AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

UniReflect: Self-Reflection Tuning for Unified Multimodal Understanding and Generation

Cong Wei, Zepeng Huang, Haoxian Tan, liang shuang, Lishuai Gao, Pengfei Yan, Xiaoming Wei

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Interleaved image-text generation with MLLMs · 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
add self-reflection test-time scalingUniReflect · GenEvaloverall: 0.83 -> 0.89Improved
add self-reflection test-time scalingUniReflect · GEdit-Bench-ENoverall: 5.14 -> 6.07Improved
add self-reflection TTSUniReflect · GenEval++overall: 0.654 -> 0.680Improved
use in-model verifier instead of OmniVerifier for TTS (ablation)UniReflect-TTS · GenEvaloverall: 0.86 -> 0.89Improved
add Semantic Perceiver (ablation)UniReflect · GenEvaloverall: 0.77 -> 0.81Improved
add decoupling verification (ablation)UniReflect · ViVerBenchoverall: 0.525 -> 0.569Improved
add decoupling verification (ablation)UniReflect · GenEvaloverall: 0.77 -> 0.78No clear effect
combine Semantic Perceiver and decoupling (ablation)UniReflect · GenEval / ViVerBenchoverall: 0.77 -> 0.83; 0.525 -> 0.579Improved
include verification task in training mixUniReflect · GEdit-Bench-ENoverall: 5.14 vs 6.41-6.70 for baselinesHurt
fine-tune 4B model on synthetic verification data vs OmniVerifier RL 7BUniReflect · ViVerBenchoverall: 0.559 -> 0.579Improved

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