Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Suggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works144 from public data
- The Sixth Visual Object Tracking VOT2018 Challenge Results804
The Visual Object Tracking challenge VOT2018 is the sixth annual tracker benchmarking activity organized by the VOT initiative; results of over eighty trackers are presented; many are state-of-the-art trackers published at major computer vision conferences or in journals in the recent years.
- The Visual Object Tracking VOT2016 Challenge Results776
The Visual Object Tracking challenge VOT2016 goes beyond its predecessors by introducing a new semi-automatic ground truth bounding box annotation methodology and extending the evaluation system with the no-reset experiment.
- 424
- RT-GENE: Real-Time Eye Gaze Estimation in Natural Environments412
This work addresses the issue of ground truth annotation by measuring head pose using a motion capture system and eye gaze using mobile eyetracking glasses and applies semantic image inpainting to the area covered by the glasses to bridge the gap between training and testing images by removing the obtrusiveness of the glasses.
- 311
- Symmetric Graph Convolutional Autoencoder for Unsupervised Graph Representation Learning291
A new numerically stable form of the Laplacian sharpening by incorporating the signed graphs is proposed and a new cost function which finds a latent representation and a latent affinity matrix simultaneously is devised to boost the performance of image clustering tasks.
- FixBi: Bridging Domain Spaces for Unsupervised Domain Adaptation275
A fixed ratio-based mixup is introduced to augment multiple intermediate domains between the source and target domain and gradually transfer domain knowledge from the source to the target domain.
- Context-Aware Deep Feature Compression for High-Speed Visual Tracking229
A new context-aware correlation filter based tracking framework to achieve both high computational speed and state-of-the-art performance among real-time trackers and introduces extrinsic denoising processes and a new orthogonality loss term for pre-training and fine-tuning of the expert autoencoders.
- FS-Net: Fast Shape-based Network for Category-Level 6D Object Pose Estimation with Decoupled Rotation Mechanism221
A fast shape-based network (FS-Net) with efficient category-level feature extraction for 6D pose estimation and a novel decoupled rotation mechanism that employs two decoders to complementarily access the rotation information.
- 189
- 130
- The Ninth Visual Object Tracking VOT2021 Challenge Results127
The Visual Object Tracking challenge VOT2021 is the ninth annual tracker benchmarking activity organized by the VOT initiative; results of 71 trackers are presented; many are state-of-the-art trackers published at major computer vision conferences or in journals in recent years.
- G2L-Net: Global to Local Network for Real-Time 6D Pose Estimation With Embedding Vector Features121
A novel real-time 6D object pose estimation framework, named G2L-Net, which operates on point clouds from RGB-D detection in a divide-and-conquer fashion and achieves state-of-the-art performance in terms of both accuracy and speed.
- 116
- HS-Pose: Hybrid Scope Feature Extraction for Category-level Object Pose Estimation94
A simple network structure, the HS-layer, which extends 3D-GC to extract hybrid scope latent features from point cloud data for category-level object pose estimation tasks and outperforms the state-of-the-art methods by a large margin.
- Collaborative Learning for Hand and Object Reconstruction with Attention-guided Graph Convolution91
This work proposes a collaborative learning strategy where two-branches of deep networks are learning from each other, and achieves beyond state-of-the-art accuracy in 3D pose estimation, as well as recovers dense 3D hand and object shapes.
- 88
- 84
- 81
- 81
- DAC-h3: A Proactive Robot Cognitive Architecture to Acquire and Express Knowledge About the World and the Self77
This paper introduces a cognitive architecture for a humanoid robot to engage in a proactive, mixed-initiative exploration and manipulation of its environment, where the initiative can originate from both human and robot.
- PMnet: learning of disentangled pose and movement for unsupervised motion retargeting72
The proposed deep learning framework for unsupervised motion retargeting outperforms the state-of-the-art (sota) method by reducing the motion retargeting error from 7.68 (sota) to 1.95 (the authors').
- Robust action recognition using local motion and group sparsity67
This paper robustly identify local motions of interest in an unsupervised manner by taking advantage of group sparsity to robustly classify action types and demonstrates that the group sparse representation with the multiple kernel method can dramatically improve the action recognition performance.
