📚 Reading-Papers

이 공간은 🎭 Pierrot Universe 안에 구현되어 있는 관련/참조 연구들과, 연구자가 관심 있는 분야의 연구들을 리뷰하는 공간입니다.

---

📰 News

• 2026-10-26 — DiffusionOPD · DanceOPD · ArcFlow · PVD (Diffusion-Distillation) 리뷰 추가
• 2026-10-22 — GroundAnything (VLM) 리뷰 추가 — 블록 확산 grounding
• 2026-10-16 — RG (Diffusion) · EmbeddingGemma 2 (VLM) 리뷰 추가
• 2026-09-30 — LocateAnything (VLM, ECCV 2026) 리뷰 추가
• 2026-09-23 — Qwen-Image-2.1 (Diffusion) 리뷰 추가 — 2.0의 첫 공개 구현 릴리스 분석

---

📂 Deep Learning

No	Title	Venue	Paper	Code	Updated
1	OpenAI Astra and Looped Transformers	Raschka Blog 2026	blog	-	2026-09-04

📂 Diffusion

No	Title	Venue	Paper	Code	Updated
36	Learning Representations to Guide Pixel Diffusion Transformers	arXiv 2026	arxiv	github	2026-10-16
35	Qwen-Image-2.1	Blog 2026	blog	github	2026-09-30
34	LLaDA-Image: Building Strong Image Generators with Fully Open Training Recipes	arXiv 2026	arxiv	github	2026-09-08
33	Mage-Flow: A 4B Image Generation and Editing Stack with Mage-VAE	arXiv 2026	arxiv	github	2026-07-29
32	DAR: Rethinking Cross-Layer Information Routing in Diffusion Transformers	arXiv 2026	arxiv	-	2026-07-15
31	SFD: Semantics Lead the Way — Harmonizing Semantic and Texture Modeling with Asynchronous Latent Diffusion	arXiv 2025	arxiv	project	2026-07-09
30	SeFi-Image: A Text-to-Image Foundation Model with Semantic-First Diffusion	arXiv 2026	arxiv	github	2026-07-06
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	2026-09-23
16	Qwen-Image Technical Report	arXiv 2025	arxiv	github	2026-07-23
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/VLM을 백본으로 재사용해, 하나의 모델이 이미지 이해와 생성을 모두 하는 unified 계열.

No	Title	Venue	Paper	Code	Updated
3	SenseNova-U1: Unifying Multimodal Understanding and Generation with NEO-unify Architecture	arXiv 2026	arxiv	github	2026-07-06
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
10	Mage-Flow: A 4B Image Generation and Editing Stack with Mage-VAE	arXiv 2026	arxiv	github	2026-07-23
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	2026-09-23
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
6	Two Halves are More than One: Phase-wise Velocity Distillation for Fast and High-Quality Image Generation	arXiv 2026	arxiv	github	2026-10-27
5	ArcFlow: Unleashing 2-Step Text-to-Image Generation via High-Precision Non-Linear Flow Distillation	arXiv 2026	arxiv	github	2026-10-26
4	DanceOPD: On-Policy Generative Field Distillation	arXiv 2026	arxiv	github	2026-10-26
3	DiffusionOPD: A Unified Perspective of On-Policy Distillation in Diffusion Models	SIGGRAPH Asia 2026	arxiv	-	2026-10-26
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

📂 LLM

No	Title	Venue	Paper	Code	Updated
1	MiniMind — Train a Tiny LLM from Scratch (Educational Repo)	GitHub 2026	-	github	2026-09-22

📂 VLM

No	Title	Venue	Paper	Code	Updated
19	GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed	arXiv 2026	arxiv	github	2026-10-26
18	EmbeddingGemma 2	Blog 2026	blog	hf	2026-10-16
17	LocateAnything: Fast and High-Quality Vision-Language Grounding with Parallel Box Decoding	ECCV 2026	paper	github	2026-09-30
16	Scaffolding Minds: Optimizing Latent Visual Target Representations for Multimodal Reasoning	arXiv 2026	arxiv	-	2026-09-14
15	Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution	arXiv 2024	arxiv	github	2026-09-08
14	Qwen2.5-VL Technical Report	arXiv 2025	arxiv	github	2026-09-08
13	Qwen-Drive-1.0: An Initial Step towards a Vision-Language Foundation Model for Autonomous Driving	arXiv 2026	arxiv	github	2026-09-08
12	Penguin-VL: Exploring the Efficiency Limits of VLM with LLM-based Vision Encoders	arXiv 2026	arxiv	github	2026-08-24
11	Qwen3-VL Technical Report	arXiv 2025	arxiv	github	2026-09-10
10	LFM2 Technical Report	arXiv 2025	arxiv	-	2026-07-29
9	BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models	ICML 2023	arxiv	github	2026-07-29
8	Florence-2: Advancing a Unified Representation for a Variety of Vision Tasks	CVPR 2024	arxiv	-	2026-07-29
7	nanoVLM: A Lightweight Vision Language Model Toolkit	HF Blog 2025	blog	github	2026-07-22
6	SmolVLM: Redefining small and efficient multimodal models	arXiv 2025	arxiv	hf	2026-07-15
5	PaliGemma 2: A Family of Versatile VLMs for Transfer	arXiv 2024	arxiv	github	2026-07-22
4	PaliGemma: A versatile 3B VLM for transfer	arXiv 2024	arxiv	-	2026-07-15
3	Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs	arXiv 2025	arxiv	-	2026-07-09
2	Gemma 3 Technical Report	arXiv 2025	arxiv	-	2026-07-09
1	Gemma 4 Technical Report	arXiv 2026	arxiv	-	2026-08-28

📂 VLM-OCR

No	Title	Venue	Paper	Code	Updated
7	Multimodal OCR: Parse Anything from Documents	arXiv 2026	arxiv	github	2026-09-08
6	MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsing	arXiv 2025	arxiv	github	2026-08-13
5	ColPali: Efficient Document Retrieval with Vision Language Models	ICLR 2025	arxiv	github	2026-08-04
4	PaddleOCR-VL-1.6: Expanding the Frontier of Document Parsing with Under-Optimized Region Refinement and Progressive Post-Training	arXiv 2026	arxiv	github	2026-08-13
3	MinerU2.5-Pro: Pushing the Limits of Data-Centric Document Parsing at Scale	arXiv 2026	arxiv	github	2026-08-13
2	GLM-OCR Technical Report	arXiv 2026	arxiv	github	2026-08-13
1	DeepSeek-OCR: Contexts Optical Compression	arXiv 2025	arxiv	github	2026-08-04

📂 OCR

No	Title	Venue	Paper	Code	Updated
2	TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models	AAAI 2023	arxiv	github	2026-07-31
1	CRAFT: Character Region Awareness for Text Detection	CVPR 2019	arxiv	github	2026-07-29

📂 3D

No	Title	Venue	Paper	Code	Updated
9	R³: 3D Reconstruction via Relative Regression	arXiv 2026	arxiv	github	2026-09-04
8	Revisiting Local Context for Long-Horizon Streaming 3D Reconstruction	arXiv 2026	arxiv	github	2026-09-04
7	4DAnyone: Create Anyone in 4D from a Casual Monocular Video	SIGGRAPH Asia 2026	arxiv	github	2026-09-02
6	DepthLab: From Partial to Complete	arXiv 2024	arxiv	github	2026-09-02
5	DUSt3R: Geometric 3D Vision Made Easy	CVPR 2024	arxiv	github	2026-08-24
4	Multi-view Reconstruction via SfM-guided Monocular Depth Estimation	CVPR 2025 (Oral)	arxiv	github	2026-08-24
3	SAM 3D Body: Robust Full-Body Human Mesh Recovery	arXiv 2026	arxiv	github	2026-08-04
2	Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data	CVPR 2024	arxiv	github	2026-07-31
1	Depth Anything V2	NeurIPS 2024	arxiv	github	2026-07-31

📂 Recommendation

No	Title	Venue	Paper	Code	Updated
2	Text-Conditioned Outfit Recommendation with Hybrid Attention Layer	IEEE Access 2023	paper	github	2026-09-04
1	Fashion Outfit Complementary Item Retrieval	CVPR 2020	paper	-	2026-09-04

📂 Forecasting

No	Title	Venue	Paper	Code	Updated
6	SAMformer: Unlocking the Potential of Transformers in Time Series Forecasting with Sharpness-Aware Minimization and Channel-Wise Attention	ICML 2024 (Oral)	arxiv	github	2026-09-22
5	TimesFM-3: A zero-shot foundation model for multivariate forecasting	Google Blog 2026	blog	github	2026-09-14
4	Chronos: Learning the Language of Time Series	TMLR 2024	arxiv	github	2026-09-08
3	A decoder-only foundation model for time-series forecasting	ICML 2024	arxiv	github	2026-09-14
2	iTransformer: Inverted Transformers Are Effective for Time Series Forecasting	ICLR 2024 (Spotlight)	arxiv	github	2026-09-08
1	A Time Series is Worth 64 Words: Long-term Forecasting with Transformers	ICLR 2023	arxiv	github	2026-09-08