papers

Publications (107)

cs.LG2026

Alignment Risks from Capability-Seeking RL Training

Yujun Zhou, Yue Huang, Han Bao +8

While most AI alignment research focuses on preventing models from generating explicitly harmful content, a more subtle risk arises from capability-seeking RL training in vulnerabl…

cs.LG2026

Non-Stationary Online Structured Prediction with Surrogate Losses

Shinsaku Sakaue, Han Bao, Yuzhou Cao

Online structured prediction, including online classification as a special case, is the task of sequentially predicting labels from input features. In this setting, the surrogate r…

cs.CL2026

PolicyLLM: Towards Excellent Comprehension of Public Policy for Large Language Models

Han Bao, Penghao Zhang, Yue Huang +9

Large Language Models (LLMs) are increasingly integrated into real-world decision-making, including in the domain of public policy. Yet, their ability to comprehend and reason abou…

cs.SE2026

UXBench: Measuring the Actionability of LLM-Generated UX Critiques

Wenjie Wang, Yue Huang, Zipeng Ling +11

Large language models (LLMs) are increasingly deployed as UX judges that inspect interfaces, diagnose usability problems, and propose repairs. Yet no controlled benchmark measures…

cs.CL2025

FlatQuant: Flatness Matters for LLM Quantization

Yuxuan Sun, Ruikang Liu, Haoli Bai +10

Recently, quantization has been widely used for the compression and acceleration of large language models (LLMs). Due to the outliers in LLMs, it is crucial to flatten weights and…

cs.CY2026

On the Trustworthiness of Generative Foundation Models: Guideline, Assessment, and Perspective

Yue Huang, Chujie Gao, Siyuan Wu +63

Generative Foundation Models (GenFMs) have emerged as transformative tools. However, their widespread adoption raises critical concerns regarding trustworthiness across dimensions.…

cs.LG2024

Referee-Meta-Learning for Fast Adaptation of Locational Fairness

Weiye Chen, Yiqun Xie, Xiaowei Jia +4

When dealing with data from distinct locations, machine learning algorithms tend to demonstrate an implicit preference of some locations over the others, which constitutes biases t…

cs.AI2026

Confidence Laundering in Agent Systems: Why Uncertainty Needs a Latent Carrier

Kaiwen Shi, Zheyuan Zhang, Han Bao +2

Modern agent systems can turn uncertainty into overconfidence. Fragile upstream decisions are often exposed to downstream components as clean intermediate artifacts, while the unce…

cs.SE2022

Failure Mechanism Traceability and Application in Human System Interface of Nuclear Power Plants using RESHA

Edward Chen, Han Bao, Tate Shorthill +2

In recent years, there has been considerable effort to modernize existing and new nuclear power plants with digital instrumentation and control systems. However, there has also bee…

cs.CL2026

Unleashing Low-Bit Inference on Ascend NPUs: A Comprehensive Evaluation of HiFloat Formats

Pengxiang Zhao, Hui-Ling Zhen, Xing Li +10

As LLMs scale, low-bit floating-point formats like MXFP and NVFP4 offer new opportunities for precision and efficiency. In this work, we evaluate HiFloat (HiF8 and HiF4), a family…

cs.CL2026

Boosting Large Language Models for Mental Manipulation Detection via Data Augmentation and Distillation

Yuansheng Gao, Peng Gao, Han Bao +4

Mental manipulation on social media poses a covert yet serious threat to individuals' psychological well-being and the integrity of online interactions. Detecting such behavior is…

cs.CL2023

Unbalanced Optimal Transport for Unbalanced Word Alignment

Yuki Arase, Han Bao, Sho Yokoi

Monolingual word alignment is crucial to model semantic interactions between sentences. In particular, null alignment, a phenomenon in which words have no corresponding counterpart…

cs.AI2026

SpecAlign: Efficient Specification-Grounded Alignment of Large Language Models via Synthetic Data

