A new 40K-pair vision-language dataset for social navigation and a fine-tuned VLM that reportedly beats GPT-4V and Gemini in human-judged scene reasoning.
Socially compliant navigation dataset (scand): A large-scale dataset of demonstrations for social navigation,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
dataset 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
REJECT 1roles
dataset 1polarities
support 1representative citing papers
citing papers explorer
-
Social-LLaVA: Enhancing Robot Navigation through Human-Language Reasoning in Social Spaces
A new 40K-pair vision-language dataset for social navigation and a fine-tuned VLM that reportedly beats GPT-4V and Gemini in human-judged scene reasoning.