WildBox provides over 237k 3D wildlife annotations from drone video and benchmarks reveal zero-shot 3D detection at 0 AP but fine-tuned performance of 8.68 AP-BEV and 13.17 AP3D, with depth estimation causing most errors.
super hub Canonical reference
Emogen: Emotional image content generation with text-to-image diffusion models
Canonical reference. 91% of citing Pith papers cite this work as background.
279
Pith papers citing it
Background
91% of classified citations
hub tools
citation-role summary
background 83
dataset 6
baseline 2
method 2
citation-polarity summary
co-cited works
representative citing papers
MoHallBench is a new benchmark evaluating motion hallucination in VideoLLMs from co-occurrence priors, sequential inference, and similarity confusion, revealing decoupling from action recognition performance.
SpheRoPE modifies rotary position embeddings in diffusion transformers to enforce spherical topology for zero-shot 360 panorama generation across multiple backbones.
RESOLVE provides a controlled multi-resolution LiDAR and camera benchmark for evaluating 3D detection and tracking under point sparsity variations in roadside cooperative perception.
A method to decompose 3D Gaussian splats into independent albedo and shading components for consistent texture editing in radiance fields.
An asynchronous architecture decouples incremental voxel-based mapping from VLM-based semantic enrichment to produce queryable open-vocabulary 3D scene graphs that match or exceed prior methods on segmentation and grounding benchmarks.
MLLMs drop from over 85% accuracy on action presence to under 50% on matched action-denial videos, exposing a causal verification gap that causal graph prompts partially close.
A regularization technique that treats diffusion model outputs as a similarity kernel during material optimization in inverse rendering, enabling joint reconstruction of geometry, materials, and illumination that satisfies the rendering equation and generalizes to new lighting.
Introduces VG-GUIBench benchmark and TASKER keyframe extraction algorithm that improves performance on VideoQA and video-guided agentic tasks.
ScaLe-INR is a multi-branch INR architecture that applies directional scaling per the Fourier inverse theorem and a directional edge guidance loss to disentangle scales and improve reconstruction fidelity.
MATCH is the first flow matching method for multi-view anomaly detection, reporting SOTA results on Real-IAD and the first comprehensive evaluation on MANTA-Tiny while enabling real-time use by omitting the divergence term.
GeoFidelity-Bench shows text-to-image models gain city-level plausibility from local names but achieve near-zero improvement in exact segment identity, with GPS coordinates adding no benefit.
Arbor attaches constraint mesh tokens to a frozen text-to-3D denoiser to enable controllable generation obeying hull, avoidance, and touch constraints.
Target dynamics provide an intrinsic source of variation equivalent to controlled illumination changes, enabling scattering-compensated reconstruction of dynamic scenes with one acquisition per frame in holographic and fluorescence imaging.
The paper defines the 4DVLT task for worldline-centered 4D scene understanding, releases Instruct-4D with 129.4K QA pairs, and presents 4DTrack achieving 62.68 TGA_Top1, outperforming adapted baselines by 19.62 points.
FLM-Occ reformulates indoor occupancy prediction as feed-forward likelihood maximization over a mixture model with volume-normalized weights, achieving superior accuracy on Occ-ScanNet using only 32 superquadrics.
HERO maps DNA methylation and miRNA to a 16-dimensional intent vector for TF-IDF caption retrieval and cosine-gated repair in VLM-based multi-task breast cancer prediction, claiming SOTA on TCGA-BRCA.
StylisticBias benchmark shows 15 visual attributes explain nearly 80% of bias variation in six MLLMs by isolating single cues like age and fashion in generated images.
CloudLULC-Net is an end-to-end heterogeneous SAR-optical fusion network for LULC mapping under cloud contamination that achieves 86.60% OA, 83.29% F1, and 73.51% mIoU on a new global benchmark of 40,223 samples.
A two-stage generative model (Graph CVAE + flow matching) learns topology-agnostic motion codes from a new 5k-topology dataset and retargets video motion to arbitrary unseen skeletons.
FisherAdapTune uses temporal drift in Fisher geometry, measured by scale-invariant Jensen-Shannon distance, to progressively freeze stabilized parameter groups during fine-tuning, reporting gains on segmentation and zero-shot transfer.
An ILP-based oracle applied to seven VIS methods on YouTube-VIS and OVIS shows tracking instability as the dominant bottleneck, producing gaps exceeding 20 AP under occlusion while classification impact is secondary.
Attributed Feature Graphs (AFGs) represent CAD features as attributed nodes and relations as directed edges to enable GNN surrogate models that predict design performance with feature-level interpretability on the CarHoods10K dataset.