Wenjie Wang, Yue Huang, Zhengqing Yuan +6

As large language models (LLMs) are increasingly deployed in real-world applications, alignment is no longer governed by a single universal notion of safety or helpfulness, but ins…

stat.ME2026

Feasible Dose-Response Curves for Continuous Treatments Under Positivity Violations

Han Bao, Michael Schomaker

Positivity violations can complicate estimation and interpretation of causal dose-response curves (CDRCs) for continuous interventions. Weighting-based methods are designed to hand…

cs.CL2026

Evaluating Cross-Modal Reasoning Ability and Problem Characteristics with Multimodal Item Response Theory

Shunki Uebayashi, Kento Masui, Kyohei Atarashi +5

Multimodal Large Language Models (MLLMs) have recently emerged as general architectures capable of reasoning over diverse modalities. Benchmarks for MLLMs should measure their abil…

cs.CV2023

BEVStereo++: Accurate Depth Estimation in Multi-view 3D Object Detection via Dynamic Temporal Stereo

Yinhao Li, Jinrong Yang, Jianjian Sun +3

Bounded by the inherent ambiguity of depth perception, contemporary multi-view 3D object detection methods fall into the performance bottleneck. Intuitively, leveraging temporal mu…

cs.SE2021

Uncertainty Quantification and Software Risk Analysis for Digital Twins in the Nearly Autonomous Management and Control Systems: A Review

Linyu Lin, Han Bao, Nam Dinh

A nearly autonomous management and control (NAMAC) system is designed to furnish recommendations to operators for achieving particular goals based on NAMAC's knowledge base. As a c…

cs.LG2018

Unsupervised Domain Adaptation Based on Source-guided Discrepancy

Seiichi Kuroki, Nontawat Charoenphakdee, Han Bao +3

Unsupervised domain adaptation is the problem setting where data generating distributions in the source and target domains are different, and labels in the target domain are unavai…

cs.LG2020

Using Deep Learning to Explore Local Physical Similarity for Global-scale Bridging in Thermal-hydraulic Simulation

Han Bao, Nam Dinh, Linyu Lin +3

Current system thermal-hydraulic codes have limited credibility in simulating real plant conditions, especially when the geometry and boundary conditions are extrapolated beyond th…

stat.ML2021

Calibrated Surrogate Losses for Adversarially Robust Classification

Han Bao, Clayton Scott, Masashi Sugiyama

Adversarially robust classification seeks a classifier that is insensitive to adversarial perturbations of test patterns. This problem is often formulated via a minimax objective,…

cs.CV2026

Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective

Zhenfeng Su, Kang Zhao, Han Bao +4

While prior studies have successfully compressed vision Transformers (ViTs) through various pruning techniques, most have concentrated on width pruning to achieve significant reduc…

cs.DC2025

HAP: Hybrid Adaptive Parallelism for Efficient Mixture-of-Experts Inference

Haoran Lin, Xianzhi Yu, Kang Zhao +7

Current inference systems for Mixture-of-Experts (MoE) models primarily employ static parallelization strategies. However, these static approaches cannot consistently achieve optim…

cs.CV2026

Enhancing Video Representations with Spatiotemporal-Semantic Residual to Mitigate Hallucinations in Video Large Multimodal Models

Yuansheng Gao, Jinman Zhao, Tong Zhang +5

Although Video Large Multimodal Models have achieved strong performance in video understanding, they still suffer from hallucination. Existing inference-time intervention methods u…

cs.CL2025

Necessary and Sufficient Watermark for Large Language Models

Yuki Takezawa, Ryoma Sato, Han Bao +2

In recent years, large language models (LLMs) have achieved remarkable performances in various NLP tasks. They can generate texts that are indistinguishable from those written by h…

cs.LG2024

FastAttention: Extend FlashAttention2 to NPUs and Low-resource GPUs

Haoran Lin, Xianzhi Yu, Kang Zhao +17

FlashAttention series has been widely applied in the inference of large language models (LLMs). However, FlashAttention series only supports the high-level GPU architectures, e.g.,…

cs.CV2022

A Semi-Supervised Framework for Automatic Pixel-Wise Breast Cancer Grading of Histological Images

