ViperGPT generates executable Python code to compose pre-trained vision-and-language modules into programs that answer visual queries, reaching state-of-the-art results with no additional training.
arXiv preprint arXiv:2005.00724
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
RecPIE jointly optimizes recommendation predictions and LLM-generated natural-language explanations via alternating training and reinforcement learning, yielding 3-4% accuracy gains and higher human preference on Google Maps POI data.
citing papers explorer
-
ViperGPT: Visual Inference via Python Execution for Reasoning
ViperGPT generates executable Python code to compose pre-trained vision-and-language modules into programs that answer visual queries, reaching state-of-the-art results with no additional training.
-
Can Explanations Improve Recommendations? Evidence from Prediction-Informed Explanations
RecPIE jointly optimizes recommendation predictions and LLM-generated natural-language explanations via alternating training and reinforcement learning, yielding 3-4% accuracy gains and higher human preference on Google Maps POI data.