ORBIS uses output-guided token reduction and DATM to achieve 2x higher token reduction than AsymRnR, with up to 4.5x speedup and 79.3% energy savings versus A100 GPU for video DiT models.
Mixed citations
Title resolution pending
Mixed citation behavior. Most common role is background (64%).
35
Pith papers citing it
Background
64% of classified citations
citation-role summary
background 7
baseline 1
dataset 1
method 1
other 1
citation-polarity summary
representative citing papers
ShadeBench is a multimodal benchmark dataset for urban shade understanding that includes temporally varying shade maps, satellite imagery, building representations, and text to support shade generation, segmentation, and 3D reconstruction tasks.
CrossMPI steers both visual and textual interpretations in LVLMs through image-only perturbations by optimizing in hidden-state space at selected middle layers with distance-based budget allocation.
A method using attention head vectors detects and suppresses risky content generation in Diffusion Transformers at inference time.
LEGO uses multiple generator-specific LoRA modules modulated by an MLP and fused with attention to detect synthetic images, achieving better performance than prior methods while using under 10% of the training data.
ResetEdit embeds a recoverable discrepancy signal during image generation in diffusion models to reconstruct an approximate original latent for high-fidelity text-guided editing.
Hallo-Live achieves 20.38 FPS real-time text-to-audio-video avatar generation with 0.94s latency using asynchronous dual-stream diffusion and HP-DMD preference distillation, matching teacher model quality at 16x higher throughput.
FlowAnchor stabilizes editing signals in flow-based inversion-free video editing via spatial-aware attention refinement and adaptive magnitude modulation for improved faithfulness and temporal coherence.
UniEditBench unifies image and video editing evaluation with a nine-plus-eight operation taxonomy and cost-effective 4B/8B distilled MLLM evaluators that align with human judgments.
IAD-Unify unifies industrial anomaly segmentation, region-grounded language understanding, and mask-guided generation in one framework using DINOv2 token injection into Qwen3.5, supported by the new Anomaly-56K dataset of 59,916 images.
MAST is a mask-guided attention allocation method that enables artifact-free multi-style transfer in diffusion models by anchoring layout, distributing attention mass, scaling sharpness, and injecting details.
SCOPE accelerates autoregressive video diffusion up to 4.73x by using a tri-modal cache-predict-recompute scheduler with Taylor extrapolation and selective active-frame computation while preserving output quality.
SHARP applies a spectrum-aware dynamic RoPE scaling schedule that promotes resolution more strongly in early denoising stages and relaxes it later, outperforming static baselines on quality metrics for remote sensing images.
Fully aligned instructional videos for physical tasks yield 11.1% better completion quality and 15.5% faster times, with four decomposable visual attributes whose isolated misalignments degrade performance without users noticing.
ClickRemoval delivers click-driven object removal and background restoration in diffusion models through self-attention modulation without additional training or inputs.
Post-generation control in AI-assisted math visual creation yields higher teacher ratings for predictability and correctness than pre- or mid-generation control, with qualitative trade-offs in agency and effort.
SpatialFusion internalizes 3D geometric awareness into unified image generation models by pairing an MLLM with a spatial transformer that produces depth maps to constrain diffusion generation.
A latent denoising objective with saliency-aware corruption and contrastive distillation improves visual alignment and corruption robustness in large multimodal models.
Task-aware localization via attention cues and feature centroids from source/target streams in IIE models improves non-edit consistency while preserving instruction following.
FASA bridges low-level forensic frequency signals and high-level semantic consistency to achieve state-of-the-art localization of both conventional and diffusion-generated image manipulations.
Precise Shield identifies safety neurons in VLLMs via activation contrasts and aligns only them with gradient masking, boosting safety, preserving generalization, and enabling zero-shot cross-lingual and cross-modal transfer.
VersaVogue unifies garment generation and virtual dressing via trait-routing attention with mixture-of-experts and an automated multi-perspective preference optimization pipeline that uses DPO without human labels.
CAGE uses LLM-generated code for label-correct diagrams followed by ControlNet-conditioned diffusion refinement to produce both accurate and visually engaging educational graphics, backed by the new EduDiagram-2K dataset.
InsTraj generates realistic, instruction-faithful GPS trajectories by using an LLM to parse natural-language travel intent and a multimodal diffusion transformer to produce the paths.
citing papers explorer
-
ORBIS: Output-Guided Token Reduction with Distribution-Aware Matching for Video Diffusion Acceleration
ORBIS uses output-guided token reduction and DATM to achieve 2x higher token reduction than AsymRnR, with up to 4.5x speedup and 79.3% energy savings versus A100 GPU for video DiT models.
-
ShadeBench: A Benchmark Dataset for Building Shade Simulation in Sustainable Society
ShadeBench is a multimodal benchmark dataset for urban shade understanding that includes temporally varying shade maps, satellite imagery, building representations, and text to support shade generation, segmentation, and 3D reconstruction tasks.
-
What Concepts Lie Within? Detecting and Suppressing Risky Content in Diffusion Transformers
A method using attention head vectors detects and suppresses risky content generation in Diffusion Transformers at inference time.