Yanyuet Man, Xiangyun Ding, Xingcheng Yao +1

Throughout the world, breast cancer is one of the leading causes of female death. Recently, deep learning methods are developed to automatically grade breast cancer of histological…

cs.LG2025

TAID: Temporally Adaptive Interpolated Distillation for Efficient Knowledge Transfer in Language Models

Makoto Shing, Kou Misaki, Han Bao +2

Causal language models have demonstrated remarkable capabilities, but their size poses significant challenges for deployment in resource-constrained environments. Knowledge distill…

cs.LG2018

Convex Formulation of Multiple Instance Learning from Positive and Unlabeled Bags

Han Bao, Tomoya Sakai, Issei Sato +1

Multiple instance learning (MIL) is a variation of traditional supervised learning problems where data (referred to as bags) are composed of sub-elements (referred to as instances)…

nlin.CD2024

Novel nonlinear system family generated from coupling effect of Sin-Cosine function

Fangfang Zhang, Jinyi Ge, Cuimei Jiang +4

The Sine-Cosine function, which is widely adopted in mathematics and physics, has attracted our attention due to its unique properties. By delving into the coupling effect of the S…

physics.comp-ph2019

A Data-driven Framework for Error Estimation and Mesh-Model Optimization in System-level Thermal-Hydraulic Simulation

Han Bao, Nam Dinh, Jeffrey Lane +1

Over the past decades, several computer codes were developed for simulation and analysis of thermal-hydraulics of system behaviors in nuclear reactors under operating, abnormal tra…

cs.CV2025

VModA: An Effective Framework for Adaptive NSFW Image Moderation

Han Bao, Qinying Wang, Zhi Chen +6

Not Safe/Suitable for Work (NSFW) content is rampant on social networks and poses serious harm to citizens, especially minors. Current detection methods mainly rely on deep learnin…

cs.LG2023

Momentum Tracking: Momentum Acceleration for Decentralized Deep Learning on Heterogeneous Data

Yuki Takezawa, Han Bao, Kenta Niwa +2

SGD with momentum is one of the key components for improving the performance of neural networks. For decentralized learning, a straightforward approach using momentum is Distribute…

cs.LG2026

Guardian-as-an-Advisor: Advancing Next-Generation Guardian Models for Trustworthy LLMs

Yue Huang, Haomin Zhuang, Jiayi Ye +6

Hard-gated safety checkers often over-refuse and misalign with a vendor's model spec; prevailing taxonomies also neglect robustness and honesty, yielding safer-on-paper yet less us…

nlin.CD2019

Generating Multi-Scroll Chua's Attractors via Simplified Piecewise-Linear Chua's Diode

Ning Wang, Chengqing Li, Han Bao +2

High implementation complexity of multi-scroll circuit is a bottleneck problem in real chaos-based communication. Especially, in multi-scroll Chua's circuit, the simplified impleme…

cs.LG2019

Classification from Pairwise Similarities/Dissimilarities and Unlabeled Data via Empirical Risk Minimization

Takuya Shimada, Han Bao, Issei Sato +1

Pairwise similarities and dissimilarities between data points might be easier to obtain than fully labeled data in real-world classification problems, e.g., in privacy-aware situat…

cs.AR2024

A Power-Efficient Hardware Implementation of L-Mul

Ruiqi Chen, Yangxintong Lyu, Han Bao +1

Multiplication is a core operation in modern neural network (NN) computations, contributing significantly to energy consumption. The linear-complexity multiplication (L-Mul) algori…

cs.HC2022

Systems-theoretic Hazard Analysis of Digital Human-System Interface Relevant to Reactor Trip

