Li Fuxin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Li Fuxin has a total of 166 co-authors. Here are the top co-authors by number of shared publications:
- Jialin Yuan
Oregon State University
7 shared publications - Damanpreet Kaur
Oregon State University
4 shared publications - Michael F. Nagle
Department of Forest Ecosystems and Society, Oregon State University
4 shared publications - Steven H. Strauss
Department of Forest Ecosystems and Society, Oregon State University, Corvallis, USA
4 shared publications - Ali Behnoudfar
Oregon State University
3 shared publications - Cathleen Ma
Department of Forest Ecosystems & Society Oregon State University Corvallis OR 97331 USA
3 shared publications
Latest Publications
- Open Access
- 3 November 2026
- Published by MDPI AG in Preprints.org
Agentic Laboratories of the Future: Towards World Models for Scientific Discovery
Advances in scientific AI have accelerated hypothesis generation, prediction, and experimental design, but translating these capabilities into reliable physical experimentation remains a major bottleneck in scientific discovery. Although automation, high-throughput experimentation, foundation models, and cloud infrastructure have accelerated individual stages of the scientific workflow, they have not unified the discovery process. Scientists must still coordinate between computational models and physical laboratory systems to translate scientific designs into executable experiments and ensure their results are reliable. We argue that the next generation of laboratories will be agentic: environments in which scientists, AI systems, equipment and robotic platforms operate as collaborative discovery partners. We propose an architecture built around two complementary components: an agentic foundation that translates scientific designs into coordinated execution plans, and an evolving laboratory world model that represents physical state, predicts action outcomes, and incorporates experimental feedback. Together, they connect computational reasoning to observable and verifiable physical execution. Multimodal perception, robotics, and human–AI interfaces can make laboratory actions and evolving experimental conditions accessible to these systems, supporting adaptive execution and learning. We further propose an evaluation framework centered on physical grounding, execution fidelity, scientific validity, and adaptive coordination under human oversight, rather than autonomy alone. This Perspective outlines how agentic laboratories could connect scientific reasoning, physical experimentation, and trustworthy evidence within a human-in-the-lead discovery loop.
- 3 November 2026
- Published by MDPI AG in Preprints.org
Open AccessAgentic Laboratories of the Future: Towards World Models for Scientific Discovery
Advances in scientific AI have accelerated hypothesis generation, prediction, and experimental design, but translating these capabilities into reliable physical experimentation remains a major bottleneck in scientific discovery. Although automation, high-throughput experimentation, foundation models, and cloud infrastructure have accelerated individual stages of the scientific workflow, they have not unified the discovery process. Scientists must still coordinate between computational models and physical laboratory systems to translate scientific designs into executable experiments and ensure their results are reliable. We argue that the next generation of laboratories will be agentic: environments in which scientists, AI systems, equipment and robotic platforms operate as collaborative discovery partners. We propose an architecture built around two complementary components: an agentic foundation that translates scientific designs into coordinated execution plans, and an evolving laboratory world model that represents physical state, predicts action outcomes, and incorporates experimental feedback. Together, they connect computational reasoning to observable and verifiable physical execution. Multimodal perception, robotics, and human–AI interfaces can make laboratory actions and evolving experimental conditions accessible to these systems, supporting adaptive execution and learning. We further propose an evaluation framework centered on physical grounding, execution fidelity, scientific validity, and adaptive coordination under human oversight, rather than autonomy alone. This Perspective outlines how agentic laboratories could connect scientific reasoning, physical experimentation, and trustworthy evidence within a human-in-the-lead discovery loop.
- 2 September 2025
- Published by Taylor & Francis in Engineering Studies
- Vol. 17 (3) , 184-198
Breaking Boundaries through Collaboration: A Human-Centered Framework for Fair AI Design
- 2 September 2025
- Published by Taylor & Francis in Engineering Studies
- Vol. 17 (3) , 184-198
Breaking Boundaries through Collaboration: A Human-Centered Framework for Fair AI Design
- Open Access
Enhancing Diversity in Bayesian Deep Learning via Hyperspherical Energy Minimization of CKA
Particle-based Bayesian deep learning often requires a similarity metric to compare two networks. However, naive similarity metrics lack permutation invariance and are inappropriate for comparing networks. Centered Kernel Alignment (CKA) on feature kernels has been proposed to compare deep networks but has not been used as an optimization objective in Bayesian deep learning. In this paper, we explore the use of CKA in Bayesian deep learning to generate diverse ensembles and hypernetworks that output a network posterior. Noting that CKA projects kernels onto a unit hypersphere and that directly optimizing the CKA objective leads to diminishing gradients when two networks are very similar. We propose adopting the approach of hyperspherical energy (HE) on top of CKA kernels to address this drawback and improve training stability. Additionally, by leveraging CKA-based feature kernels, we derive feature repulsive terms applied to synthetically generated outlier examples. Experiments on both diverse ensembles and hypernetworks show that our approach significantly outperforms baselines in terms of uncertainty quantification in both synthetic and realistic outlier detection tasks.
