Author name cluster

Bo Xiong

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

8 papers

2 author rows

AAAI Conference 2026 Conference Paper

Spike Imaging Velocimetry: Dense Motion Estimation of Fluids Using Spike Streams

Yunzhong Zhang
You Zhou
Changqing Su
Zhen Cheng
Zhaofei Yu
Bo Xiong
Tiejun Huang
Xun Cao

Particle Image Velocimetry (PIV) is a widely adopted non-invasive imaging technique that tracks the motion of tracer particles across image sequences to capture the velocity distribution of fluid flows. It is commonly employed to analyze complex flow structures and validate numerical simulations. This study explores the untapped potential of spike cameras—ultra-high-speed, high-dynamic-range vision sensors—in high-speed fluid velocimetry. We propose a deep learning framework, Spike Imaging Velocimetry (SIV), tailored for high-resolution fluid motion estimation. To enhance the network’s performance, we design three novel modules specifically adapted to the characteristics of fluid dynamics and spike streams: the Detail-Preserving Hierarchical Transform (DPHT), the Graph Encoder (GE), and the Multi-scale Velocity Refinement (MSVR). Furthermore, we introduce a spike-based PIV dataset, Particle Scenes with Spike and Displacement (PSSD), which contains labeled samples from three representative fluid-dynamics scenarios: steady turbulence, high-speed flow, and high-dynamic-range conditions. Our proposed method outperforms existing baselines across all these scenarios, demonstrating its effectiveness.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

Hyperbolic Fine-Tuning for Large Language Models

Menglin Yang
Ram B
Aosong Feng
Bo Xiong
Jiahong Liu
Irwin King
Rex Ying

Large language models (LLMs) have demonstrated remarkable performance across various tasks. However, it remains an open question whether the default Euclidean space is the most suitable choice for LLMs. In this study, we investigate the geometric characteristics of LLMs, focusing specifically on tokens and their embeddings. Our findings reveal that token frequency follows a power-law distribution, where high-frequency tokens (e. g. , the, that ) constitute the minority, while low-frequency tokens (e. g. , apple, dog) constitute the majority. Furthermore, high-frequency tokens cluster near the origin, whereas low-frequency tokens are positioned farther away in the embedding space. Additionally, token embeddings exhibit hyperbolic characteristics, indicating a latent tree-like structure within the embedding space. Motivated by these observations, we propose HypLoRA, an efficient fine-tuning approach that operates in hyperbolic space to exploit these underlying hierarchical structures better. HypLoRA performs low-rank adaptation directly in hyperbolic space, thereby preserving hyperbolic modeling capabilities throughout the fine-tuning process. Extensive experiments across various base models and reasoning benchmarks, specifically arithmetic and commonsense reasoning tasks, demonstrate that HypLoRA substantially improves LLM performance.

PDF Details

AAAI Conference 2024 Conference Paper

NestE: Modeling Nested Relational Structures for Knowledge Graph Reasoning

Bo Xiong
Mojtaba Nayyeri
Linhao Luo
Zihao Wang
Shirui Pan
Steffen Staab

Reasoning with knowledge graphs (KGs) has primarily focused on triple-shaped facts. Recent advancements have been explored to enhance the semantics of these facts by incorporating more potent representations, such as hyper-relational facts. However, these approaches are limited to atomic facts, which describe a single piece of information. This paper extends beyond atomic facts and delves into nested facts, represented by quoted triples where subjects and objects are triples themselves (e.g., ((BarackObama, holds_position, President), succeed_by, (DonaldTrump, holds_position, President))). These nested facts enable the expression of complex semantics like situations over time and logical patterns} over entities and relations. In response, we introduce NestE, a novel KG embedding approach that captures the semantics of both atomic and nested factual knowledge. NestE represents each atomic fact as a 1*3 matrix, and each nested relation is modeled as a 3*3 matrix that rotates the 1*3 atomic fact matrix through matrix multiplication. Each element of the matrix is represented as a complex number in the generalized 4D hypercomplex space, including (spherical) quaternions, hyperbolic quaternions, and split-quaternions. Through thorough analysis, we demonstrate the embedding's efficacy in capturing diverse logical patterns over nested facts, surpassing the confines of first-order logic-like expressions. Our experimental results showcase NestE's significant performance gains over current baselines in triple prediction and conditional link prediction. The code and pre-trained models are open available at https://github.com/xiongbo010/NestE.

PDF Details DOI

NeurIPS Conference 2022 Conference Paper

Hyperbolic Embedding Inference for Structured Multi-Label Prediction

Bo Xiong
Michael Cochez
Mojtaba Nayyeri
Steffen Staab

We consider a structured multi-label prediction problem where the labels are organized under implication and mutual exclusion constraints. A major concern is to produce predictions that are logically consistent with these constraints. To do so, we formulate this problem as an embedding inference problem where the constraints are imposed onto the embeddings of labels by geometric construction. Particularly, we consider a hyperbolic Poincaré ball model in which we encode labels as Poincaré hyperplanes that work as linear decision boundaries. The hyperplanes are interpreted as convex regions such that the logical relationships (implication and exclusion) are geometrically encoded using the insideness and disjointedness of these regions, respectively. We show theoretical groundings of the method for preserving logical relationships in the embedding space. Extensive experiments on 12 datasets show 1) significant improvements in mean average precision; 2) lower number of constraint violations; 3) an order of magnitude fewer dimensions than baselines.

