Hong

papers

1,666

total citations

papers (24)

LRM: Large Reconstruction Model for Single Image to 3D

ICLR 2024arXiv

711

citations

BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions

ICLR 2025arXiv

410

citations

Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond

ICML 2025arXiv

citations

Wear-Any-Way: Manipulable Virtual Try-on via Sparse Correspondence Alignment

ECCV 2024arXiv

citations

Inference-Time Alignment of Diffusion Models with Direct Noise Optimization

ICML 2025arXiv

citations

BountyBench: Dollar Impact of AI Agent Attackers and Defenders on Real-World Cybersecurity Systems

NEURIPS 2025arXiv

citations

Differentially Private SGD Without Clipping Bias: An Error-Feedback Approach

ICLR 2024arXiv

citations

Trajectory Inference with Smooth Schrödinger Bridges

ICML 2025arXiv

citations

KLASS: KL-Guided Fast Inference in Masked Diffusion Models

NEURIPS 2025arXiv

citations

DiSK: Differentially Private Optimizer with Simplified Kalman Filter for Noise Reduction

ICLR 2025arXiv

citations

A Implies B: Circuit Analysis in LLMs for Propositional Logical Reasoning

NEURIPS 2025arXiv

citations

ReGen: Generative Robot Simulation via Inverse Design

ICLR 2025arXiv

citations

Overcoming Distribution Mismatch in Quantizing Image Super-Resolution Networks

ECCV 2024arXiv

citations

Mitigating Hallucination in VideoLLMs via Temporal-Aware Activation Engineering

NEURIPS 2025arXiv

citations

Unlabeled Data Improves Fine-Grained Image Zero-shot Classification with Multimodal LLMs

NEURIPS 2025arXiv

citations

CryptoMoE: Privacy-Preserving and Scalable Mixture of Experts Inference via Balanced Expert Routing

NEURIPS 2025arXiv

citations

Multi-Agent Retrieval-Augmented Framework for Evidence-Based Counterspeech Against Health Misinformation

COLM 2025arXiv

citations

Rebalancing Contrastive Alignment with Bottlenecked Semantic Increments in Text-Video Retrieval

NEURIPS 2025arXiv

citations

Universal Few-shot Spatial Control for Diffusion Models

NEURIPS 2025arXiv

citations

A Principled Path to Fitted Distributional Evaluation

NEURIPS 2025arXiv

citations

Hong

papers (24)

LRM: Large Reconstruction Model for Single Image to 3D

BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions

AFlow: Automating Agentic Workflow Generation

MoBA: Mixture of Block Attention for Long-Context LLMs

RazorAttention: Efficient KV Cache Compression Through Retrieval Heads

DP-OPT: Make Large Language Model Your Privacy-Preserving Prompt Engineer

Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond

Wear-Any-Way: Manipulable Virtual Try-on via Sparse Correspondence Alignment

Inference-Time Alignment of Diffusion Models with Direct Noise Optimization

BountyBench: Dollar Impact of AI Agent Attackers and Defenders on Real-World Cybersecurity Systems

Differentially Private SGD Without Clipping Bias: An Error-Feedback Approach

Trajectory Inference with Smooth Schrödinger Bridges

KLASS: KL-Guided Fast Inference in Masked Diffusion Models

DiSK: Differentially Private Optimizer with Simplified Kalman Filter for Noise Reduction

A Implies B: Circuit Analysis in LLMs for Propositional Logical Reasoning

ReGen: Generative Robot Simulation via Inverse Design

Overcoming Distribution Mismatch in Quantizing Image Super-Resolution Networks

Mitigating Hallucination in VideoLLMs via Temporal-Aware Activation Engineering

Unlabeled Data Improves Fine-Grained Image Zero-shot Classification with Multimodal LLMs

CryptoMoE: Privacy-Preserving and Scalable Mixture of Experts Inference via Balanced Expert Routing

Multi-Agent Retrieval-Augmented Framework for Evidence-Based Counterspeech Against Health Misinformation

Rebalancing Contrastive Alignment with Bottlenecked Semantic Increments in Text-Video Retrieval

Universal Few-shot Spatial Control for Diffusion Models

A Principled Path to Fitted Distributional Evaluation

papers (24)

LRM: Large Reconstruction Model for Single Image to 3D

BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions

AFlow: Automating Agentic Workflow Generation

MoBA: Mixture of Block Attention for Long-Context LLMs

RazorAttention: Efficient KV Cache Compression Through Retrieval Heads

DP-OPT: Make Large Language Model Your Privacy-Preserving Prompt Engineer

Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond

Wear-Any-Way: Manipulable Virtual Try-on via Sparse Correspondence Alignment

Inference-Time Alignment of Diffusion Models with Direct Noise Optimization

BountyBench: Dollar Impact of AI Agent Attackers and Defenders on Real-World Cybersecurity Systems

Differentially Private SGD Without Clipping Bias: An Error-Feedback Approach

Trajectory Inference with Smooth Schrödinger Bridges

KLASS: KL-Guided Fast Inference in Masked Diffusion Models

DiSK: Differentially Private Optimizer with Simplified Kalman Filter for Noise Reduction

A Implies B: Circuit Analysis in LLMs for Propositional Logical Reasoning

ReGen: Generative Robot Simulation via Inverse Design

Overcoming Distribution Mismatch in Quantizing Image Super-Resolution Networks

Mitigating Hallucination in VideoLLMs via Temporal-Aware Activation Engineering

Unlabeled Data Improves Fine-Grained Image Zero-shot Classification with Multimodal LLMs

CryptoMoE: Privacy-Preserving and Scalable Mixture of Experts Inference via Balanced Expert Routing

Multi-Agent Retrieval-Augmented Framework for Evidence-Based Counterspeech Against Health Misinformation

Rebalancing Contrastive Alignment with Bottlenecked Semantic Increments in Text-Video Retrieval

Universal Few-shot Spatial Control for Diffusion Models

A Principled Path to Fitted Distributional Evaluation