Edward Chen, Han Bao, Tate Shorthill +2

Human-system interface is one of the key advanced design features applied to modern digital instrumentation and control systems of nuclear power plants. The conventional design is…

cs.LG2018

Classification from Pairwise Similarity and Unlabeled Data

Han Bao, Gang Niu, Masashi Sugiyama

Supervised learning needs a huge amount of labeled data, which can be a big bottleneck under the situation where there is a privacy concern or labeling cost is high. To overcome th…

cs.LG2026

Denoise First, Orthogonalize Later: Understanding Momentum in Muon via Spectral Filtering

Xianliang Li, Zihan Zhang, Weiyang Liu +1

Muon has recently demonstrated strong empirical performance in large language model training, but the theoretical role of momentum in Muon remains unclear. Existing analyses of Muo…

cs.LG2026

Treatment Effect Estimation with Differentiated Networked Effect on Graph Data

Xiaofeng Lin, Han Bao, Hisashi Kashima

Estimating individual treatment effect (ITE) from observational graph data is crucial for decision-making in the fields such as commerce and medicine. This task is challenging due…

cs.LG2023

Embarrassingly Simple Text Watermarks

Ryoma Sato, Yuki Takezawa, Han Bao +2

We propose Easymark, a family of embarrassingly simple yet effective watermarks. Text watermarking is becoming increasingly important with the advent of Large Language Models (LLM)…

cs.LG2026

Non-asymptotic implicit bias of logistic regression at early-stage gradient descent dynamics

Han Bao

Gradient descent has been of particular interest in modern machine learning beyond sole focus on optimization. Implicit bias emerging from optimization, though not being encoded by…

cs.CV2025

AutoBench-V: Can Large Vision-Language Models Benchmark Themselves?

Han Bao, Yue Huang, Yanbo Wang +7

Large Vision-Language Models (LVLMs) have become essential for advancing the integration of visual and linguistic information. However, the evaluation of LVLMs presents significant…

quant-ph2016

Two-axis-twisting spin squeezing by multi-pass quantum erasure

Mingfeng Wang, Weizhi Qu, Pengxiong Li +3

Many-body entangled states are key elements in quantum information science and quantum metrology. One important problem in establishing a high degree of many-body entanglement usin…

cs.LG2025

Establishing Linear Surrogate Regret Bounds for Convex Smooth Losses via Convolutional Fenchel-Young Losses

Yuzhou Cao, Han Bao, Lei Feng +1

Surrogate regret bounds, also known as excess risk bounds, bridge the gap between the convergence rates of surrogate and target losses. The regret transfer is lossless if the surro…

cs.LG2024

Parameter-free Clipped Gradient Descent Meets Polyak

Yuki Takezawa, Han Bao, Ryoma Sato +2

Gradient descent and its variants are de facto standard algorithms for training machine learning models. As gradient descent is sensitive to its hyperparameters, we need to tune th…

cs.LG2019

Imitation Learning from Imperfect Demonstration

Yueh-Hua Wu, Nontawat Charoenphakdee, Han Bao +2

Imitation learning (IL) aims to learn an optimal policy from demonstrations. However, such demonstrations are often imperfect since collecting optimal ones is costly. To effectivel…

stat.ML2026

Spectral Gradient Descent Mitigates Anisotropy-Driven Misalignment: A Case Study in Phase Retrieval

Guillaume Braun, Han Bao, Wei Huang +1

Spectral gradient methods, such as the Muon optimizer, modify gradient updates by preserving directional information while discarding scale, and have shown strong empirical perform…

cs.LG2026

SkillGen: Verified Inference-Time Agent Skill Synthesis

Yuchen Ma, Yue Huang, Han Bao +5

Skills are a promising way to improve LLM agent capabilities without retraining, while keeping the added procedure reusable and controllable. However, high-quality skills are still…

physics.comp-ph2019

Computationally Efficient CFD Prediction of Bubbly Flow using Physics-Guided Deep Learning