Empirical study of five LVR variants finds cosine alignment negatively correlates with accuracy (r=-0.94), supervised latents are bypassed under corruption (max 4-point shift), and answers are decodable downstream but not at the latent.
citing papers explorer
-
WildBox: A Dataset and Benchmark for Aerial Monocular 3D Detection of African Savanna Wildlife
WildBox provides over 237k 3D wildlife annotations from drone video and benchmarks reveal zero-shot 3D detection at 0 AP but fine-tuned performance of 8.68 AP-BEV and 13.17 AP3D, with depth estimation causing most errors.
-
MoHallBench: A Benchmark for Motion Hallucination in Video Large Language Models
MoHallBench is a new benchmark evaluating motion hallucination in VideoLLMs from co-occurrence priors, sequential inference, and similarity confusion, revealing decoupling from action recognition performance.
-
SpheRoPE: Zero-Shot Optimization-Free 360 Panorama Generation with Spherical RoPE
SpheRoPE modifies rotary position embeddings in diffusion transformers to enforce spherical topology for zero-shot 360 panorama generation across multiple backbones.
-
Think While You Map: Asynchronous Vision-Language Agents for Incremental 3D Scene Graphs
An asynchronous architecture decouples incremental voxel-based mapping from VLM-based semantic enrichment to produce queryable open-vocabulary 3D scene graphs that match or exceed prior methods on segmentation and grounding benchmarks.
-
Learning to Deny: Action Denial in Multimodal Large Language Models
MLLMs drop from over 85% accuracy on action presence to under 50% on matched action-denial videos, exposing a causal verification gap that causal graph prompts partially close.
-
Diffusion-Based Material Regularization for Physics-Based Inverse Rendering
A regularization technique that treats diffusion model outputs as a similarity kernel during material optimization in inverse rendering, enabling joint reconstruction of geometry, materials, and illumination that satisfies the rendering equation and generalizes to new lighting.
-
Bridging VideoQA and Video-Guided Agentic Tasks via Generalized Keyframe Extraction
Introduces VG-GUIBench benchmark and TASKER keyframe extraction algorithm that improves performance on VideoQA and video-guided agentic tasks.
-
ScaLe-INR: Scale and Learn Implicit Neural Representations
ScaLe-INR is a multi-branch INR architecture that applies directional scaling per the Fourier inverse theorem and a directional edge guidance loss to disentangle scales and improve reconstruction fidelity.
-
MATCH: Flow Matching for Multi-View Anomaly Detection
MATCH is the first flow matching method for multi-view anomaly detection, reporting SOTA results on Real-IAD and the first comprehensive evaluation on MANTA-Tiny while enabling real-time use by omitting the divergence term.
-
GeoFidelity-Bench: Evaluating Segment-Level Geographic Fidelity in Text-to-Image Street-View Generation
GeoFidelity-Bench shows text-to-image models gain city-level plausibility from local names but achieve near-zero improvement in exact segment identity, with GPS coordinates adding no benefit.
-
Arbor: Explicit Geometric Conditioning for Controllable 3D Asset Generation
Arbor attaches constraint mesh tokens to a frozen text-to-3D denoiser to enable controllable generation obeying hull, avoidance, and touch constraints.
-
FLM-Occ: Feed-forward Likelihood Maximization for Efficient Indoor Occupancy Prediction
FLM-Occ reformulates indoor occupancy prediction as feed-forward likelihood maximization over a mixture model with volume-normalized weights, achieving superior accuracy on Occ-ScanNet using only 32 superquadrics.
-
HERO: Hypothesis-Driven Evidence Retrieval from Omics for Multi-Task Breast Cancer Analysis
HERO maps DNA methylation and miRNA to a 16-dimensional intent vector for TF-IDF caption retrieval and cosine-gated repair in VLM-based multi-task breast cancer prediction, claiming SOTA on TCGA-BRCA.
-
TopoCap: Learning Topology-Agnostic Motion Priors for Monocular Video-to-Animation
A two-stage generative model (Graph CVAE + flow matching) learns topology-agnostic motion codes from a new 5k-topology dataset and retargets video motion to arbitrary unseen skeletons.
-
Fisher-Guided Progressive Parameter Selection for Adaptive Fine-Tuning
FisherAdapTune uses temporal drift in Fisher geometry, measured by scale-invariant Jensen-Shannon distance, to progressively freeze stabilized parameter groups during fine-tuning, reporting gains on segmentation and zero-shot transfer.
-
Mind the Gap: Disentangling Performance Bottlenecks in Video Instance Segmentation
An ILP-based oracle applied to seven VIS methods on YouTube-VIS and OVIS shows tracking instability as the dominant bottleneck, producing gaps exceeding 20 AP under occlusion while classification impact is secondary.
