paper-with-me

홈 › Papers

Don't Buy it! Reassessing the Ad Understanding Abilities of Contrastive Multimodal Models

2024-05-31 · A. Bavaresco, A. Testoni, R. Fernández

Image-based advertisements are complex multimodal stimuli that often contain unusual visual elements and figurative language. Previous research on automatic ad understanding has reported impressive zero-shot accuracy of contrastive vision-and-language models (VLMs) on an ad-explanation retrieval task. Here, we examine the original task setup and show that contrastive VLMs can solve it by exploiting grounding heuristics. To control for this confound, we introduce TRADE, a new evaluation test set with adversarial grounded explanations. While these explanations look implausible to humans, we show that they "fool" four different contrastive VLMs. Our findings highlight the need for an improved operationalisation of automatic ad understanding that truly evaluates VLMs' multimodal reasoning abilities. We make our code and TRADE available at https://github.com/dmg-illc/trade .

📄 PDF Abstract BibTeX arXiv:2405.20846

Code (1)

dmg-illc/trade 공식 구현 jax

Tasks

Multimodal ReasoningRetrieval

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

Hard Negative Contrastive Learning for Fine-Grained Geometric Understanding in Large Multimodal Models

2025-05-26 · Kai Sun, Yushi Bai, Zhen Yang, Jiajie Zhang 외

Benefiting from contrastively trained visual encoders on large-scale natural scene images, Large Multimodal Models (LMMs) have achieved remarkable performance across various visual perception tasks. However, the inherent…

Contrastive LearningMath

Retrieval or Representation? Reassessing Benchmark Gaps in Multilingual and Visually Rich RAG

2026-03-04 · Martin Asenov, Kenza Benkirane, Dan Goldwater, Aneiss Ghodsi arxiv

Retrieval-augmented generation (RAG) is a common way to ground language models in external documents and up-to-date information. Classical retrieval systems relied on lexical methods such as BM25, which rank documents by…

MaskCD: Mitigating LVLM Hallucinations by Image Head Masked Contrastive Decoding

2025-10-03 · Jingyuan Deng, Yujiu Yang arxiv

Large vision-language models (LVLMs) have shown remarkable performance in visual-language understanding for downstream multimodal tasks. While their capabilities are improving, problems emerge simultaneously. Among those…

X-Former: Unifying Contrastive and Reconstruction Learning for MLLMs

2024-07-18 · Sirnam Swetha, Jinyu Yang, Tal Neiman, Mamshad Nayeem Rizve 외

Recent advancements in Multimodal Large Language Models (MLLMs) have revolutionized the field of vision-language understanding by integrating visual perception capabilities into Large Language Models (LLMs). The prevaili…

Contrastive LearningRepresentation LearningVisual Reasoning

Are They the Same? Exploring Visual Correspondence Shortcomings of Multimodal LLMs

2025-01-08 · Yikang Zhou, Tao Zhang, Shilin Xu, Shihao Chen 외

Recent advancements in multimodal models have shown a strong ability in visual perception, reasoning abilities, and vision-language understanding. However, studies on visual matching ability are missing, where finding th…

Contrastive Learning