2024
Aligning CodeLLMs with Direct Preference Optimization.
CoRR, 2024

MatryoshkaKV: Adaptive KV Compression via Trainable Orthogonal Projection.
CoRR, 2024

In-context KV-Cache Eviction for LLMs via Attention-Gate.
CoRR, 2024

Meta-Unlearning on Diffusion Models: Preventing Relearning Unlearned Concepts.
CoRR, 2024

Lost in Translation: Latent Concept Misalignment in Text-to-Image Diffusion Models.
CoRR, 2024

Fine-tuning Diffusion Models for Enhancing Face Quality in Text-to-image Generation.
CoRR, 2024

Optimizing Speculative Decoding for Serving Large Language Models Using Goodput.
CoRR, 2024

AdaMoE: Token-Adaptive Routing with Null Experts for Mixture-of-Experts Language Models.
CoRR, 2024

3D-Properties: Identifying Challenges in DPO and Charting a Path Forward.
CoRR, 2024

MLCM: Multistep Consistency Distillation of Latent Diffusion Model.
CoRR, 2024

SCott: Accelerating Diffusion Models with Stochastic Consistency Distillation.
CoRR, 2024

SpikeZIP-TF: Conversion is All You Need for Transformer-based SNN.
Proceedings of the Forty-first International Conference on Machine Learning, 2024

Improved Operator Learning by Orthogonal Attention.
Proceedings of the Forty-first International Conference on Machine Learning, 2024

Online Speculative Decoding.
Proceedings of the Forty-first International Conference on Machine Learning, 2024

CLLMs: Consistency Large Language Models.
Proceedings of the Forty-first International Conference on Machine Learning, 2024

BayesDiff: Estimating Pixel-wise Uncertainty in Diffusion via Bayesian Inference.
Proceedings of the Twelfth International Conference on Learning Representations, 2024

AdaMoE: Token-Adaptive Routing with Null Experts for Mixture-of-Experts Language Models.
Proceedings of the Findings of the Association for Computational Linguistics: EMNLP 2024, 2024

Bayesian Exploration of Pre-Trained Models for Low-Shot Image Classification.
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024

Efficient Detection of LLM-generated Texts with a Bayesian Surrogate Model.
Proceedings of the Findings of the Association for Computational Linguistics, 2024

2023
Heterogeneous multi-task Gaussian Cox processes.
Mach. Learn., December, 2023

Batch virtual adversarial training for graph convolutional networks.
AI Open, January, 2023

LOVECon: Text-driven Training-Free Long Video Editing with ControlNet.
CoRR, 2023

Evaluating the Robustness of Text-to-image Diffusion Models against Real-world Attacks.
CoRR, 2023

On Calibrating Diffusion Probabilistic Models.
Proceedings of the Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, 2023

Towards Accelerated Model Training via Bayesian Data Selection.
Proceedings of the Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, 2023

Learning Sample Difficulty from Pre-trained Models for Reliable Prediction.
Proceedings of the Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, 2023

Learning Neural Eigenfunctions for Unsupervised Semantic Segmentation.
Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023

2022
Efficient Inference for Dynamic Flexible Interactions of Neural Populations.
J. Mach. Learn. Res., 2022

Neural Eigenfunctions Are Structured Representation Learners.
CoRR, 2022

Deep Ensemble as a Gaussian Process Approximate Posterior.
CoRR, 2022

Accelerated Linearized Laplace Approximation for Bayesian Deep Learning.
Proceedings of the Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, 2022

Confidence-based Reliable Learning under Dual Noises.
Proceedings of the Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, 2022

NeuralEF: Deconstructing Kernels by Deep Neural Networks.
Proceedings of the International Conference on Machine Learning, 2022

Exploring Memorization in Adversarial Training.
Proceedings of the Tenth International Conference on Learning Representations, 2022

BayesAdapter: Being Bayesian, Inexpensively and Reliably, via Bayesian Fine-tuning.
Proceedings of the Asian Conference on Machine Learning, 2022

2021
Accurate and Reliable Forecasting using Stochastic Differential Equations.
CoRR, 2021

Black-box Detection of Backdoor Attacks with Limited Information and Data.
Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision, 2021

LiBRe: A Practical Bayesian Approach to Adversarial Detection.
Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2021

2020
BayesAdapter: Being Bayesian, Inexpensively and Robustly, via Bayeisan Fine-tuning.
CoRR, 2020

Adversarial Distributional Training for Robust Deep Learning.
Proceedings of the Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, 2020

Understanding and Exploring the Network with Stochastic Architectures.
Proceedings of the Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, 2020

AutoSync: Learning to Synchronize for Data-Parallel Distributed Deep Learning.
Proceedings of the Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, 2020

2019
DBSN: Measuring Uncertainty through Bayesian Learning of Deep Neural Network Structures.
CoRR, 2019

Cluster Alignment With a Teacher for Unsupervised Domain Adaptation.
Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision, 2019

2018
Cavs: An Efficient Runtime System for Dynamic Neural Networks.
Proceedings of the 2018 USENIX Annual Technical Conference, 2018

2017
Structured Generative Adversarial Networks.
Proceedings of the Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017, 2017