via the prediction target simplification.

cs.CV↗

arXiv2026

Texture Image Classification Using DWT AlexNet Feature Fusion and Deep Neural Networks

Texture image classification plays a significant role in computer vision applications, including industrial inspection, medical image analysis, remote sensing, and object recognition. Handcrafted features can capture local texture characteristics but may have limited capability to represent complex visual patterns. In contrast, deep learning models automatically learn discriminative representations but may not fully exploit the multiscale spatial-frequency information inherent in texture images. This paper proposes a hybrid feature fusion framework, termed DWT_AlexNet_DNN, which combines Discrete Wavelet Transform (DWT) features with deep features extracted using AlexNet for texture image classification.

cs.CV↗