- 63
- SeqHAND: RGB-Sequence-Based 3D Hand Pose and Shape Estimation62
This paper proposes a novel method that generates a synthetic dataset that mimics natural human hand movements by re-engineering annotations of an extant static hand pose dataset into pose-flows and shows that utilizing temporal information for 3D hand pose estimation significantly enhances general pose estimations by outperforming state-of-the-art methods in experiments on hand poses estimation benchmarks.
- DiffPose: SpatioTemporal Diffusion Model for Video-Based Human Pose Estimation61
DiffPose is presented, a novel diffusion architecture that formulates video-based human pose estimation as a conditional heatmap generation problem and a mechanism called Lookup-based Multi-Scale Feature Interaction that determines the correlations between local joints and global contexts across multiple scales.
- VaB-AL: Incorporating Class Imbalance and Difficulty with Variational Bayes for Active Learning55
A method based on the Bayes’ rule, that can naturally incorporate class imbalance into the Active Learning framework and be applied to classification tasks on multiple different datasets – including one that is a real-world dataset with heavy data imbalance – significantly outperforming the state of the art.
- Learning a Model-Driven Variational Network for Deformable Image Registration54
VR-Net is proposed, a novel cascaded variational network for unsupervised deformable image registration that outperforms state-of-the-art deep learning methods on registration accuracy, while maintains the fast inference speed of deep learning and the data-efficiency of variational model.
- 54
- 48
- 47
- Robust moving object detection against fast illumination change46
A novel moving object detection algorithm is proposed for which an illumination change model, a chromaticity difference model and a brightness ratio model are developed that estimates the intensity difference and intensity ratio of false foreground pixels, respectively.
- 42
- 41
- Mutual Information-Based Temporal Difference Learning for Human Pose Estimation in Video39
A novel multi-frame human pose estimation framework is presented, which employs temporal differences across frames to model dynamic contexts and engages mutual information objectively to facilitate useful motion information disentanglement.
- Self-Supervised Spatio-Temporal Representation Learning Using Variable Playback Speed Prediction39
A self-supervised learning method by predicting the variable playback speeds of a video and learning the spatio-temporal representation of the video by leveraging the variations in the visual appearance according to different playback speeds under the assumption of temporal coherence outperforms state-of-the-art self- supervised learning methods in action recognition.
- Spatio-Temporal Hough Forest for efficient detection–localisation–recognition of fingerwriting in egocentric camera37
A novel framework to recognise fingerwriting in mid-air using Spatio-Temporal Hough Forest that takes sequential data as input and performs regression on both spatial and temporal domain and can perform character recognition and localisation simultaneously.
- 35
- Variational Autoencoded Regression: High Dimensional Regression of Visual Data on Complex Manifold35
This paper proposes a new high dimensional regression method by merging Gaussian process regression into a variational autoencoder framework, and introduces a strategy to learn the latent space as well as the encoder and decoder so that the result of the regressed responded coincide with the corresponding response in the data space.
- Nesterov Accelerated ADMM for Fast Diffeomorphic Image Registration34
This paper proposes a simple iterative scheme that functionally composes intermediate non-stationary velocity fields to handle large deformations in images whilst guaranteeing diffeomorphisms in the resultant deformation, and proposes a convex optimisation model that uses a regularisation term of arbitrary order to impose smoothness on these velocity fields.
- Unsupervised Hyperbolic Representation Learning via Message Passing Auto-Encoders33
A novel hyperbolic message passing auto-encoder whose overallAuto-encoding is performed inhyperbolic space and is validated through extensive quantitative and qualitative analyses to validate the properties and benefits of the unsupervisedHyperbolic representations.
- MoST: Motion Style Transformer Between Diverse Action Contents31
A novel motion style transformer that effectively disentangles style from content and generates a plausible motion with transferred style from a source motion without the need for heuristic post-processing is proposed.
- What Do You See in Vehicle? Comprehensive Vision Solution for In-Vehicle Gaze Estimation30
This work introduces IVGaze, a pioneering dataset capturing in-vehicle gaze, and proposes a new vision-based solution for in-vehicle gaze collection, introducing a refined gaze target calibration method to tackle annotation challenges and explores a novel strategy for gaze zone classification by extending the GazeDPTR.
