Jing Huang

Affiliations

Stanford University

papers

1,160

total citations

papers (20)

Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives

CVPR 2024arXiv

343

citations

TextOCR: Towards Large-Scale End-to-End Reasoning for Arbitrary-Shaped Scene Text

CVPR 2021arXiv

222

citations

Mask TextSpotter v3: Segmentation Proposal Network for Robust Scene Text Spotting

ECCV 2020arXiv

204

citations

AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders

ICML 2025arXiv

118

citations

Improving Weakly Supervised Visual Grounding by Contrastive Knowledge Distillation

CVPR 2021arXiv

citations

PoseTriplet: Co-Evolving 3D Human Pose Estimation, Imitation, and Hallucination Under Self-Supervision

CVPR 2022arXiv

citations

A Multiplexed Network for End-to-End, Multilingual OCR

CVPR 2021arXiv

citations

Tight Mutual Information Estimation With Contrastive Fenchel-Legendre Optimization

NEURIPS 2022arXiv

citations

INTER: Mitigating Hallucination in Large Vision-Language Models by Interaction Guidance Sampling

ICCV 2025arXiv

citations

Blackbox Model Provenance via Palimpsestic Membership Inference

NEURIPS 2025arXiv

citations

MoLE:Decoding by Mixture of Layer Experts Alleviates Hallucination in Large Vision-Language Models

AAAI 2025

citations

RESCUE: Crowd Evacuation Simulation via Controlling SDM-United Characters

ICCV 2025arXiv

citations

Diffusion Epistemic Uncertainty with Asymmetric Learning for Diffusion-Generated Image Detection

ICCV 2025arXiv

citations

Self-Supervised Object Detection from Egocentric Videos

ICCV 2023

citations

A Plug-and-Play Bregman ADMM Module for Inferring Event Branches in Temporal Point Processes

AAAI 2025arXiv

citations

Jing Huang

Affiliations

papers (20)

Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives

TextOCR: Towards Large-Scale End-to-End Reasoning for Arbitrary-Shaped Scene Text

Mask TextSpotter v3: Segmentation Proposal Network for Robust Scene Text Spotting

AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders

Improving Weakly Supervised Visual Grounding by Contrastive Knowledge Distillation

PoseTriplet: Co-Evolving 3D Human Pose Estimation, Imitation, and Hallucination Under Self-Supervision

A Multiplexed Network for End-to-End, Multilingual OCR

Tight Mutual Information Estimation With Contrastive Fenchel-Legendre Optimization

Real-Time Simulated Avatar from Head-Mounted Sensors

MIB: A Mechanistic Interpretability Benchmark

Crowd3D: Towards Hundreds of People Reconstruction From a Single Image

LLMs Encode Harmfulness and Refusal Separately

Querying as Prompt: Parameter-Efficient Learning for Multimodal Language Model

INTER: Mitigating Hallucination in Large Vision-Language Models by Interaction Guidance Sampling

Blackbox Model Provenance via Palimpsestic Membership Inference

MoLE:Decoding by Mixture of Layer Experts Alleviates Hallucination in Large Vision-Language Models

RESCUE: Crowd Evacuation Simulation via Controlling SDM-United Characters

Diffusion Epistemic Uncertainty with Asymmetric Learning for Diffusion-Generated Image Detection

Self-Supervised Object Detection from Egocentric Videos

A Plug-and-Play Bregman ADMM Module for Inferring Event Branches in Temporal Point Processes

papers (20)

Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives

TextOCR: Towards Large-Scale End-to-End Reasoning for Arbitrary-Shaped Scene Text

Mask TextSpotter v3: Segmentation Proposal Network for Robust Scene Text Spotting

AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders

Improving Weakly Supervised Visual Grounding by Contrastive Knowledge Distillation

PoseTriplet: Co-Evolving 3D Human Pose Estimation, Imitation, and Hallucination Under Self-Supervision

A Multiplexed Network for End-to-End, Multilingual OCR

Tight Mutual Information Estimation With Contrastive Fenchel-Legendre Optimization

Real-Time Simulated Avatar from Head-Mounted Sensors

MIB: A Mechanistic Interpretability Benchmark

Crowd3D: Towards Hundreds of People Reconstruction From a Single Image

LLMs Encode Harmfulness and Refusal Separately

Querying as Prompt: Parameter-Efficient Learning for Multimodal Language Model

INTER: Mitigating Hallucination in Large Vision-Language Models by Interaction Guidance Sampling

Blackbox Model Provenance via Palimpsestic Membership Inference

MoLE:Decoding by Mixture of Layer Experts Alleviates Hallucination in Large Vision-Language Models

RESCUE: Crowd Evacuation Simulation via Controlling SDM-United Characters

Diffusion Epistemic Uncertainty with Asymmetric Learning for Diffusion-Generated Image Detection

Self-Supervised Object Detection from Egocentric Videos

A Plug-and-Play Bregman ADMM Module for Inferring Event Branches in Temporal Point Processes