Han Bao, Jinyong Feng, Nam Dinh +1

To realize efficient computational fluid dynamics (CFD) prediction of two-phase flow, a multi-scale framework was proposed in this paper by applying a physics-guided data-driven ap…

cs.CL2026

AI Alignment Breaks at the Edge

Han Bao, Yue Huang, Xiaoda Wang +5

General Alignment has improved average-case helpfulness and safety, but current alignment practice still rewards confident, single-turn responses. The problem is not only that mode…

cs.CV2022

BEVStereo: Enhancing Depth Estimation in Multi-view 3D Object Detection with Dynamic Temporal Stereo

Yinhao Li, Han Bao, Zheng Ge +3

Bounded by the inherent ambiguity of depth perception, contemporary camera-based 3D object detection methods fall into the performance bottleneck. Intuitively, leveraging temporal…

cs.DC2026

UBEP: Re-architecting Expert Parallelism Communication Library for Production Superpods

Yipeng Liu, Chang Liu, Si Shen +16

The deployment of Mixture-of-Experts (MoE) models on production high-bandwidth superpods, such as NVIDIA's NVL72/576 and Huawei's CloudMatrix384, introduces critical challenges bey…

cs.LG2023

STORM-GAN: Spatio-Temporal Meta-GAN for Cross-City Estimation of Human Mobility Responses to COVID-19

Han Bao, Xun Zhou, Yiqun Xie +2

Human mobility estimation is crucial during the COVID-19 pandemic due to its significant guidance for policymakers to make non-pharmaceutical interventions. While deep learning app…

cs.LG2020

Calibrated Surrogate Maximization of Linear-fractional Utility in Binary Classification

Han Bao, Masashi Sugiyama

Complex classification performance metrics such as the F-measure and Jaccard index are often used, in order to handle class-imbalanced cases such as information retrieval an…

eess.SY2020

A Redundancy-Guided Approach for the Hazard Analysis of Digital Instrumentation and Control Systems in Advanced Nuclear Power Plants

Tate Shorthill, Han Bao, Hongbin Zhang +1

Digital instrumentation and control (I&C) upgrades are a vital research area for nuclear industry. Despite their performance benefits, deployment of digital I&C in nuclear power pl…

cs.LG2025

Revisiting Online Learning Approach to Inverse Linear Optimization: A FenchelYoung Loss Perspective and Gap-Dependent Regret Analysis

Shinsaku Sakaue, Han Bao, Taira Tsuchiya

This paper revisits the online learning approach to inverse linear optimization studied by Bärmann et al. (2017), where the goal is to infer an unknown linear objective function o…

cs.LG2022

On the Surrogate Gap between Contrastive and Supervised Losses

Han Bao, Yoshihiro Nagano, Kento Nozawa

Contrastive representation learning encourages data representation to make semantically similar pairs closer than randomly drawn negative samples, which has been successful in vari…

cs.RO2026

NavIsaacLab: Generating Realistic Crowd via Parallel Robot Learning for Benchmarking Human-aware Navigation

Bingyi Xia, Han Bao, Jingyu Zhu +6

Robot autonomous navigation that accounts for surrounding human activities is crucial for ensuring both safety and natural human-robot interaction in real-world environments shared…

cs.LG2025

Online Inverse Linear Optimization: Efficient Logarithmic-Regret Algorithm, Robustness to Suboptimality, and Lower Bound

Shinsaku Sakaue, Taira Tsuchiya, Han Bao +1

In online inverse linear optimization, a learner observes time-varying sets of feasible actions and an agent's optimal actions, selected by solving linear optimization over the fea…

cs.SE2022

Application of Orthogonal Defect Classification for Software Reliability Analysis

Edward Chen, Han Bao, Tate Shorthill +3

The modernization of existing and new nuclear power plants with digital instrumentation and control systems (DI&C) is a recent and highly trending topic. However, there lacks stron…

cs.LG2025

PhiNets: Brain-inspired Non-contrastive Learning Based on Temporal Prediction Hypothesis

Satoki Ishikawa, Makoto Yamada, Han Bao +1

Predictive coding is a theory which hypothesises that cortex predicts sensory inputs at various levels of abstraction to minimise prediction errors. Inspired by predictive coding,…

cs.LG2023

Beyond Exponential Graph: Communication-Efficient Topologies for Decentralized Learning via Finite-time Convergence

Yuki Takezawa, Ryoma Sato, Han Bao +2

Decentralized learning has recently been attracting increasing attention for its applications in parallel computation and privacy preservation. Many recent studies stated that the…

cs.LG2022

Robust computation of optimal transport by -potential regularization

Shintaro Nakamura, Han Bao, Masashi Sugiyama

Optimal transport (OT) has become a widely used tool in the machine learning field to measure the discrepancy between probability distributions. For instance, OT is a popular loss…

cs.LG2026

Why Semantic Entropy Fails: Geometry-Aware and Calibrated Uncertainty for Policy Optimization

Zheyuan Zhang, Kaiwen Shi, Han Bao +3

Post-training has become central to improving reasoning and alignment in large language models, where critic-free models enable scalable learning from model-generated outputs but l…

quant-ph2024

Concurrent spin squeezing and light squeezing in an atomic ensemble

Shenchao Jin, Junlei Duan, Youwei Zhang +7

Squeezed spin states and squeezed light are both key resources for quantum metrology and quantum information science, but have been separately investigated in experiments so far. S…

math.ST2021

Robust covariance estimation for distributed principal component analysis

Kangqiang Li, Han Bao, Lixin Zhang

Fan et al. [ (6) (2019) 3009-3031] constructed a distributed principal component analysis (PCA) algorithm to reduc…

quant-ph2019

Measurements with prediction and retrodiction on the collective spin of 10^{11} atoms beat the standard quantum limit

Han Bao, Junlei Duan, Xingda Lu +10

Quantum probes using uncorrelated particles give a limit on the measurement sensitivity referred to as the standard quantum limit (SQL). The SQL, however, can be overcome by ex…

cs.AI2026

MemoHarness: Agent Harnesses That Learn from Experience

Yue Huang, Wenjie Wang, Han Bao +7

MemoHarness is a framework that automatically adapts the control layer (harness) of large language model agents by learning from past executions, using a dual‑layer experience bank…

#llm agents#adaptive harness#experience replay#prompt optimization
cs.CL2026

BATQuant: Outlier-resilient MXFP4 Quantization via Learnable Block-wise Optimization

Ji-Fu Li, Manyi Zhang, Xiaobo Xia +4

Microscaling floating-point (MXFP) formats have emerged as a promising standard for deploying Multi-modal Large Language Models (MLLMs) and Large Language Models (LLMs) on modern a…

cs.AI2026

Drift-Bench: Diagnosing Cooperative Breakdowns in LLM Agents under Input Faults via Multi-Turn Interaction

Han Bao, Zheyuan Zhang, Pengcheng Jing +3

As Large Language Models transition to autonomous agents, user inputs frequently violate cooperative assumptions (e.g., implicit intent, missing parameters, false presuppositions,…

physics.chem-ph2023

DeePMD-kit v2: A software package for Deep Potential models

Jinzhe Zeng, Duo Zhang, Denghui Lu +44

DeePMD-kit is a powerful open-source software package that facilitates molecular dynamics simulations using machine learning potentials (MLP) known as Deep Potential (DP) models. T…

cs.LG2020

Learning from Noisy Similar and Dissimilar Data

Soham Dan, Han Bao, Masashi Sugiyama

With the widespread use of machine learning for classification, it becomes increasingly important to be able to use weaker kinds of supervision for tasks in which it is hard to obt…

physics.atom-ph2025

Microwave-Dressing of Rydberg States in a Trapped Calcium Ion

Han Bao, Alexander Schulze-Makuch, Ferdinand Schmidt-Kaler

We are using optical- and microwave-fields to excite Rydberg states in trapped cold 40Ca+ ions. We employ a single ion and observe spectroscopically in the manifold of a principal…