-
LEGO: LoRA-Enabled Generator-Oriented Framework for Synthetic Image Detection
LEGO uses multiple generator-specific LoRA modules modulated by an MLP and fused with attention to detect synthetic images, achieving better performance than prior methods while using under 10% of the training data.
-
ResetEdit: Precise Text-guided Editing of Generated Image via Resettable Starting Latent
ResetEdit embeds a recoverable discrepancy signal during image generation in diffusion models to reconstruct an approximate original latent for high-fidelity text-guided editing.
-
Hallo-Live: Real-Time Streaming Joint Audio-Video Avatar Generation with Asynchronous Dual-Stream and Human-Centric Preference Distillation
Hallo-Live achieves 20.38 FPS real-time text-to-audio-video avatar generation with 0.94s latency using asynchronous dual-stream diffusion and HP-DMD preference distillation, matching teacher model quality at 16x higher throughput.
-
FlowAnchor: Stabilizing the Editing Signal for Inversion-Free Video Editing
FlowAnchor stabilizes editing signals in flow-based inversion-free video editing via spatial-aware attention refinement and adaptive magnitude modulation for improved faithfulness and temporal coherence.
-
UniEditBench: A Unified and Cost-Effective Benchmark for Image and Video Editing via Distilled MLLMs
UniEditBench unifies image and video editing evaluation with a nine-plus-eight operation taxonomy and cost-effective 4B/8B distilled MLLM evaluators that align with human judgments.
-
IAD-Unify: A Region-Grounded Unified Model for Industrial Anomaly Segmentation, Understanding, and Generation
IAD-Unify unifies industrial anomaly segmentation, region-grounded language understanding, and mask-guided generation in one framework using DINOv2 token injection into Qwen3.5, supported by the new Anomaly-56K dataset of 59,916 images.
-
MAST: Mask-Guided Attention Mass Allocation for Training-Free Multi-Style Transfer
MAST is a mask-guided attention allocation method that enables artifact-free multi-style transfer in diffusion models by anchoring layout, distributing attention mass, scaling sharpness, and injecting details.
-
Not All Frames Deserve Full Computation: Accelerating Autoregressive Video Generation via Selective Computation and Predictive Extrapolation
SCOPE accelerates autoregressive video diffusion up to 4.73x by using a tri-modal cache-predict-recompute scheduler with Taylor extrapolation and selective active-frame computation while preserving output quality.
-
SHARP: Spectrum-aware Highly-dynamic Adaptation for Resolution Promotion in Remote Sensing Synthesis
SHARP applies a spectrum-aware dynamic RoPE scaling schedule that promotes resolution more strongly in early denoising stages and relaxes it later, outperforming static baselines on quality metrics for remote sensing images.
-
ClickRemoval: An Interactive Open-Source Tool for Object Removal in Diffusion Models
ClickRemoval delivers click-driven object removal and background restoration in diffusion models through self-attention modulation without additional training or inputs.
-
SpatialFusion: Endowing Unified Image Generation with Intrinsic 3D Geometric Awareness
SpatialFusion internalizes 3D geometric awareness into unified image generation models by pairing an MLLM with a spatial transformer that produces depth maps to constrain diffusion generation.
-
Latent Denoising Improves Visual Alignment in Large Multimodal Models
A latent denoising objective with saliency-aware corruption and contrastive distillation improves visual alignment and corruption robustness in large multimodal models.
-
Rethinking Where to Edit: Task-Aware Localization for Instruction-Based Image Editing
Task-aware localization via attention cues and feature centroids from source/target streams in IIE models improves non-edit consistency while preserving instruction following.
-
Bridging the Micro--Macro Gap: Frequency-Aware Semantic Alignment for Image Manipulation Localization
FASA bridges low-level forensic frequency signals and high-level semantic consistency to achieve state-of-the-art localization of both conventional and diffusion-generated image manipulations.
-
Precise Shield: Explaining and Aligning VLLM Safety via Neuron-Level Guidance
Precise Shield identifies safety neurons in VLLMs via activation contrasts and aligns only them with gradient masking, boosting safety, preserving generalization, and enabling zero-shot cross-lingual and cross-modal transfer.
-
VersaVogue: Visual Expert Orchestration and Preference Alignment for Unified Fashion Synthesis
VersaVogue unifies garment generation and virtual dressing via trait-routing attention with mixture-of-experts and an automated multi-perspective preference optimization pipeline that uses DPO without human labels.
-
CAGE: Bridging the Accuracy-Aesthetics Gap in Educational Diagrams via Code-Anchored Generative Enhancement
CAGE uses LLM-generated code for label-correct diagrams followed by ControlNet-conditioned diffusion refinement to produce both accurate and visually engaging educational graphics, backed by the new EduDiagram-2K dataset.
-
Hallucination of Multimodal Large Language Models: A Survey
The survey organizes causes of hallucinations in MLLMs, reviews evaluation benchmarks and metrics, and outlines mitigation approaches plus open questions.
-
Face-D(^2)CL: Multi-Domain Synergistic Representation with Dual Continual Learning for Facial DeepFake Detection
Face-D²CL fuses spatial and frequency features and uses dual continual learning to reduce forgetting while adapting to new DeepFakes, cutting average error rates by 60.7% and raising unseen-domain AUC by 7.9% over prior SOTA.