Reading-Papers

논문 리뷰 정리. 주제별 폴더 안에 PAPER_*.md 형식으로 보관합니다. 테이블은 최신 업데이트가 위쪽.

---

📂 Diffusion

No	Title	Venue	Paper	Code	Updated
29	i1: A Recipe for Text-to-Image Diffusion from Public Materials	arXiv 2026	arxiv	github	2026-06-19
28	Ovis-Image Technical Report	arXiv 2025	arxiv	github	2026-06-19
27	Query-Kontext: An Unified Multimodal Model for Image Generation and Editing	arXiv 2025	arxiv	-	2026-06-19
26	MiniT2I: A Minimalist Baseline for Text-to-Image Generation	Blog 2026	blog	github	2026-06-18
24	TREAD: Token Routing for Efficient Architecture-agnostic Diffusion Training	arXiv 2025	arxiv	-	2026-06-17
23	REPA: Representation Alignment for Generation	ICLR 2025 (Oral)	arxiv	github	2026-06-17
22	EMA: Exponential Moving Average of Model Weights (Algorithm Reference)	Algorithm Ref	ddpm	-	2026-06-16
21	LPIPS: The Unreasonable Effectiveness of Deep Features as a Perceptual Metric	CVPR 2018	arxiv	github	2026-06-16
20	PRX (Photoroom eXperimental) — Open-source T2I Research	HF Blog 2026	blog	github	2026-06-12
19	DDT: Decoupled Diffusion Transformer	arXiv 2025	arxiv	github	2026-06-10
18	PixelDiT: Towards Pixel-Space Image Generation via Single-Stage Diffusion Transformer	CVPR 2026	arxiv	github	2026-06-10
17	Qwen-Image-2.0 Technical Report	arXiv 2026	arxiv	github	2026-06-04
16	Qwen-Image Technical Report	arXiv 2025	arxiv	github	2026-06-16
15	Nucleus-Image: Sparse MoE for Image Generation	arXiv 2026	arxiv	-	2026-06-02
14	PiD: Fast and High-Resolution Latent Decoding with Pixel Diffusion	arXiv 2026	arxiv	github	2026-06-19
13	Lumina-Image 2.0: A Unified and Efficient Image Generative Model	arXiv 2025	arxiv	github	2026-05-29
12	SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion Transformer	arXiv 2025	arxiv	github	2026-05-29
11	LongCat-Image: 6B Bilingual Efficient Image Generation Foundation Model	arXiv 2025	arxiv	github	2026-05-28
10	UniCustom: Unified Visual Conditioning for Multi-Reference Image Generation	arXiv 2026	arxiv	-	2026-05-27
9	DreamLite: A Lightweight On-Device Unified Model for Image Generation and Editing	arXiv 2026	arxiv	github	2026-05-26
8	Asymmetric Flow Models	arXiv 2026	arxiv	github	2026-05-22
7	Lumina-Next: Making Lumina-T2X Stronger and Faster with Next-DiT	arXiv 2024	arxiv	github	2026-05-29
6	Efficient Diffusion Training via Min-SNR Weighting Strategy	ICCV 2023	arxiv	github	2026-05-15
5	SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformers	arXiv 2024	arxiv	github	2026-05-29
4	PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis	ICLR 2024 (Spotlight)	arxiv	github	2026-05-14
2	Z-Image: An Efficient Image Generation Foundation Model	arXiv 2025	arxiv	github	2026-05-26
1	Mean Mode Screaming: Mean–Variance Split Residuals for 1000-Layer Diffusion Transformers	arXiv 2026	arxiv	github	2026-05-13

📂 Diffusion-Native-Unified

사전학습된 LLM/VLM 백본을 그대로 재사용하고, 여기에 픽셀 어댑터 + diffusion head/모듈을 얹어 이해(understanding)와 생성(generation)을 한 모델로 통합한 native unified multimodal 계열. VLM 백본(HiDream-O1, SenseNova-U1) / LLM 백본 + 인코더 제거(TUNA-2) 모두 포함.

No	Title	Venue	Paper	Code	Updated
3	SenseNova-U1: Unifying Multimodal Understanding and Generation with NEO-unify Architecture	arXiv 2026	arxiv	github	2026-06-19
2	TUNA-2: Encoder-free Native Unified Multimodal Model	arXiv 2026	arxiv	github	2026-06-17
1	HiDream-O1-Image: A Natively Unified Image Generative Foundation Model with Pixel-level Unified Transformer	Tech Report 2026	pdf	github	2026-06-02

📂 Diffusion-Edit

No	Title	Venue	Paper	Code	Updated
9	Query-Kontext: An Unified Multimodal Model for Image Generation and Editing	arXiv 2025	arxiv	-	2026-06-19
8	Tstars-Tryon 1.0: Industrial-Grade Virtual Try-On	arXiv 2026	arxiv	-	2026-06-15
7	UniRef-Image-Edit: Towards Scalable and Consistent Multi-Reference Image Editing	arXiv 2026	arxiv	-	2026-06-10
6	FireRed-Image-Edit-1.0 Technical Report	arXiv 2026	arxiv	github	2026-06-10
5	Qwen-Image-2.0 Technical Report	arXiv 2026	arxiv	github	2026-06-04
4	HiDream-O1-Image: A Natively Unified Image Generative Foundation Model with Pixel-level Unified Transformer	Tech Report 2026	pdf	github	2026-06-02
3	DreamLite: A Lightweight On-Device Unified Model for Image Generation and Editing	arXiv 2026	arxiv	github	2026-05-26
2	UniCustom: Unified Visual Conditioning for Multi-Reference Image Generation	arXiv 2026	arxiv	-	2026-05-27
1	Any2AnyTryon: Leveraging Adaptive Position Embeddings for Versatile Virtual Clothing Tasks	ICCV 2025	arxiv	github	2026-05-27

📂 Diffusion-Post-Training

No	Title	Venue	Paper	Code	Updated
1	Qwen-Image-2.0-RL: RLHF + On-Policy Distillation Recipe	arXiv 2026	arxiv	-	2026-06-19

📂 Diffusion-Distillation

No	Title	Venue	Paper	Code	Updated
2	Improved Distribution Matching Distillation for Fast Image Synthesis	NeurIPS 2024	arxiv	github	2026-05-22
1	One-step Diffusion with Distribution Matching Distillation	CVPR 2024	arxiv	-	2026-05-22