A multimodal language model with special tokens for deciding when to retrieve and which passages to trust beats earlier retrieval-augmented methods on knowledge-based visual question answering.
Improving Language Models by Retrieving from Trillions of Tokens
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Augmenting Multimodal LLMs with Self-Reflective Tokens for Knowledge-based Visual Question Answering
A multimodal language model with special tokens for deciding when to retrieve and which passages to trust beats earlier retrieval-augmented methods on knowledge-based visual question answering.