paper-with-me

홈 › Papers

Probing Contextual Language Models for Common Ground with Visual Representations

2020-05-01 · NAACL 2021 4 · Gabriel Ilharco, Rowan Zellers, Ali Farhadi, Hannaneh Hajishirzi

The success of large-scale contextual language models has attracted great interest in probing what is encoded in their representations. In this work, we consider a new question: to what extent contextual representations of concrete nouns are aligned with corresponding visual representations? We design a probing model that evaluates how effective are text-only representations in distinguishing between matching and non-matching visual representations. Our findings show that language representations alone provide a strong signal for retrieving image patches from the correct object categories. Moreover, they are effective in retrieving specific instances of image patches; textual context plays an important role in this process. Visually grounded language models slightly outperform text-only language models in instance retrieval, but greatly under-perform humans. We hope our analyses inspire future research in understanding and improving the visual capabilities of language models.

📄 PDF Abstract BibTeX arXiv:2005.00619

Code (0)

등록된 구현이 없습니다.

Tasks

Representation LearningRetrieval

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Weight Decay 설명 없음
Residual Connection 설명 없음
Adam 설명 없음
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

Similar Papers 제목 키워드 기반

CommonsenseVIS: Visualizing and Understanding Commonsense Reasoning Capabilities of Natural Language Models

2023-07-23 · Xingbo Wang, Renfei Huang, Zhihua Jin, Tianqing Fang 외

Recently, large pretrained language models have achieved compelling performance on commonsense benchmarks. Nevertheless, it is unclear what commonsense knowledge the models learn and whether they solely exploit spurious …

Question AnsweringRelational Reasoning

VIPHY: Probing "Visible" Physical Commonsense Knowledge

2022-09-15 · Shikhar Singh, Ehsan Qasemi, Muhao Chen

In recent years, vision-language models (VLMs) have shown remarkable performance on visual reasoning tasks (e.g. attributes, location). While such tasks measure the requisite knowledge to ground and reason over a given v…

Visual Reasoning

Probing Physical Reasoning with Counter-Commonsense Context

2023-06-04 · Kazushi Kondo, Saku Sugawara, Akiko Aizawa

In this study, we create a CConS (Counter-commonsense Contextual Size comparison) dataset to investigate how physical commonsense affects the contextualized size comparison task; the proposed dataset consists of both con…

Probing Biomedical Embeddings from Language Models

2019-04-03 · WS 2019 6 · Qiao Jin, Bhuwan Dhingra, William W. Cohen, Xinghua Lu

Contextualized word embeddings derived from pre-trained language models (LMs) show significant improvements on downstream NLP tasks. Pre-training on domain-specific corpora, such as biomedical articles, further improves …

ArticlesNERWord Embeddings

Rice-VL: Evaluating Vision-Language Models for Cultural Understanding Across ASEAN Countries

2025-12-01 · Tushar Pranav, Eshan Pandey, Austria Lyka Diane Bala, Aman Chadha 외 arxiv

Vision-Language Models (VLMs) excel in multimodal tasks but often exhibit Western-centric biases, limiting their effectiveness in culturally diverse regions like Southeast Asia (SEA). To address this, we introduce RICE-V…

Visual Question AnsweringVisual Grounding