AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

DETONATE – A Benchmark for Text-to-Image Alignment and Kernelized Direct Preference Optimization

Renjith Prasad Kaippilly Mana, Abhilekh Borah, Hasnat Md Abdullah, Chathurangi Shyalika, Gurpreet Singh, Ritvik Garimella, Rajarshi Roy, Harshul Raj Surana, Nasrin Imanpour, Suranjana Trivedy, Amit Sheth, Amitava Das

AAAI 2026 · Image generation › Text-to-image diffusion · approach: Preference optimization for diffusion · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add DPO-Kernels (RBF + Rényi) in place of DDPOSD-XL · DETONATE held-out promptsToxicity: 0.19 -> 0.12Improved
add DPO-Kernels (RBF + Rényi) in place of DDPOSD-XL · DETONATE held-out promptsAQI: 0.28 -> 0.80Improved
add DPO-Kernels (RBF + Rényi) in place of DDPOSD-v1.5 · DETONATE held-out promptsToxicity: 0.17 -> 0.11Improved
use Rényi divergence instead of KL with RBF kernel (ablation)DPO-Kernels (SD-XL) · DETONATE held-out promptsToxicity: 0.14 -> 0.12Improved
use Rényi divergence instead of KL with polynomial kernel (ablation)DPO-Kernels (SD-XL) · DETONATE held-out promptsToxicity: 0.16 -> 0.14Improved
use Wasserstein divergence instead of KL with RBF kernel (ablation)DPO-Kernels (SD-XL) · DETONATE held-out promptsToxicity: 0.14 -> 0.15No clear effect
use Rényi instead of KL with wavelet kernel (ablation)DPO-Kernels (SD-XL) · DETONATE held-out promptsAQI: 0.76 -> 0.76No clear effect
add DPO-Kernels RBF + KL in place of vanilla DPO baselineSD-XL · DETONATE held-out promptsCLIP Score: 0.325 -> 0.410Improved
add SAFREE (training-free filter) over vanilla modelSD-XL · DETONATE held-out promptsToxicity: 0.31 -> 0.24Improved
use Wavelet + Wasserstein kernel instead of RBF + KL (ablation)DPO-Kernels · HT-SR weighted alphaweighted alpha: 2.02 -> 3.64Hurt
use RBF + KL kernel with weighted alpha below generalization threshold (ablation)DPO-Kernels · HT-SR weighted alphaweighted alpha: 1.82 -> 2.02Mixed
add DPO-Kernels (RBF + Rényi) over DDPO on Disability axisDPO-Kernels · DETONATE Disability axisAQI: 0.16 -> 0.71Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy