Stateful visual encoders condition each visual representation on prior features, yielding consistent gains on multi-image tasks under supervised finetuning across model sizes and domains.
hub Mixed citations
Editscore: Unlocking online rl for image editing via high-fidelity reward modeling
Mixed citation behavior. Most common role is background (62%).
21
Pith papers citing it
Background
62% of classified citations
hub tools
citation-role summary
background 5
baseline 1
method 1
other 1
citation-polarity summary
representative citing papers
A planner-orchestrator system learns long-horizon image editing by maximizing outcome-based rewards from a vision-language judge and refining plans from successful trajectories.
Edit-Compass and EditReward-Compass are new unified benchmarks for fine-grained image editing evaluation and realistic reward modeling in reinforcement learning optimization.
RewardHarness self-evolves a tool-and-skill library from 100 preference examples to reach 47.4% accuracy on image-edit evaluation, beating GPT-5, and yields stronger RL-tuned models.
EditRefiner uses a perception-reasoning-action-evaluation agent loop and the EditFHF-15K human feedback dataset to refine text-guided image edits more accurately than prior methods.
Sparkle supplies a large-scale dataset and benchmark for instruction-driven video background replacement, enabling models that generate more natural and temporally consistent new scenes than earlier approaches.
ParetoSlider conditions diffusion models on continuous preference weights to approximate the full Pareto front, providing dynamic control over multi-objective rewards at inference time.
Formalizes Fashion Detail Generation task, releases FDBench benchmark with 40K+ pairs, and proposes CFAD distillation method plus RL consistency reward that outperforms open-source baselines.
SpatialFlow-GRPO adds region-level reward feedback and spatial alignment to Flow-GRPO-style RL for image editing, reporting gains on GEdit-Bench, ImgEdit-Bench, and a new MultiEditBench.
Auto-Rubric as Reward externalizes VLM preferences into structured rubrics and applies Rubric Policy Optimization to create more reliable binary rewards for multimodal generation, outperforming pairwise models on text-to-image and editing benchmarks.
Edit-R1 builds a CoT-based reasoning reward model (RRM) via SFT and GCPO, then applies it with GRPO to improve image editing models such as FLUX.1-kontext.
DDA-Thinker decouples planning from generation and applies dual-atomic RL with checklist-based rewards to boost reasoning in image editing, yielding competitive results on RISE-Bench and KRIS-Bench.
RC-GRPO-Editing constrains GRPO exploration to editing regions via localized noise and attention rewards, improving instruction adherence and non-target preservation in flow-based image editing.
InsEdit adapts a video diffusion backbone for text-instruction video editing via Mutual Context Attention, achieving SOTA open-source results with O(100K) data while also supporting image editing.
A 235B VLM trained with human-refined SFT and hardness-adaptive error-aware DPO cuts critical instruction errors from ~48% to ~18% and beats Gemini-3-Pro on three editing-instruction benchmarks.
ConsistencySolver enables high-quality low-step diffusion previews by adapting general linear multistep methods into a lightweight RL-optimized solver, matching multistep DPM-Solver FID with 47% fewer steps and cutting user interaction time by nearly 50%.
UniWorld-V2 applies policy optimization via DiffusionNFT and MLLM logit feedback with group filtering to reach state-of-the-art scores of 4.49 on ImgEdit and 7.83 on GEdit-Bench while remaining model-agnostic.
DivRL introduces negative self-similarity measure and visual semantic matching with an explore-and-suppress gated strategy to jointly improve structural diversity and identity consistency in subject-driven generation.
Introduces ProductConsistency dataset, benchmark, and Cyclic Consistency reward to fine-tune image editing models, achieving a 5x reduction in character error rate for product identity preservation.
FineEdit adds multi-level bounding box injection to diffusion image editing, releases a 1.2M-pair dataset with box annotations, and shows better instruction following and background consistency than prior open models on new and existing benchmarks.
JoyAI-Image unifies visual understanding and generation via an MLLM-MMDiT architecture with spatial training signals to reach competitive benchmark performance and stronger spatial intelligence.
citing papers explorer
-
Stateful Visual Encoders for Vision-Language Models
Stateful visual encoders condition each visual representation on prior features, yielding consistent gains on multi-image tasks under supervised finetuning across model sizes and domains.
-
From Plans to Pixels: Learning to Plan and Orchestrate for Open-Ended Image Editing
A planner-orchestrator system learns long-horizon image editing by maximizing outcome-based rewards from a vision-language judge and refining plans from successful trajectories.
-
Edit-Compass & EditReward-Compass: A Unified Benchmark for Image Editing and Reward Modeling
Edit-Compass and EditReward-Compass are new unified benchmarks for fine-grained image editing evaluation and realistic reward modeling in reinforcement learning optimization.
-
RewardHarness: Self-Evolving Agentic Post-Training
RewardHarness self-evolves a tool-and-skill library from 100 preference examples to reach 47.4% accuracy on image-edit evaluation, beating GPT-5, and yields stronger RL-tuned models.
-
EditRefiner: A Human-Aligned Agentic Framework for Image Editing Refinement
EditRefiner uses a perception-reasoning-action-evaluation agent loop and the EditFHF-15K human feedback dataset to refine text-guided image edits more accurately than prior methods.
-
Sparkle: Realizing Lively Instruction-Guided Video Background Replacement via Decoupled Guidance
Sparkle supplies a large-scale dataset and benchmark for instruction-driven video background replacement, enabling models that generate more natural and temporally consistent new scenes than earlier approaches.
-
ParetoSlider: Diffusion Models Post-Training for Continuous Reward Control
ParetoSlider conditions diffusion models on continuous preference weights to approximate the full Pareto front, providing dynamic control over multi-objective rewards at inference time.
-
DetailAnywhere: Fashion Detail Generation via Cross-Modal Feature Alignment Distillation
Formalizes Fashion Detail Generation task, releases FDBench benchmark with 40K+ pairs, and proposes CFAD distillation method plus RL consistency reward that outperforms open-source baselines.
-
SpatialFlow-GRPO: Where Spatial Credit Drives Image Editing
SpatialFlow-GRPO adds region-level reward feedback and spatial alignment to Flow-GRPO-style RL for image editing, reporting gains on GEdit-Bench, ImgEdit-Bench, and a new MultiEditBench.
-
Auto-Rubric as Reward: From Implicit Preferences to Explicit Multimodal Generative Criteria
Auto-Rubric as Reward externalizes VLM preferences into structured rubrics and applies Rubric Policy Optimization to create more reliable binary rewards for multimodal generation, outperforming pairwise models on text-to-image and editing benchmarks.
-
Leveraging Verifier-Based Reinforcement Learning in Image Editing
Edit-R1 builds a CoT-based reasoning reward model (RRM) via SFT and GCPO, then applies it with GRPO to improve image editing models such as FLUX.1-kontext.
-
DDA-Thinker: Decoupled Dual-Atomic Reinforcement Learning for Reasoning-Driven Image Editing
DDA-Thinker decouples planning from generation and applies dual-atomic RL with checklist-based rewards to boost reasoning in image editing, yielding competitive results on RISE-Bench and KRIS-Bench.
-
Region-Constrained Group Relative Policy Optimization for Flow-Based Image Editing
RC-GRPO-Editing constrains GRPO exploration to editing regions via localized noise and attention rewards, improving instruction adherence and non-target preservation in flow-based image editing.
-
InsEdit: Towards Instruction-based Visual Editing via Data-Efficient Video Diffusion Models Adaptation
InsEdit adapts a video diffusion backbone for text-instruction video editing via Mutual Context Attention, achieving SOTA open-source results with O(100K) data while also supporting image editing.
-
EditCaption: Human-Refined SFT and HAE-DPO for Image Editing Instruction Synthesis
A 235B VLM trained with human-refined SFT and hardness-adaptive error-aware DPO cuts critical instruction errors from ~48% to ~18% and beats Gemini-3-Pro on three editing-instruction benchmarks.
-
Image Diffusion Preview with Consistency Solver
ConsistencySolver enables high-quality low-step diffusion previews by adapting general linear multistep methods into a lightweight RL-optimized solver, matching multistep DPM-Solver FID with 47% fewer steps and cutting user interaction time by nearly 50%.
-
Uniworld-V2: Reinforce Image Editing with Diffusion Negative-aware Finetuning and MLLM Implicit Feedback
UniWorld-V2 applies policy optimization via DiffusionNFT and MLLM logit feedback with group filtering to reach state-of-the-art scores of 4.49 on ImgEdit and 7.83 on GEdit-Bench while remaining model-agnostic.
-
DivRL: Disentangled Self-Similarity Rewards for Diverse Subject-Driven Generation
DivRL introduces negative self-similarity measure and visual semantic matching with an explore-and-suppress gated strategy to jointly improve structural diversity and identity consistency in subject-driven generation.
-
ProductConsistency: Improving Product Identity Preservation in Instruction-Based Image Editing via SFT and RL
Introduces ProductConsistency dataset, benchmark, and Cyclic Consistency reward to fine-tune image editing models, achieving a 5x reduction in character error rate for product identity preservation.
-
FineEdit: Fine-Grained Image Edit with Bounding Box Guidance
FineEdit adds multi-level bounding box injection to diffusion image editing, releases a 1.2M-pair dataset with box annotations, and shows better instruction following and background consistency than prior open models on new and existing benchmarks.
-
JoyAI-Image: Awaking Spatial Intelligence in Unified Multimodal Understanding and Generation
JoyAI-Image unifies visual understanding and generation via an MLLM-MMDiT architecture with spatial training signals to reach competitive benchmark performance and stronger spatial intelligence.