-
TIDES: Time-Derivative Event Simulation via Deformable Reconstruction
TIDES simulates realistic event camera streams in continuous time via dynamic Gaussian splatting with adaptive occlusion handling and sensor artifact modeling, claiming SOTA fidelity and better downstream transfer than prior methods.
-
SuperMemory-VQA: An Egocentric Visual Question-Answering Benchmark for Long-Horizon Memory
SuperMemory-VQA provides 4,853 human-verified QA pairs from 52.9 hours of egocentric AI glasses recordings to benchmark AI systems on realistic long-horizon memory tasks including an unanswerable option.
-
RS2AD-LiDAR: End-to-End Autonomous Driving LiDAR Data Generation from Roadside Sensor Observations
RS2AD-LiDAR reconstructs vehicle LiDAR data from roadside observations via coordinate transformation, virtual LiDAR modeling and resampling, claimed as the first such method, with experiments showing improved object detection when mixed with real data.
-
AIGaitor: Privacy-preserving and cloud-free motion analysis for everyone, using edge computing
AIGaitor is the first claimed end-to-end on-device monocular motion-capture and deep-learning gait analysis pipeline demonstrated on consumer smartphones.
-
SDM: A Powerful Tool for Evaluating Model Robustness
SDM is a new staged gradient attack that reconstructs the adversarial objective around probability differences and reports stronger performance than prior methods like APGD.
-
LMM-Track4D: Eliciting 4D Dynamic Reasoning in LMMs via Trajectory-Grounded Dialogue
LMM-Track4D formulates a trajectory-grounded dialogue task, releases Track4D-Bench with 526 samples, and proposes RTGE encoding, TRK state token, and OSK-RA decoder to elicit better 4D spatiotemporal reasoning in LMMs.
-
HL-OutPaint: Coarse-to-Fine Video Outpainting for High-Resolution Long-Range Videos
HL-OutPaint enables high-resolution outpainting of long video sequences via a coarse-to-fine pipeline that first builds Global Coarse Guidance through global-local frame swapping then synthesizes details.
-
Pareto-Guided Optimal Transport for Multi-Reward Alignment
PG-OT builds prompt-specific Pareto frontiers and applies distribution-aware optimal transport to improve multi-reward alignment while introducing JDR and JCR metrics to measure synergy and hacking.
-
Field-Localized Forgery Detection for Digital Identity Documents
FLiD is a field-localized forgery detection method for identity documents that outperforms full-document baselines and general detectors with significantly fewer parameters.
-
AniMatrix: An Anime Video Generation Model that Thinks in Art, Not Physics
AniMatrix generates anime videos by structuring artistic production rules into a controllable taxonomy and training the model to prioritize those rules over physical realism, achieving top scores from professional animators on prompt understanding and artistic motion.
-
Does it Really Count? Assessing Semantic Grounding in Text-Guided Class-Agnostic Counting
Text-guided class-agnostic counting models exhibit significant weaknesses in grounding textual prompts to visual objects, as demonstrated by new negative-label and distractor tests on a multi-category dataset.
-
MIRL: Mutual Information-Guided Reinforcement Learning for Vision-Language Models
MIRL uses mutual information to guide trajectory selection and provide separate rewards for visual perception in RLVR for VLMs, achieving 70.22% average accuracy with 25% fewer full trajectories.
-
CSGuard: Toward Forgery-Resistant Watermarking in Diffusion Models via Compressed Sensing Constraint
CSGuard binds diffusion-model watermarks to a secret matrix via compressed sensing, cutting forgery attack success from 100% to 28.12% while preserving 100% detection on legitimate images.
-
Towards Temporal Compositional Reasoning in Long-Form Sports Videos
SportsTime benchmark and CoTR method improve multimodal AI's temporal compositional reasoning and evidence grounding in long-form sports videos.
-
HumanScore: Benchmarking Human Motions in Generated Videos
HumanScore defines six metrics for kinematic plausibility, temporal stability, and biomechanical consistency to benchmark human motions in videos from thirteen state-of-the-art generation models, revealing gaps between visual appeal and physical fidelity.
-
Divide-and-Conquer Approach to Holistic Cognition in High-Similarity Contexts with Limited Data
DHCNet improves ultra-fine-grained visual categorization by progressively building holistic cognition from local discrepancies using self-shuffling and refinement on limited data.
-
Towards Symmetry-sensitive Pose Estimation: A Rotation Representation for Symmetric Object Classes
SARR modifies trigonometric rotation encodings with object symmetry orders to produce unique continuous poses, enabling standard CNNs to outperform existing methods on symmetry-aware 6D pose estimation without custom losses or 3D models.
