paper-with-me

홈 › Papers

Scaling Inference-Time Search with Vision Value Model for Improved Visual Comprehension

2024-12-04 · Xiyao Wang, Zhengyuan Yang, Linjie Li, Hongjin Lu, Yuancheng Xu, Chung-Ching Lin, Kevin Lin, Furong Huang, Lijuan Wang

Despite significant advancements in vision-language models (VLMs), there lacks effective approaches to enhance response quality by scaling inference-time computation. This capability is known to be a core step towards the self-improving models in recent large language model studies. In this paper, we present Vision Value Model (VisVM) that can guide VLM inference-time search to generate responses with better visual comprehension. Specifically, VisVM not only evaluates the generated sentence quality in the current search step, but also anticipates the quality of subsequent sentences that may result from the current step, thus providing a long-term value. In this way, VisVM steers VLMs away from generating sentences prone to hallucinations or insufficient detail, thereby producing higher quality responses. Experimental results demonstrate that VisVM-guided search significantly enhances VLMs' ability to generate descriptive captions with richer visual details and fewer hallucinations, compared with greedy decoding and search methods with other visual reward signals. Furthermore, we find that self-training the model with the VisVM-guided captions improve VLM's performance across a wide range of multimodal benchmarks, indicating the potential for developing self-improving VLMs. Our value model and code are available at https://github.com/si0wang/VisVM.

📄 PDF Abstract BibTeX arXiv:2412.03704

Code (1)

si0wang/visvm 공식 구현 pytorch

Tasks

DescriptiveLanguage ModelingLanguage ModellingLarge Language ModelSentence

Similar Papers 제목 키워드 기반

ScalingNoise: Scaling Inference-Time Search for Generating Infinite Videos

2025-03-20 · Haolin Yang, Feilong Tang, Ming Hu, Yulong Li 외

Video diffusion models (VDMs) facilitate the generation of high-quality videos, with current research predominantly concentrated on scaling efforts during training through improvements in data quality, computational reso…

DenoisingDiversityVideo Generation

Value-Guided Search for Efficient Chain-of-Thought Reasoning

2025-05-23 · Kaiwen Wang, Jin Peng Zhou, Jonathan Chang, Zhaolin Gao 외

In this paper, we propose a simple and efficient method for value model training on long-context reasoning traces. Compared to existing process reward models (PRMs), our method does not require a fine-grained notion of "…

Math

AVIS: Adaptive Test-Time Scaling for Vision-Language Models

2026-06-10 · Ahmadreza Jeddi, Minh Ngoc Le, Amirhossein Kazerouni, Hakki Can Karaimer 외 arxiv

Modern Vision-Language Models (VLMs) benefit from chain-of-thought prompting and test-time scaling, but these gains often come with prohibitive inference cost due to large visual contexts and long decoding chains. We vie…

Visual Reasoning

Learning Probabilistic Multi-Modal Actor Models for Vision-Based Robotic Grasping

2019-04-15 · Mengyuan Yan, Adrian Li, Mrinal Kalakrishnan, Peter Pastor

Many previous works approach vision-based robotic grasping by training a value network that evaluates grasp proposals. These approaches require an optimization process at run-time to infer the best action from the value …

Robotic Grasping

A*-Decoding: Token-Efficient Inference Scaling

2025-05-19 · Giannis Chatziveroglou

Inference-time scaling has emerged as a powerful alternative to parameter scaling for improving language model performance on complex reasoning tasks. While existing methods have shown strong performance gains under fixe…

Language ModelingLanguage Modelling