AIDE, a four-stage agentic pipeline, uses the VLM itself to select and enrich training examples with expert tool outputs, yielding modest single-digit benchmark gains on Eagle-8B.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
AIDE: Agentically Improve Visual Language Model with Domain Experts
AIDE, a four-stage agentic pipeline, uses the VLM itself to select and enrich training examples with expert tool outputs, yielding modest single-digit benchmark gains on Eagle-8B.