papers

Publications (20)

cs.AI2026

TriLens: Per-Layer Logit-Lens Entropy for White-Box Hallucination Detection

Bohan Yang, Yijun Gong, Zhi Zhang +3

When a language model hallucinates, the final answer is wrong, but the mistake is not necessarily invisible inside the model. Different internal pathways may remain uncertain, disa…

cs.LG2026

Off-Policy Value-Based Reinforcement Learning for Large Language Models

Peng-Yuan Wang, Ziniu Li, Tian Xu +8

Improving data utilization efficiency is critical for scaling reinforcement learning (RL) for long-horizon tasks where generating trajectories is expensive. However, the dominant R…

cs.RO2023

Modal-graph 3D shape servoing of deformable objects with raw point clouds

Bohan Yang, Congying Sui, Fangxun Zhong +1

Deformable object manipulation (DOM) with point clouds has great potential as non-rigid 3D shapes can be measured without detecting and tracking image features. However, robotic sh…

cs.RO2026

Beyond Failure Recovery: An Engagement-Aware Human-in-the-loop Framework for Robotic Systems

Jiaying Fang, Joyce Yang, Zhanxin Wu +2

Conventional human-in-the-loop approaches typically involve users only when a robot encounters failure or uncertainty, treating humans primarily as tools for improving robot perfor…

cs.RO2025

Prescribed Performance Control of Deformable Object Manipulation in Spatial Latent Space

Ning Han, Gu Gong, Bin Zhang +4

Manipulating three-dimensional (3D) deformable objects presents significant challenges for robotic systems due to their infinite-dimensional state space and complex deformable dyna…

cs.CL2026

Latent Fusion Jailbreak: Blending Harmful and Harmless Representations to Elicit Unsafe LLM Outputs

Wenpeng Xing, Mohan Li, Bohan Yang +5

Safety-aligned large language models can still be manipulated through white-box interventions that modify their internal representations. We introduce Latent Fusion Jailbreak (LFJ)…

cs.RO2026

A Human-in-the-Loop Confidence-Aware Failure Recovery Framework for Modular Robot Policies

Rohan Banerjee, Krishna Palempalli, Bohan Yang +5

Robots operating in unstructured human environments inevitably encounter failures, especially in robot caregiving scenarios. While humans can often help robots recover, excessive o…

math.NT2024

On identities concerning integer parts

Zichang Wang, Chengyang Wu, Bohan Yang

In 2007 V. Zhuravlev discovered a family of identities concerning integer parts which are satisfied by the number . Some of these identities turned out to be…

cs.CV2024

Leveraging CORAL-Correlation Consistency Network for Semi-Supervised Left Atrium MRI Segmentation

Xinze Li, Runlin Huang, Zhenghao Wu +4

Semi-supervised learning (SSL) has been widely used to learn from both a few labeled images and many unlabeled images to overcome the scarcity of labeled samples in medical image s…

cs.CV2026

QuadBox: Accelerating 3D Gaussian Splatting with Geometry-Aware Boxes

Xinze Li, Bohan Yang, Pengxu Chen +4

3D Gaussian Splatting (3DGS) has emerged as an advanced technique for real-time novel view synthesis by representing scene geometry and appearance using differentiable Gaussian pri…

cs.RO2023

Model-Free 3D Shape Control of Deformable Objects Using Novel Features Based on Modal Analysis

Bohan Yang, Bo Lu, Wei Chen +2

Shape control of deformable objects is a challenging and important robotic problem. This paper proposes a model-free controller using novel 3D global deformation features based on…

cs.LG2022

Machine Learning-based EEG Applications and Markets

Weiqing Gu, Bohan Yang, Ryan Chang

This paper addresses both the various EEG applications and the current EEG market ecosystem propelled by machine learning. Increasingly available open medical and health datasets u…

cs.RO2026

A Unified Calibration Framework for Coordinate and Kinematic Parameters in Dual-Arm Robots

Tianyu Huang, Bohan Yang, Bin Li +4

Precise collaboration in vision-based dual-arm robot systems requires accurate system calibration. Recent dual-robot calibration methods have achieved strong performance by simulta…

eess.SP2022

EEG Machine Learning for Analysis of Mild Traumatic Brain Injury: A survey

Weiqing Gu, Ryan Chang, Bohan Yang

Mild Traumatic Brain Injury (mTBI) is a common brain injury and affects a diverse group of people: soldiers, constructors, athletes, drivers, children, elders, and nearly everyone.…

math.NT2026

Maximal Gaps for Dilated Lacunary Integer Sequences

Yuval Peres, Bohan Yang

Let \((a_n)_{n\ge1}\subset\mathbb{N}\) be a lacunary sequence, \(a_{n+1}\ge q a_n\) for \(q>1\). For \(x\in\mathbb{T}\), we study the maximal empty circular gap \(G_N(x)\) of the f…

math.NT2026

Slowly Divergent Trajectories for Weighted Singular Vectors in R^2

Bohan Yang

Let satisfy and . For every function with , we prove that the set of -singular vectors whose weighted…

math.DS2023

Pointwise equidistribution for almost smooth functions with an error rate and Weighted Lévy-Khintchin theorem

Bohan Yang, Han Zhang

The purpose of this article is twofold: to prove a pointwise equidistribution theorem with an error rate for almost smooth functions, which strengthens the main result of Kleinbock…

cs.CV2022

Stereo Dense Scene Reconstruction and Accurate Localization for Learning-Based Navigation of Laparoscope in Minimally Invasive Surgery

Ruofeng Wei, Bin Li, Hangjie Mo +6

Objective: The computation of anatomical information and laparoscope position is a fundamental block of surgical navigation in Minimally Invasive Surgery (MIS). Recovering a dense…

math.CO2025

Markov's equation is not partition regular

Tianyi Tao, Bohan Yang

Markov's equation x^2 + y^2 + z^2 = 3xyz is a widely studied topic in number theory, and the structure of its solutions has profound connections with mathematical fields such as co…

physics.med-ph2025

Unsupervised deep learning model for fast energy layer pre-selection of delivery-efficient proton arc therapy plan optimization of nasopharyngeal carcinoma

Bohan Yang, Gang Liu, Yang Zhong +7

Proton arc therapy (PAT) is an emerging and promising modality in radiotherapy, offering improved dose distribution and treatment robustness over intensity-modulated proton therapy…