- 29
- Self-Supervised Visual Learning by Variable Playback Speeds Prediction of a Video28
A self-supervised visual learning method by predicting the variable playback speeds of a video and learning the spatio-temporal visual representation of the video by leveraging the variations in the visual appearance according to different playback speeds under the assumption of temporal coherence.
- GazeNeRF: 3D-Aware Gaze Redirection with Neural Radiance Fields27
The method leverages recent advancements in conditional image-based neural radiance fields and proposes a two-stream architecture that predicts volumetric features for the face and eye regions separately, which outperforms naively conditioned NeRF baselines as well as previous state-of-the-art 2D gaze redirection methods in terms of redirection accuracy and identity preservation.
- 27
- Contrastive Vicinal Space for Unsupervised Domain Adaptation27
This paper proposes an instance-wise minimax strategy that minimizes the entropy of high uncertainty instances in the vicinal space to tackle the stated problem of equilibrium collapse of labels.
- 25
- STARE: Spatio-Temporal Attention Relocation for Multiple Structured Activities Detection24
The main contributions can be summarized as follows: an information-theoretic dynamic attention relocation framework that allows the detection of multiple activities efficiently by exploiting the activity structure information and a new high-resolution data set of temporally-structured concurrent activities.
- 21
- 21
- 21
- 21
- 21
- Clothes Grasping and Unfolding Based on RGB-D Semantic Segmentation20
A novel Bi-directional Fractal Cross Fusion Network (BiFCNet) is proposed for semantic segmentation, enabling recognition of graspable regions in order to provide more possibilities for grasping, and a data augmentation method based on an adversarial strategy to reduce the cost of real data collection is proposed.
- 20
- PersonaBooth: Personalized Text-to-Motion Generation19
To tackle the dataset distribution gap, a persona token is introduced to accept new persona features and perform multi-modal adaptation for both text and visuals during finetuning, and a context-aware fusion mechanism to maximize the integration of persona cues from multiple input motions is introduced.
- 19
- 19
- Fast Support Vector Data Description Using K-Means Clustering18
A new fast SVDD method using K-means clustering method that trains each decomposed sub-problems to get support vectors and retrains with the support vectors to find a global data description of a whole target class.
- Repurposing existing deep networks for caption and aesthetic-guided image cropping17
Through various quantitative and qualitative evaluations, it is shown that the proposed novel optimization framework can produce crops that are well-aligned to intended user descriptions and aesthetically pleasing.
- 17
- 16
- 16
- 15
- Spectral Graphormer: Spectral Graph-based Transformer for Egocentric Two-Hand Reconstruction using Multi-View Color Images15
A novel transformer-based framework that reconstructs two high fidelity hands from multi-view RGB images that is able to produce realistic two-hand reconstructions and demonstrate the generalisation of synthetic-trained models to real data, as well as real-time AR/VR applications.
- 15
- Class-Attentive Diffusion Network for Semi-Supervised Classification15
Adaptive aggregation with Class-Attentive Diffusion (AdaCAD) is proposed, a new aggregation scheme that adaptively aggregates nodes probably of the same class among K-hop neighbors and significantly outperforms the state-of-the-art methods.
- User Modelling Using Multimodal Information for Personalised Dressing Assistance15
A user modelling method that combines vision and force data to enable the robot to search for an optimised dressing path for each user and improve as the human-robot interaction progresses is proposed.
- AMDANet: Attention-Driven Multi-Perspective Discrepancy Alignment for RGB-Infrared Image Fusion and Segmentation14
An AttentionDriven Multimodal Discrepancy Alignment Network (AMDANet) is proposed, which reallocates weights to reduce the saliency of discrepant features and utilizes low-weight features as cues to mitigate discrepancies between modalities, thereby achieving multi-modal feature alignment.
- GeoReF: Geometric Alignment Across Shape Variation for Category-level Object Pose Refinement14
This work introduces a novel architecture for category-level object pose refinement and introduces a cross-cloud transformation mechanism that efficiently merges di-verse data sources and pushes the limits of the model by incorporating the shape prior information for translation and size error prediction.
- 13
- 13
- 13
- 3D Prior is All You Need: Cross-Task Few-shot 2D Gaze Estimation12
A novel cross-task few-shot 2D gaze estimation approach, aiming to adapt a pre-trained 3D gaze estimation network for 2D gaze prediction on unseen devices using only a few training images, and introduces a dynamic pseudo-labelling strategy for flipped images.
