paper-with-me

Papers

Can Language Models Understand Physical Concepts?

2023-05-23 · Lei LI, Jingjing Xu, Qingxiu Dong, Ce Zheng, Qi Liu, Lingpeng Kong, Xu sun

Language models~(LMs) gradually become general-purpose interfaces in the interactive and embodied world, where the understanding of physical concepts is an essential prerequisite. However, it is not yet clear whether LMs can understand physical concepts in the human world. To investigate this, we design a benchmark VEC that covers the tasks of (i) Visual concepts, such as the shape and material of objects, and (ii) Embodied Concepts, learned from the interaction with the world such as the temperature of objects. Our zero (few)-shot prompting results show that the understanding of certain visual concepts emerges as scaling up LMs, but there are still basic concepts to which the scaling law does not apply. For example, OPT-175B performs close to humans with a zero-shot accuracy of 85\% on the material concept, yet behaves like random guessing on the mass concept. Instead, vision-augmented LMs such as CLIP and BLIP achieve a human-level understanding of embodied concepts. Analysis indicates that the rich semantics in visual representation can serve as a valuable source of embodied knowledge. Inspired by this, we propose a distillation method to transfer embodied knowledge from VLMs to LMs, achieving performance gain comparable with that by scaling up the parameters of LMs 134x. Our dataset is available at \url{https://github.com/TobiasLee/VEC}

📄 PDF Abstract BibTeX arXiv:2305.14057

Code (1)

tobiaslee/vec 공식 구현

Methods 이 논문이 사용한 방법론

CLIP Contrastive Language-Image Pre-training (CLIP), consisting of a simplified version of ConVIRT trained from scratch, is an efficient method of image representation learning…
BLIP Vision-Language Pre-training (VLP) has advanced the performance for many vision-language tasks. However, most existing pre-trained models only excel in either understanding-based…

Similar Papers 제목 키워드 기반

Physically Grounded Vision-Language Models for Robotic Manipulation

2023-09-05 · Jensen Gao, Bidipta Sarkar, Fei Xia, Ted Xiao 외

Recent advances in vision-language models (VLMs) have led to improved performance on tasks such as visual question answering and image captioning. Consequently, these models are now well-positioned to reason about the ph…

Image CaptioningLanguage ModellingLarge Language ModelObject+2

Digital Gene: Learning about the Physical World through Analytic Concepts

2025-04-05 · Jianhua Sun, Cewu Lu

Reviewing the progress in artificial intelligence over the past decade, various significant advances (e.g. object detection, image generation, large language models) have enabled AI systems to produce more semantically m…

Image Generationobject-detectionObject DetectionPhilosophy

Intrinsic Physical Concepts Discovery with Object-Centric Predictive Models

2023-03-03 · CVPR 2023 1 · Qu Tang, Xiangyu Zhu, Zhen Lei, Zhaoxiang Zhang

The ability to discover abstract physical concepts and understand how they work in the world through observing lies at the core of human intelligence. The acquisition of this ability is based on compositionally perceivin…

On the Learnability of Physical Concepts: Can a Neural Network Understand What's Real?

2022-07-25 · Alessandro Achille, Stefano Soatto

We revisit the classic signal-to-symbol barrier in light of the remarkable ability of deep neural networks to generate realistic synthetic data. DeepFakes and spoofing highlight the feebleness of the link between physica…

WorldBench: Benchmarking Physical Understanding of World Models by Isolating Physics Concepts

2026-01-29 · Rishi Upadhyay, Howard Zhang, Jim Solomon, Ayush Agrawal 외 arxiv

Recent advances in generative foundational models, often termed "world models," have propelled interest in applying them to critical tasks like robotic planning and autonomous system training. For reliable deployment, th…

Video Generation