PDF Details

NeurIPS Conference 2022 Conference Paper

Pseudo-Riemannian Graph Convolutional Networks

Bo Xiong
Shichao Zhu
Nico Potyka
Shirui Pan
Chuan Zhou
Steffen Staab

Graph Convolutional Networks (GCNs) are powerful frameworks for learning embeddings of graph-structured data. GCNs are traditionally studied through the lens of Euclidean geometry. Recent works find that non-Euclidean Riemannian manifolds provide specific inductive biases for embedding hierarchical or spherical data. However, they cannot align well with data of mixed graph topologies. We consider a larger class of pseudo-Riemannian manifolds that generalize hyperboloid and sphere. We develop new geodesic tools that allow for extending neural network operations into geodesically disconnected pseudo-Riemannian manifolds. As a consequence, we derive a pseudo-Riemannian GCN that models data in pseudo-Riemannian manifolds of constant nonzero curvature in the context of graph neural networks. Our method provides a geometric inductive bias that is sufficiently flexible to model mixed heterogeneous topologies like hierarchical graphs with cycles. We demonstrate the representational capabilities of this method by applying it to the tasks of graph reconstruction, node classification, and link prediction on a series of standard graphs with mixed topologies. Empirical results demonstrate that our method outperforms Riemannian counterparts when embedding graphs of complex topologies.

PDF Details

ICRA Conference 2013 Conference Paper

Exploiting domain knowledge for Object Discovery

Alvaro Collet
Bo Xiong
Corina Gurau
Martial Hebert
Siddhartha S. Srinivasa

In this paper, we consider the problem of Lifelong Robotic Object Discovery (LROD) as the long-term goal of discovering novel objects in the environment while the robot operates, for as long as the robot operates. As a first step towards LROD, we automatically process the raw video stream of an entire workday of a robotic agent to discover objects. We claim that the key to achieve this goal is to incorporate domain knowledge whenever available, in order to detect and adapt to changes in the environment. We propose a general graph-based formulation for LROD in which generic domain knowledge is encoded as constraints. Our formulation enables new sources of domain knowledge—metadata—to be added dynamically to the system, as they become available or as conditions change. By adding domain knowledge, we discover 2. 7· more objects and decrease processing time 190 times. Our optimized implementation, HerbDisc, processes 6 h 20 min of RGBD video of real human environments in 18 min 30 s, and discovers 121 correct novel objects with their 3D models.

Details

ICRA Conference 2005 Conference Paper

Group Elevator Scheduling with Advanced Traffic Information for Normal Operations and Coordinated Emergency Evacuation

Bo Xiong
Peter B. Luh
Shi-Chung Chang

In a building, effective operations of transportation systems including elevators, escalators, and stairs are vital. Among them, group elevator scheduling has long been recognized as an important issue for transportation efficiency. The problem, however, is difficult because of the large state space, various traffic profiles, and uncertainties. With the progress in information technology and sensor networks, one potential way is to use advanced traffic information to reduce uncertainties and optimize the performance. How to effectively utilize such information remains an open and challenging issue. This paper presents the optimized scheduling of a group of elevators with advanced traffic information for normal operations and coordinated emergency evacuation. A look-ahead time window is first introduced to model advanced information. Key characteristics of group elevator scheduling are abstracted to establish an innovative formulation. The objective function is transformed into an additive form to facilitate the decomposition of the problem into individual car subproblems. Subproblems are independently solved by using a local search method in conjunction with dynamic programming with a novel definition of stages, states, decisions, and costs to optimize single car dispatching. With surrogate optimization, local search is “good enough” to set multiplier updating directions. Individual cars are then coordinated through the updating of multipliers by using surrogate optimization for near-optimal solutions. Numerical testing results demonstrate that near-optimal solutions are obtained for problems of moderate sizes under selected traffic patterns. The results also show the value of advanced information through testing different window sizes and rescheduling intervals.

Details

IROS Conference 2004 Conference Paper

An inventory control policy for maintenance networks

Ming Ni
Peter B. Luh
Bo Xiong
Shi-Chung Chang

Many industries rely on maintenance networks to maintain their key assets, and a key characteristic of such maintenance services is the wide use of rotable parts. In view of today's time-based competition, efficiently managing the rotable inventory becomes imperative for a maintenance network to achieve short turn-around-times and low costs. This, however, is difficult in view of complicated and uncertain maintenance processes, and hybrid inventory replenishments from both new and refurbished parts. As the information on demands and replenishments is highly dependent on maintenance processes, and can be obtained, utilizing this information opens a new way to improve the inventory operational efficiency. This paper presents a new approach for rotable inventory control with the demand and replenishment information. Key characteristics of maintenance processes and hybrid replenishments are abstracted to form a novel model within the context of stochastic optimal control, where the demand and replenishment information is incorporated in the system state. Comparing to traditional approaches with a large-size augmented state, an aggregated state variable is defined based on inventory dynamics to reduce the number of required state components. A solution methodology based on stochastic dynamic programming (SDP) is developed, with stage-wise costs obtained in terms of aggregated state variables. Steady state solutions are computed offline, and are stored as inventory policies to be implemented by table lookup. Simulation results demonstrate the effectiveness of the approach on reducing inventory costs with the demand and replenishment information.

Details