- Multi-Modal Gaze Following in Conversational Scenarios12
A novel multimodal gaze following framework based on the observation "audiences tend to focus on the speaker" is proposed, which significantly outperforms existing methods in VGS datasets and proves the advantage of audio cues in gaze following tasks.
- 12
- 11
- Apparently Irrational Choice as Optimal Sequential Decision Making10
A normative approach to modeling apparently human irrational decision making (cognitive biases) that makes use of inherently rational computational mechanisms and makes quantitative predictions that correspond well to evidence about human choice is proposed.
- 9
- PIL-EYE: Integrated System for Sustainable Development of Intelligent Visual Surveillance Algorithms8
- TextGaze: Gaze-Controllable Face Generation with Natural Language7
This paper presents a novel gaze-controllable face generation task that overcomes limitations of existing approaches and introduces a text-of-gaze dataset containing over 90k text descriptions spanning a dense distribution of gaze and head poses.
- 7
- High-Fidelity Eye Animatable Neural Radiance Fields for Human Face7
This paper introduces a new Dynamic Eye-aware NeRF (DeNeRF) model, which is capable of generating high-fidelity images with accurate eyeball rotation and non-rigid periocular deformation, even under novel viewing angles and can effectively enhance gaze estimation performance.
- Towards Generic 3D Tracking in RGBD Videos: Benchmark and Baseline7
This paper investigates a novel problem: is it possible to track a generic 3D object in RGBD videos and predict 3D bounding boxes of the object of interest and proposes an effective tracking baseline to estimate 3D bounding boxes for arbitrary objects in RGBD videos, by fusing appearance and spatial information effectively.
- 7
- Towards cost-effective and safe contact-rich robotic manipulation with reinforcement learning: A review of techniques for future industrial automation6
It was found that current research efforts have significantly improved the cost and safety of RL-based control for contact-rich tasks, but further improvements can be made by progressing research towards improving knowledge transfer between tasks, improving inter-robot policy transfer and facilitating real-world and continual RL.
- 6
- 6
- 6
- Instruction-Grounded Visual Projectors for Continual Learning of Generative Vision-Language Models5
This work proposes a novel framework that grounds the translation of visual information on instructions for language models, and introduces a mixture of visual projectors, each serving as a specialized visual-to-language translation expert based on the given instruction context to adapt to new tasks.
- High-Resolution Spatiotemporal Modeling with Global-Local State Space Models for Video-Based Human Pose Estimation5
A novel framework is presented that extends Mamba from two aspects to separately learn global and local high-resolution spatiotemporal representations for VHPE and introduces a windowed space-time scan-based Local Refinement Mamba to enhance the high-frequency details of localized keypoint motions.
- BoIR: Box-Supervised Instance Representation for Multi-Person Pose Estimation5
A bounding box-level instance representation learning called BoIR, which simultaneously solves instance detection, instance disentanglement, and instance-keypoint association problems, and achieves globally consistent and disentangled instance representation.
- 5
- 5
- DarkSeg: Infrared-Driven Semantic Segmentation for Garment Grasping Detection in Low-Light Conditions4
Extensive experiments deploying DarkSeg on a Baxter robot demonstrate that DarkSeg achieves a 22% improvement in the grasping success rate while reducing the model parameters by 99.08 million compared to traditional methods, validating the practical viability of DarkSeg for robotic garment grasping in low-light conditions.
- PoseBH: Prototypical Multi-Dataset Training Beyond Human Pose Estimation4
PoseBH is introduced, a new MDT framework that tackles keypoint heterogeneity and limited supervision through two key techniques, and develops a cross-type self-supervision mechanism that aligns keypoint predictions with keypoint embedding prototypes, providing supervision without relying on teacher-student models or additional augmentations.
- NCRF: Neural Contact Radiance Fields for Free-Viewpoint Rendering of Hand-Object Interaction4
This work presents a novel free-viewpoint rendering framework, Neural Contact Radiance Field (NCRF), to reconstruct hand-object interactions from a sparse set of videos, and shows that this approach outperforms the current state-of-the-art in terms of both rendering quality and pose estimation accuracy.
