AI Paper Atlas

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

LLM Thought Divergence and Convergence for Dialogue-Based Image Generation Control

Hui Li

ICCV 2025 · Image generation › Text-to-image diffusion · approach: T2I evaluation and controllable adapters · 0 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
use LTDC instead of GILLGILL · VisDial valCLIP similarity: 64.50 -> 69.37Improved
add LTDC divergence and convergence to the vanilla ChatGPT-4o-mini pipelineVanilla(SC) · VisDial valCLIP similarity: 67.41 -> 69.37Improved
add LTDC to the vanilla Janus-Pro + Deepseek-R1 pipelineVanilla(JD) · VisDial valCLIP similarity: 68.61 -> 70.03Improved
add LTDC to vanilla Stable Diffusion + ChatGPT-4o-miniVanilla(SC) · VisDial valLPIPS: 71.38 -> 70.71No clear effect
replace Stable Diffusion by Janus-Pro with Deepseek-R1 (vanilla)Vanilla(SD) · VisDial valCLIP similarity: 67.37 -> 68.61Improved
replace Stable Diffusion by Janus-Pro with ChatGPT-4o-mini (vanilla) (ablation)Vanilla(SC) · VisDial valCLIP similarity: 67.41 -> 67.09No clear effect
use Deepseek-R1 instead of Llama3.3 for task breakdown (Janus-Pro) (ablation)Vanilla(JL) · VisDial valCLIP similarity: 68.30 -> 68.61Improved
use ChatGPT-4o-latest instead of ChatGPT-4o-mini as evaluator (ablation)LTDC evaluator · Multi-modal evaluator accuracy (100 VisDial examples)accuracy (%): 73.82 -> 81.19Improved
use Llama3.2 (90b) instead of Qwen-vl-plus (7b) as evaluator (ablation)LTDC evaluator · Multi-modal evaluator accuracy (100 VisDial examples)accuracy (%): 66.13 -> 77.56Improved
use ChatGPT-4o as evaluator under the same prompt (ablation)LTDC evaluator · Multi-modal evaluator accuracy (100 VisDial examples)accuracy (%): FailHurt
use LTDC instead of GILL in human evaluationGILL · VisDial val (100 samples, 3 human evaluators)correct ratio (%): 58.3 -> 72.9Improved

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