The paper creates FISD, a controlled benchmark for composed image retrieval that removes query ambiguity via generative models, and proposes a multi-round agentic evaluation to assess models in interactive settings.
Conceptual captions: A cleaned, hypernymed, im- age alt-text dataset for automatic image captioning
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 4roles
background 1polarities
background 1representative citing papers
MVBench is a benchmark of 20 temporal video understanding tasks built by transforming static tasks into dynamic ones, with VideoChat2 outperforming prior MLLMs by over 15%.
Sparse autoencoders are derived as MAP estimators for a continuous topic model, yielding a reusable topic modeling framework that produces coherent topics on text and image datasets.
InternVL scales a vision model to 6B parameters and aligns it with LLMs using web data to achieve state-of-the-art results on 32 visual-linguistic benchmarks.
citing papers explorer
-
A Sanity Check on Composed Image Retrieval
The paper creates FISD, a controlled benchmark for composed image retrieval that removes query ambiguity via generative models, and proposes a multi-round agentic evaluation to assess models in interactive settings.
-
MVBench: A Comprehensive Multi-modal Video Understanding Benchmark
MVBench is a benchmark of 20 temporal video understanding tasks built by transforming static tasks into dynamic ones, with VideoChat2 outperforming prior MLLMs by over 15%.
-
Sparse Autoencoders are Topic Models
Sparse autoencoders are derived as MAP estimators for a continuous topic model, yielding a reusable topic modeling framework that produces coherent topics on text and image datasets.
-
InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks
InternVL scales a vision model to 6B parameters and aligns it with LLMs using web data to achieve state-of-the-art results on 32 visual-linguistic benchmarks.