-
Efficient Video Diffusion Models: Advancements and Challenges
A survey that groups efficient video diffusion methods into four paradigms—step distillation, efficient attention, model compression, and cache/trajectory optimization—and outlines open challenges for practical use.
-
DPC-VQA: Decoupling Quality Perception and Residual Calibration for Video Quality Assessment
DPC-VQA decouples a frozen MLLM perceptual prior from a lightweight residual calibration branch to adapt video quality assessment to new scenarios with under 2% trainable parameters and 20% of typical MOS labels.
-
Beyond Perception Errors: Semantic Fixation in Large Vision-Language Models
VLMs display semantic fixation, with higher accuracy on standard rule mappings than inverse ones across 14 models, narrowed by neutral prompts but widened by loaded ones and affected by post-training alignment.
-
DetailVerifyBench: A Benchmark for Dense Hallucination Localization in Long Image Captions
DetailVerifyBench supplies 1,000 images and densely annotated long captions to evaluate precise hallucination localization in multimodal large language models.
-
From Plausibility to Verifiability: Risk-Controlled Generative OCR with Vision-Language Models
A model-agnostic Geometric Risk Controller reduces extreme errors in VLM-based OCR by requiring cross-view consensus before accepting outputs.
-
Zero-shot Human Pose Estimation using Diffusion-based Inverse solvers
InPose formulates pose estimation as an inverse problem solved by guiding a rotation-conditioned diffusion prior with a location-based likelihood term for zero-shot generalization across users.
-
CamPVG: Camera-Controlled Panoramic Video Generation with Epipolar-Aware Diffusion
CamPVG is the first diffusion-based framework for generating geometrically consistent panoramic videos from camera pose inputs using a panoramic Plücker embedding and spherical epipolar attention module.
-
FaSTA$^*$: Fast-Slow Toolpath Agent with Subroutine Mining for Efficient Multi-turn Image Editing
FaSTA* combines LLM fast planning with A* search and inductive subroutine mining to create an efficient agent for multi-turn image editing tasks.
-
BEVCALIB: LiDAR-Camera Calibration via Geometry-Guided Bird's-Eye View Representations
BEVCALIB performs LiDAR-camera calibration from raw data by fusing camera and LiDAR bird's-eye view features with a novel feature selector and reports state-of-the-art accuracy on KITTI and NuScenes.
-
Fleet: Few Shots Lead Effective AI-generated Image Detection
Fleet achieves dynamic few-shot adaptation for AIGI detection via avoidance routing in decoupled subspaces, raising accuracy from 20.4% to 73.1% on new generators like Doubao Seedream 4.0 with 10 shots.
-
Anomaly Factory 3D: A Modular Framework for Diverse Pseudo-Anomaly Synthesis in Unsupervised 3D Anomaly Detection
AF3AD is a modular synthesis framework using center-conditioned parametric deformations in local PCA frames to create diverse pseudo-anomalies, improving unsupervised 3D anomaly detection on AnomalyShapeNet and Real3D-AD.
-
OSOR: One-Step Diffusion Inpainting for Effect-Aware Object Removal
OSOR is a one-step diffusion inpainting method using an occupancy-guided discriminator, alpha head, and semantic-anchored verification pipeline to achieve effect-aware object removal, outperforming multi-step baselines in quality at 4-30x speed.
-
HarmVideoBench: Benchmarking Harmful Video Understanding in Large Multimodal Models
HarmVideoBench is a multi-layered benchmark for harmful video understanding in LVLMs with three hierarchical dimensions, and BCR is a method that raises average model performance from 61.7% to 84.4%.
-
HANCLIP: A Family of Hyperbolic Angular Negation Vision Language Models
HANCLIP restructures VLM embeddings with hyperbolic space and angular negation objectives to raise negation sensitivity on NegBench while keeping standard retrieval and classification performance.
-
Interpretable Uncertainty Routing Separating Emotion Ambiguity from Distribution Shift in Facial Expression Recognition
Uncertainty decomposition via deep ensembles separates annotator disagreement from distribution shift in FER, enabling a routing mechanism that retains 1.8x more ambiguous faces at matched OOD rejection compared to single-uncertainty baselines.
-
Spectral Query-Key Product Weight Steering for Training-Free VLM Hallucination Mitigation
QK Product Steering suppresses dominant singular modes in the per-head QK product of selected middle layers via a closed-form query-only update, yielding 4.0% average relative CHAIR_s reduction on three GQA VLMs.
-
FATE: Pillar Encoding and Frequency-Aware Training for Event-Based Object Detection
FATE combines pillar encoding via orthogonal polynomial basis with frequency-aware training to enable event-based object detection at up to 200 Hz without internal temporal sub-binning.