MEGA Hub

Mine, Sample, Render: Test-Time Diversity Alignment Of Generative Models

Authors

Do you know Kostas Triaridis?You can claim authorship or link another user.Do you know Morris E. Sharp?You can claim authorship or link another user.Do you know Nebojsa Jojic?You can claim authorship or link another user.

Abstract

Modern generative models produce high-quality individual outputs but struggle to sample diversely, even when explicitly instructed to do so. We introduce MINE-SAMPLE-RENDER (MSR), a test-time approach that externalizes the diversity into an explicit, corpus-derived sampling procedure. Our pipeline (i) mines an interpretable categorical schema from a corpus of real artifacts and supplies statistics for sampling; (ii) samples attribute configurations from the corpus-derived statistics; (iii) uses a generative model purely as a conditional renderer of a sampled configuration. Across weather report generation, user simulation, webpage design, text-to-image generation, and agent-generated 3D assets, we evaluate the diversity and output quality of MSR-guided generation and show that MSR improves alignment with corpus distributions and diversity of generated outputs.

Community

00