English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Combining the visual modality with pretrained language models has beensurprisingly effective for simple descriptive tasks such as image captioning.More general text generation however remains elusive. We take a step back andask: How do these models work for more complex generative tasks, i.e.conditioning on both text and images? Are multimodal models simply visuallyadapted language models, or do they combine they reason jointly overmodalities?  We investigate these questions in the context of self-rationalization(jointly generating task labels/answers and free-text explanations) of threetasks: (i) visual question answering in VQA-X, (ii) visual commonsensereasoning in VCR, and (iii) visual-textual entailment in e-SNLI-VE. We showthat recent unimodal advances, CLIP image representations and scaling oflanguage models, do not consistently improve self-rationalization in multimodaltasks. We find that no single model type works universally best across tasks,datasets, and finetuning data sizes. Our findings motivate the need for novelgeneral backbones approach that move text generation from images and textbeyond image captioning.

On Advances in Text Generation from Images Beyond Captioning: A Case Study in Self-Rationalization