Wispr Flow - Don’t type, just speak
Mini toolsPapers
Filter by company
-
Shifting Adaptation from Weight Space to Memory Space: A Memory-Augmented Agent for Medical Image SegmentationHarvard Medical School, New Jersey Institute of Technology, Northeastern University, University of California, University of Georgia
-
Stochastic Event Prediction via Temporal Motif TransitionsUniversity at Buffalo
-
Systematic Evaluation of Novel View Synthesis for Video Place RecognitionFordham University
-
ROSE: Reordered SparseGPT for More Accurate One-Shot Large Language Models PruningWestlake University
-
Confidence Before Answering: A Paradigm Shift for Efficient LLM Uncertainty Estimation
-
CylinderSplat: 3D Gaussian Splatting with Cylindrical Triplanes for Panoramic Novel View SynthesisNanjing University of Science and Technology, ShanghaiTech University
-
VerChol -- Grammar-First Tokenization for Agglutinative Languages
-
Computational Pathology in the Era of Emerging Foundation and Agentic AI -- International Expert Perspectives on Clinical Integration and Translational ReadinessAntGroup / Carnegie Mellon University, Duke University, Emory University, Georgia Institute of Technology, Harvard Medical School, Harvard University, Oncode Institute, Qingdao City University, Shanghai Jiao Tong University, Sichuan University, Stanford University, Technische Universität Dresden, Tsinghua University, University of Chicago, University of Texas, University of Warwick
-
HERO: Hierarchical Embedding-Refinement for Open-Vocabulary Temporal Sentence Grounding in VideosHangzhou Dianzi University, Tsinghua University
-
Reconstruct! Don't Encode: Self-Supervised Representation Reconstruction Loss for High-Intelligibility and Low-Latency Streaming Neural Audio CodecJohns Hopkins University, University of Southern California
-
PixARMesh: Autoregressive Mesh-Native Single-View Scene ReconstructionUniversity of California
-
Lost in Stories: Consistency Bugs in Long Story Generation by LLMs
-
Building an Ensemble LLM Semantic Tagger for UN Security Council Resolutions
-
InnoAds-Composer: Efficient Condition Composition for E-Commerce Poster GenerationChongqing University of Posts and Telecommunications, Hong Kong University of Science and Technology, Zhejiang University
-
Mitigating Bias in Concept Bottleneck Models for Fair and Interpretable Image ClassificationMassachusetts Institute of Technology
-
Reference-guided Policy Optimization for Molecular Optimization via LLM ReasoningAlibaba / Central Michigan University, DAMO Academy, Hong Kong Baptist University, Shanghai Jiao Tong University
-
Calibrated Credit Intelligence: Shift-Robust and Fair Risk Scoring with Bayesian Uncertainty and Gradient Boosting
-
LUMINA: LLM-Guided GPU Architecture Exploration via Bottleneck Analysis
-
CollabOD: Collaborative Multi-Backbone with Cross-scale Vision for UAV Small Object DetectionShenzhen ATC Technology / Hong Kong University of Science and Technology, The University of International Business and Economics Beijing, The University of Sydney
-
Beyond Geometry: Artistic Disparity Synthesis for Immersive 2D-to-3D
-
Pano3DComposer: Feed-Forward Compositional 3D Scene Generation from Single Panoramic ImageSun Yat-sen University
-
InfoGatherer: Principled Information Seeking via Evidence Retrieval and Strategic QuestioningCanadian Institute for Advanced Research, Dalhousie University, University of British Columbia, University of Washington, Vector Institute
-
The World Won't Stay Still: Programmable Evolution for Agent Benchmarks
-
CORE-Seg: Reasoning-Driven Segmentation for Complex Lesions via Reinforcement LearningAgency for Science, Technology and Research, Singapore, Nanjing University of Science and Technology, Southeast University
-
DeepFact: Co-Evolving Benchmarks and Agents for Deep Research Factuality
-
Stock Market Prediction Using Node Transformer Architecture Integrated with BERT Sentiment AnalysisUniversity of Ottawa
-
Design Experiments to Compare Multi-armed Bandit AlgorithmsThe Chinese University of Hong Kong, University of Toronto
-
BlackMirror: Black-Box Backdoor Detection for Text-to-Image Models via Instruction-Response DeviationBeijing Institute of Technology, Chinese Academy of Sciences, Sun Yat-sen University, University of Chinese Academy of Sciences
-
Learning Next Action Predictors from Human-Computer InteractionHasso Plattner Institute, New York University, Stanford University
-
Weak-SIGReg: Covariance Regularization for Stable Deep LearningKreasof AI
-
RAC: Rectified Flow Auto CoderNanyang Technological University, Rutgers University, University of Wisconsin-Madison
-
Towards Driver Behavior Understanding: Weakly-Supervised Risk Perception in Driving Scenes
-
Addressing the Ecological Fallacy in Larger LMs with Human ContextStony Brook University, Vanderbilt University
-
Beyond Static Frames: Temporal Aggregate-and-Restore Vision Transformer for Human Pose EstimationZhejiang Gongshang University
-
A Persistent-State Dataflow Accelerator for Memory-Bound Linear Attention Decode on FPGAUniversity of Southern California
-
FTSplat: Feed-forward Triangle Splatting NetworkNankai University
-
Implicit Style Conditioning: A Structured Style-Rewrite Framework for Low-Resource Character ModelingGuangdong University of Finance
-
OD-RASE: Ontology-Driven Risk Assessment and Safety Enhancement for Autonomous DrivingChubu University
-
Facial Expression Recognition Using Residual Masking NetworkHo Chi Minh City University of Technology
-
SLER-IR: Spherical Layer-wise Expert Routing for All-in-One Image RestorationUniversity of California San Diego
-
XAI for Coding Agent Failures: Transforming Raw Execution Traces into Actionable InsightsIslington College
-
Adaptive Radial Projection on Fourier Magnitude Spectrum for Document Image Skew EstimationHo Chi Minh City University of Technology, Vietnam National University Ho Chi Minh City
-
Vessel-Aware Deep Learning for OCTA-Based Detection of AMDStony Brook University
-
LucidNFT: LR-Anchored Multi-Reward Preference Optimization for Generative Real-World Super-ResolutionHong Kong University of Science and Technology
-
Energy-Driven Adaptive Visual Token Pruning for Efficient Vision-Language ModelsHong Kong University of Science and Technology
-
Unify the Views: View-Consistent Prototype Learning for Few-Shot SegmentationTongji University
-
Who We Are, Where We Are: Mental Health at the Intersection of Person, Situation, and Large Language ModelsOslo Metropolitan University, Stony Brook University, University of Texas
-
Domain-Adaptive Model Merging across Disconnected ModesNanchang University, Peking University, Southeast University, Tongji University
-
OVGGT: O(1) Constant-Cost Streaming Visual Geometry TransformerNational Taiwan University, National Taiwan University of Science and Technology
-
Omni-Masked Gradient Descent: Memory-Efficient Optimization via Mask Traversal with Improved ConvergencePeking University