stat.ML2026

Proper losses regret at least 1/2-order

Han Bao, Asuka Takatsu

A fundamental challenge in machine learning is the choice of a loss as it characterizes our learning task, is minimized in the training phase, and serves as an evaluation criterion…

cs.LG2025

LiD-FL: Towards List-Decodable Federated Learning

Hong Liu, Liren Shan, Han Bao +3

Federated learning is often used in environments with many unverified participants. Therefore, federated learning under adversarial attacks receives significant attention. This pap…

cs.LG2024

Online Policy Learning from Offline Preferences

Guoxi Zhang, Han Bao, Hisashi Kashima

In preference-based reinforcement learning (PbRL), a reward function is learned from a type of human feedback called preference. To expedite preference collection, recent works hav…

quant-ph2024

Quantum computing architecture with Rydberg gates in trapped ions

Han Bao, Jonas Vogel, Ulrich Poschinger +1

Fast entangling gate operations are a fundamental prerequisite for quantum simulation and computation. We propose an entangling scheme for arbitrary pairs of ions in a linear cryst…

cs.CV2026

MultiToP: Learning to Patch Visual Tokens to Mitigate Hallucinations in Video Large Multimodal Models

Yuansheng Gao, Wenbin Xing, Jiahao Yuan +4

Video Large Multimodal Models have achieved remarkable progress in video understanding, yet they remain prone to hallucinations, where generated responses are not faithfully suppor…

cs.LG2025

Many-to-Many Matching via Sparsity Controlled Optimal Transport

Weijie Liu, Han Bao, Makoto Yamada +3

Many-to-many matching seeks to match multiple points in one set and multiple points in another set, which is a basis for a wide range of data mining problems. It can be naturally r…

cs.CL2024

Zipfian Whitening

Sho Yokoi, Han Bao, Hiroto Kurita +1

The word embedding space in neural models is skewed, and correcting this can improve task performance. We point out that most approaches for modeling, correcting, and measuring the…

stat.ML2025

Any-stepsize Gradient Descent for Separable Data under Fenchel-Young Losses

Han Bao, Shinsaku Sakaue, Yuki Takezawa

The gradient descent (GD) has been one of the most common optimizer in machine learning. In particular, the loss landscape of a neural network is typically sharpened during the ini…

cs.LG2025

Feature Normalization Prevents Collapse of Non-contrastive Learning Dynamics

Han Bao

Contrastive learning is a self-supervised representation learning framework, where two positive views generated through data augmentation are made similar by an attraction force in…

cs.CV2023

Will Large-scale Generative Models Corrupt Future Datasets?

Ryuichiro Hataya, Han Bao, Hiromi Arai

Recently proposed large-scale text-to-image generative models such as DALLE 2, Midjourney, and StableDiffusion can generate high-quality and realistic images from users' pro…

eess.SY2021

An Integrated Risk Assessment Process of Safety-Related Digital I&C Systems in Nuclear Power Plants

Hongbin Zhang, Han Bao, Tate Shorthill +1

Upgrading the existing analog instrumentation and control (IC) systems to state-of-the-art digital IC (DIC) systems will greatly benefit existing light-water reactors (LWRs). Howev…

eess.IV2019

Investigations of the Influences of a CNN's Receptive Field on Segmentation of Subnuclei of Bilateral Amygdalae

Han Bao

Segmentation of objects with various sizes is relatively less explored in medical imaging, and has been very challenging in computer vision tasks in general. We hypothesize that th…

quant-ph2020

Retrodiction beyond the Heisenberg uncertainty relation

Han Bao, Shenchao Jin, Junlei Duan +4

In quantum mechanics, the Heisenberg uncertainty relation presents an ultimate limit to the precision by which one can predict the outcome of position and momentum measurements on…

cs.LG2024

Online Structured Prediction with Fenchel--Young Losses and Improved Surrogate Regret for Online Multiclass Classification with Logistic Loss