- G-DAIC: A Gaze Initialized Framework for Description and Aesthetic-Based Image Cropping4
The runtime limitation of the state-of-the-art method is addressed by implementing the Early termination strategy to reduce the number of iterations required to produce the output and the adaptive Mixed scaling method to find the optimal output despite the size of the generated initialisation region and the described part of the image.
- Investigation of Architectures and Receptive Fields for Appearance-based Gaze Estimation4
It is revealed that tuning a few simple parameters of a ResNet architecture can outperform most of the existing state-of-the-art methods for the gaze estimation task on three popular datasets and that the stride number, input image resolution, and multi-region architecture are critical for the stare estimation performance.
- Towards anchoring self-learned representations to those of other agents4
This work proposes a developmental framework, where the robot incrementally acquires knowledge, and rapidly acquires primitives and complex actions by verbal descriptions and instructions from a human partner, and discovers correspondences between the robots body and other articulated objects and agents.
- Blockwise collaborative representation‐based classification via L 2 ‐norm of query data for accurate face recognition4
A novel classification scheme based on the L 2-norm of test blocks is proposed and this scheme is jointly applied with conventional reconstruction error-based classification and Experimental results show that the proposed methods outperform other representation-based methods for face recognition.
- Detection of suspicious activity using incremental outlier detection algorithms4
Efficiency of various techniques for incremental outlier detection on detecting unusual trajectories on simulated and real-life data obtained from SENSIAC database are compared.
- 3
- VL4Gaze: Unleashing Vision-Language Models for Gaze Following3
VL4Gaze is introduced, the first large-scale benchmark designed to investigate, evaluate, and unlock the potential of VLMs for gaze understanding and shows that even large-scale VLMs struggle to reliably infer gaze semantics and spatial localization without task-specific supervision.
- Motion-aware ensemble of three-mode trackers for unmanned aerial vehicles3
The proposed motion-aware ensemble shows a 5.3% improvement in average precision compared to the baseline correlation filter tracker, which effectively employs deep features while achieving a tracking speed of at least 80fps in the authors' experimental settings.
- GraspALL: Adaptive Structural Compensation from Illumination Variation for Robotic Garment Grasping in Any Low-Light Conditions2
The innovation of GraspALL lies in encoding continuous illumination changes into quantitative references to guide adaptive feature fusion between RGB and non-RGB modalities according to varying lighting intensities, thereby generating illumination-consistent grasping representations.
- Single-view Image to Novel-view Generation for Hand-Object Interactions2
This work introduces the HO123, a novel method to synthesize novel-view hand-object interaction images from a single image, which significantly outperforms state-of-the-art novel-view synthesis for hand-object interactions.
- 2
- Non-Zero Grid for Accurate 2-Bit Additive Power-of-Two CNN Quantization2
This work proposes a new non-zero grid formulation that enables 2-bit non-uniform quantization and allow the CNN to be trained successfully in every attempt, even for a large network.
- Joint Manifold Diffusion for Combining Predictions on Decoupled Observations2
A new non-parametric task dependence estimation procedure that automatically aligns evaluations of heterogeneous predictors across disjoint feature sets and outperforms existing methods even when applied to classical predictor combination settings.
- Personalised assistive dressing by humanoid robots using multi-modal information2
An approach to enable a humanoid robot to provide personalised dressing assistance for human users using multi-modal information and shows that the robot can assist human to wear a sleeveless jacket while reacting to the force resistances.
- Using Neural Network Algorithms for Compositional Mapping in STEM EELS2
Extended abstract of a paper presented at Microscopy and Microanalysis 2009 in Richmond, Virginia, USA, July 26 – July 30, 2009 about the design and application of nanofiltration membranes for selective separation of Na6(SO4) from Na2(SO3).
- Context-Aware Autoregressive Diffusion for Gloss-Wise Sign Language Production1
The Context-aware Gloss-wise AutoRegressive Diffusion model (GARD), a gloss-wise diffusion framework that models coarticulation by conditioning on both semantic and kinematic contexts, achieves superior performance over existing SLP methods in terms of both linguistic accuracy and motion similarity.
