Zobrazeno 1 - 3
of 3
pro vyhledávání: '"Pandya, Pranshu"'
Cognitive textual and visual reasoning tasks, such as puzzles, series, and analogies, demand the ability to quickly reason, decipher, and evaluate patterns both textually and spatially. While LLMs and VLMs, through extensive training on large amounts
Externí odkaz:
http://arxiv.org/abs/2407.10380
Autor:
Singh, Shubhankar, Chaurasia, Purvi, Varun, Yerram, Pandya, Pranshu, Gupta, Vatsal, Gupta, Vivek, Roth, Dan
Existing benchmarks for visual question answering lack in visual grounding and complexity, particularly in evaluating spatial reasoning skills. We introduce FlowVQA, a novel benchmark aimed at assessing the capabilities of visual question-answering m
Externí odkaz:
http://arxiv.org/abs/2406.19237
Language models, characterized by their black-box nature, often hallucinate and display sensitivity to input perturbations, causing concerns about trust. To enhance trust, it is imperative to gain a comprehensive understanding of the model's failure
Externí odkaz:
http://arxiv.org/abs/2311.08662