Scaling Test-time Inference for Visual Grounding

• URL: http://arxiv.org/abs/2601.13633v1
• Date: Tue, 20 Jan 2026 06:07:02 GMT
• Title: Scaling Test-time Inference for Visual Grounding
• Authors: Guanqi Zhan, Changye Li, Zhijian Liu, Yao Lu, Yi Wu, Song Han, Ligeng Zhu,
• Abstract summary: 'Efficient visual Grounding language Models' (EGM) is a method to scale the test-time computation of a small model.<n> scaling the test-time computation of a small model is deployment-friendly, and yields better end-to-end latency.<n>EGM-Qwen3-VL-8B demonstrates 91.4 IoU with an average of 737ms (5.9x faster) latency while Qwen3-VL-235B demands 4,320ms to achieve 90.5 IoU.
• Score: 23.80116923122471
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Visual grounding is an essential capability of Visual Language Models (VLMs) to understand the real physical world. Previous state-of-the-art grounding visual language models usually have large model sizes, making them heavy for deployment and slow for inference. However, we notice that the sizes of visual encoders are nearly the same for small and large VLMs and the major difference is the sizes of the language models. Small VLMs fall behind larger VLMs in grounding because of the difference in language understanding capability rather than visual information handling. To mitigate the gap, we introduce 'Efficient visual Grounding language Models' (EGM): a method to scale the test-time computation (#generated tokens). Scaling the test-time computation of a small model is deployment-friendly, and yields better end-to-end latency as the cost of each token is much cheaper compared to directly running a large model. On the RefCOCO benchmark, our EGM-Qwen3-VL-8B demonstrates 91.4 IoU with an average of 737ms (5.9x faster) latency while Qwen3-VL-235B demands 4,320ms to achieve 90.5 IoU. To validate our approach's generality, we further set up a new amodal grounding setting that requires the model to predict both the visible and occluded parts of the objects. Experiments show our method can consistently and significantly improve the vanilla grounding and amodal grounding capabilities of small models to be on par with or outperform the larger models, thereby improving the efficiency for visual grounding.

Related papers

• DriveVLA-W0: World Models Amplify Data Scaling Law in Autonomous Driving [52.63591791507895]
  We propose textbfDriveVLA-W0, a training paradigm that employs world modeling to predict future images.<n>This task generates a dense, self-supervised signal that compels the model to learn the underlying dynamics of the driving environment.<n>Experiments on the NAVSIM v1/v2 benchmark and a 680x larger in-house dataset demonstrate that DriveVLA-W0 significantly outperforms BEV and VLA baselines.
  arXiv  Detail & Related papers  (2025-10-14T17:59:47Z)
• EfficientLLaVA:Generalizable Auto-Pruning for Large Vision-language Models [64.18350535770357]
  We propose an automatic pruning method for large vision-language models to enhance the efficiency of multimodal reasoning.<n>Our approach only leverages a small number of samples to search for the desired pruning policy.<n>We conduct extensive experiments on the ScienceQA, Vizwiz, MM-vet, and LLaVA-Bench datasets for the task of visual question answering.
  arXiv  Detail & Related papers  (2025-03-19T16:07:04Z)
• MUSE-VL: Modeling Unified VLM through Semantic Discrete Encoding [6.538592344967826]
  We introduce MUSE-VL, a Unified Vision-Language Model Semantic through discrete semantic.<n>Our method improved the understanding performance by 4.8% compared to the previous SOTA Emu3 and surpassed the dedicated understanding model LLaVA-NeXT 34B by 3.7%.
  arXiv  Detail & Related papers  (2024-11-26T03:33:52Z)
• Cross-model Control: Improving Multiple Large Language Models in One-time Training [34.98931804630706]
  Cross-model Control (CMC) is a method that improves multiple large language models in one-time training. Based on this insight, we incorporate a tiny language model with a minimal number of parameters. We propose a novel token mapping strategy named PM-MinED to make this tiny language model applicable to models with different vocabularies.
  arXiv  Detail & Related papers  (2024-10-23T06:52:09Z)
• Knowledge boosting during low-latency inference [20.617827647115874]
  Models for low-latency, streaming applications could benefit from the knowledge capacity of larger models, but edge devices cannot run these models due to resource constraints. We propose knowledge boosting, a novel technique that allows a large model to operate on time-delayed input during inference, while still boosting small model performance. Our results show larger gains where the performance gap between the small and large models is wide, demonstrating a promising method for large-small model collaboration for low-latency applications.
  arXiv  Detail & Related papers  (2024-07-09T22:04:23Z)
• Memory-Space Visual Prompting for Efficient Vision-Language Fine-Tuning [59.13366859237086]
  Current solutions for efficiently constructing large vision-language (VL) models follow a two-step paradigm. We consider visual prompts as additional knowledge that facilitates language models in addressing tasks associated with visual information. We introduce a novel approach, wherein visual prompts are memoryd with the weights of FFN for visual knowledge injection.
  arXiv  Detail & Related papers  (2024-05-09T08:23:20Z)
• LLaVA-Grounding: Grounded Visual Chat with Large Multimodal Models [105.7362622712606]
  Grounding capability in large multi-modal models (LMMs) is increasingly recognized. The problem is the lack of a dataset for grounded visual chat (GVC) We have created GVC data that allows for the combination of grounding and chat capabilities. Our model achieves competitive performance on classic grounding benchmarks like RefCOCO/+/g and Flickr30K Entities.
  arXiv  Detail & Related papers  (2023-12-05T18:29:31Z)
• Contrastive Alignment of Vision to Language Through Parameter-Efficient Transfer Learning [60.26952378997713]
  Contrastive vision-language models (e.g. CLIP) are created by updating all the parameters of a vision model and language model through contrastive training. We show that a minimal set of parameter updates ($$7%) can achieve the same performance as full-model training. We describe a series of experiments: we show that existing knowledge is conserved more strongly in parameter-efficient training.
  arXiv  Detail & Related papers  (2023-03-21T14:12:08Z)
• eP-ALM: Efficient Perceptual Augmentation of Language Models [70.47962271121389]
  We propose to direct effort to efficient adaptations of existing models, and propose to augment Language Models with perception. Existing approaches for adapting pretrained models for vision-language tasks still rely on several key components that hinder their efficiency. We show that by freezing more than 99% of total parameters, training only one linear projection layer, and prepending only one trainable token, our approach (dubbed eP-ALM) significantly outperforms other baselines on VQA and Captioning.
  arXiv  Detail & Related papers  (2023-03-20T19:20:34Z)
• Scaling Vision-Language Models with Sparse Mixture of Experts [128.0882767889029]
  We show that mixture-of-experts (MoE) techniques can achieve state-of-the-art performance on a range of benchmarks over dense models of equivalent computational cost. Our research offers valuable insights into stabilizing the training of MoE models, understanding the impact of MoE on model interpretability, and balancing the trade-offs between compute performance when scaling vision-language models.
  arXiv  Detail & Related papers  (2023-03-13T16:00:31Z)
• Does language help generalization in vision models? [0.0]
  We show that a visual model trained on a very large supervised image dataset (ImageNet-21k) can be as efficient for generalization as its multimodal counterpart (CLIP) When compared to other standard visual or language models, the latent representations of BiT-M were found to be just as "linguistic" as those of CLIP.
  arXiv  Detail & Related papers  (2021-04-16T18:54:14Z)

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.