Shinsaku Sakaue, Han Bao, Taira Tsuchiya +1

This paper studies online structured prediction with full-information feedback. For online multiclass classification, Van der Hoeven (2020) established \emph{finite} surrogate regr…

quant-ph2018

Entangling and squeezing atoms by weak measurement

Mingfeng Wang, Weizhi Qu, Han Bao +2

A weak measurement approach is proposed to entangle and squeeze atoms. We show that even for very small coupling strength between light and atoms, one can achieve large squeezing u…

cs.SE2022

An Application of a Modified Beta Factor Method for the Analysis of Software Common Cause Failures

Tate Shorthill, Han Bao, Edward Chen +1

This paper presents an approach for modeling software common cause failures (CCFs) within digital instrumentation and control (I&C) systems. CCFs consist of a concurrent failure be…

cs.LG2025

Beyond 2:4: exploring V:N:M sparsity for efficient transformer inference on GPUs

Kang Zhao, Tao Yuan, Han Bao +6

To date, 2:4 sparsity has stood as the only sparse pattern that can be accelerated using sparse tensor cores on GPUs. In practice, 2:4 sparsity often possesses low actual speedups…

cs.LG2023

Dynamic Model Agnostic Reliability Evaluation of Machine-Learning Methods Integrated in Instrumentation & Control Systems

Edward Chen, Han Bao, Nam Dinh

In recent years, the field of data-driven neural network-based machine learning (ML) algorithms has grown significantly and spurred research in its applicability to instrumentation…

stat.ML2022

Approximating 1-Wasserstein Distance with Trees

Makoto Yamada, Yuki Takezawa, Ryoma Sato +3

Wasserstein distance, which measures the discrepancy between distributions, shows efficacy in various types of natural language processing (NLP) and computer vision (CV) applicatio…

stat.ML2026

Brenier Isotonic Regression

Han Bao, Amirreza Eshraghi, Yutong Wang

Isotonic regression (IR) is shape-constrained regression to maintain a univariate fitting curve non-decreasing, which has numerous applications including single-index models and pr…

cs.CL2026

DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

DeepSeek-AI, Daya Guo, Dejian Yang +195

General reasoning represents a long-standing and formidable challenge in artificial intelligence. Recent breakthroughs, exemplified by large language models (LLMs) and chain-of-tho…

quant-ph2025

Realizing exceptional points by Floquet dissipative couplings in thermal atoms

Zimo Zhang, Fengbo Zhang, Zhongxiao Xu +3

Exceptional degeneracies and generically complex spectra of non-Hermitian systems are at the heart of numerous phenomena absent in the Hermitian realm. Recently, it was suggested t…

cs.CL2025

DeepSeek-V3 Technical Report

DeepSeek-AI, Aixin Liu, Bei Feng +195

We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. To achieve efficient inference and cost-effec…

eess.SY2022

Quantitative Evaluation of Common Cause Failures in High Safety-significant Safety-related Digital Instrumentation and Control Systems in Nuclear Power Plants

Han Bao, Hongbin Zhang, Tate Shorthill +2

Digital instrumentation and control (DIC) systems at nuclear power plants (NPPs) have many advantages over analog systems. They are proven to be more reliable, cheaper, and easier…

cs.LG2023

Estimating Treatment Effects Under Heterogeneous Interference

Xiaofeng Lin, Guoxi Zhang, Xiaotian Lu +3

Treatment effect estimation can assist in effective decision-making in e-commerce, medicine, and education. One popular application of this estimation lies in the prediction of the…

physics.flu-dyn2020

Deep Learning Interfacial Momentum Closures in Coarse-Mesh CFD Two-Phase Flow Simulation Using Validation Data

Han Bao, Jinyong Feng, Nam Dinh +1

Multiphase flow phenomena have been widely observed in the industrial applications, yet it remains a challenging unsolved problem. Three-dimensional computational fluid dynamics (C…