- Bidirectional regression for monocular 6DoF head pose estimation and reference system alignment1
TRGv2, a lightweight extension of the previous Translation, Rotation, and Geometry network, which explicitly models the bidirectional interaction between facial geometry and head pose is introduced, demonstrating that TRGv2 outperforms state-of-the-art methods in both accuracy and efficiency.
- Multi-Hypothesis 3D Hand Mesh Recovering from a Single Blurry Image1
A two-stage Multi-Hypothesis Hand Mesh Recovery network, consisting of a generation and selection model, which combines the classifying and ranking loss to better align with the target of the selection model.
- Audio-Visual Separation with Hierarchical Fusion and Representation Alignment1
A representation alignment approach that aligns the latent features of the audio encoder with embeddings extracted from pre-trained audio models is proposed, showing that it reduces modality gap between the audio and visual modalities.
- 1
- Category-Level 6D Object Pose Estimation with Flexible Vector-Based Rotation Representation1
A novel 3D graph convolution based pipeline for category-level 6D pose and size estimation from monocular RGB-D images achieves state-of-the-art performance on category- level tasks and is more suitable for the pose estimation tasks than other rotation representations.
- Combining Task Predictors via Enhancing Joint Predictability1
A new predictor combination algorithm is presented that improves the target by measuring the relevance of references based on their capabilities in predicting the target, and strengthening such estimated relevance by adopting a Bayesian framework.
- Action Chart: A Representation for Efficient Recognition of Complex Activity1
Experimental results show that the Action Chart gives promising recognition performance with low computational load and can be used for abstracting long video sequences.
- Hierarchical feature grouping for multiple object segmentation and tracking1
A hierarchical feature grouping method for multiple object segmentation and tracking that efficiently segments and tracks various kinds of objects without object detector, 3D model and geometry information.
- –
- –
- –
- –
- –
- –
- Enhancing Gaze Reasoning in Vision Foundation Models for Gaze Following–
This work proposes a novel training mechanism to enhance gaze reasoning in VFMs for gaze following, which includes a head-conditioned local LoRA, which enables localized adaptation to preserve scene token learning while improving head token learning for gaze reasoning.
- SimForce: Force and Surface Electromyography from Full Body Video with Graph Neural Nets–
SimForce predicts the more biomechanically accurate 3D human pose and shape of a given subject, along with their proposed muscle activations and resultant ground reaction force which leads to their input motion.
- –
- –
- AISPO: Enhancing Depth Reliability for Robotic Manipulation of Non-Lambertian Objects via Affine-Invariant Shape Prior–
AISPO is proposed, a depth completion framework that improves depth reliability for manipulation in challenging sensing conditions and combines multi-scale RGB-D feature fusion with an affine-invariant shape prior to enforce geometric consistency and mitigate catastrophic depth failures.
- Roll Your Eyes: Gaze Redirection via Explicit 3D Eyeball Rotation–
This work proposes a novel 3D gaze redirection framework that leverages an explicit 3D eyeball structure to represent the eyeballs with 3D Gaussian Splatting (3DGS), and proposes an adaptive deformation module that enables the replication of subtle muscle movements around the eyes.
- AuGQ: Augmented quantization granularity to overcome accuracy degradation for sub-byte quantized deep neural networks–
Augmented Quantization (AuGQ), a novel granularity technique tailored for low bit-width quantization, is introduced, demonstrating that AuGQ is agnostic to the training pipeline and batch normalization folding, distinguishing it from conventional quantization techniques.
- –
- Foundation Model-Driven Framework for Human-Object Interaction Prediction with Segmentation Mask Integration–
The Segmentation to Human-Object Interaction approach is introduced, a novel framework that integrates segmentation-based vision foundation models with the human-object interaction task, distinguished from traditional detection-based Human-Object Interaction (HOI) methods.
- –
- –
- –
- –
- Abnormal Behavior Detection using Simplified Social Force Model–
An abnormal behavior detection algorithm using simplified social force model, using particle advection scheme, to find interaction forces among individuals and applies threshold criteria on UMN dataset to detect abnormal crowd behavior.
- Support Vector Data Description Using Clustering Method–
This work proposes SVDD using clustering method, similar to divideand- conquer strategy; trains each decomposed subproblems to get support vectors, and uses two different clustering algorithms; K-means clustering and Mean Shift clustering.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-14. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.