Open AccessEnhancing Diversity in Bayesian Deep Learning via Hyperspherical Energy Minimization of CKA
Particle-based Bayesian deep learning often requires a similarity metric to compare two networks. However, naive similarity metrics lack permutation invariance and are inappropriate for comparing networks. Centered Kernel Alignment (CKA) on feature kernels has been proposed to compare deep networks but has not been used as an optimization objective in Bayesian deep learning. In this paper, we explore the use of CKA in Bayesian deep learning to generate diverse ensembles and hypernetworks that output a network posterior. Noting that CKA projects kernels onto a unit hypersphere and that directly optimizing the CKA objective leads to diminishing gradients when two networks are very similar. We propose adopting the approach of hyperspherical energy (HE) on top of CKA kernels to address this drawback and improve training stability. Additionally, by leveraging CKA-based feature kernels, we derive feature repulsive terms applied to synthetically generated outlier examples. Experiments on both diverse ensembles and hypernetworks show that our approach significantly outperforms baselines in terms of uncertainty quantification in both synthetic and realistic outlier detection tasks.
- Open Access
- 22 April 2024
- Published by Wiley in New Phytologist
- Vol. 242 (5) , 2059-2076
Genome‐wide association study and network analysis of in vitro transformation in Populus trichocarpa support key roles of diverse phytohormone pathways and cross talk
Summary: Wide variation in amenability to transformation and regeneration (TR) among many plant species and genotypes presents a challenge to the use of genetic engineering in research and breeding. To help understand the causes of this variation, we performed association mapping and network analysis using a population of 1204 wild trees of Populus trichocarpa (black cottonwood). To enable precise and high‐throughput phenotyping of callus and shoot TR, we developed a computer vision system that cross‐referenced complementary red, green, and blue (RGB) and fluorescent‐hyperspectral images. We performed association mapping using single‐marker and combined variant methods, followed by statistical tests for epistasis and integration of published multi‐omic datasets to identify likely regulatory hubs. We report 409 candidate genes implicated by associations within 5 kb of coding sequences, and epistasis tests implicated 81 of these candidate genes as regulators of one another. Gene ontology terms related to protein–protein interactions and transcriptional regulation are overrepresented, among others. In addition to auxin and cytokinin pathways long established as critical to TR, our results highlight the importance of stress and wounding pathways. Potential regulatory hubs of signaling within and across these pathways include GROWTH REGULATORY FACTOR 1 ( GRF1 ), PHOSPHATIDYLINOSITOL 4‐KINASE β1 ( PI‐4Kβ1 ), and OBF‐BINDING PROTEIN 1 ( OBP1 ).
- 22 April 2024
- Published by Wiley in New Phytologist
- Vol. 242 (5) , 2059-2076
Open AccessGenome‐wide association study and network analysis of in vitro transformation in Populus trichocarpa support key roles of diverse phytohormone pathways and cross talk
Summary: Wide variation in amenability to transformation and regeneration (TR) among many plant species and genotypes presents a challenge to the use of genetic engineering in research and breeding. To help understand the causes of this variation, we performed association mapping and network analysis using a population of 1204 wild trees of Populus trichocarpa (black cottonwood). To enable precise and high‐throughput phenotyping of callus and shoot TR, we developed a computer vision system that cross‐referenced complementary red, green, and blue (RGB) and fluorescent‐hyperspectral images. We performed association mapping using single‐marker and combined variant methods, followed by statistical tests for epistasis and integration of published multi‐omic datasets to identify likely regulatory hubs. We report 409 candidate genes implicated by associations within 5 kb of coding sequences, and epistasis tests implicated 81 of these candidate genes as regulators of one another. Gene ontology terms related to protein–protein interactions and transcriptional regulation are overrepresented, among others. In addition to auxin and cytokinin pathways long established as critical to TR, our results highlight the importance of stress and wounding pathways. Potential regulatory hubs of signaling within and across these pathways include GROWTH REGULATORY FACTOR 1 ( GRF1 ), PHOSPHATIDYLINOSITOL 4‐KINASE β1 ( PI‐4Kβ1 ), and OBF‐BINDING PROTEIN 1 ( OBP1 ).
- Open Access
Object Dynamics Modeling with Hierarchical Point Cloud-based Representations
Modeling object dynamics with a neural network is an important problem with numerous applications. Most recent work has been based on graph neural networks. However, physics happens in 3D space, where geometric information potentially plays an important role in modeling physical phenomena. In this work, we propose a novel U-net architecture based on continuous point convolution which naturally embeds information from 3D coordinates and allows for multi-scale feature representations with established downsampling and upsampling procedures. Bottleneck layers in the downsampled point clouds lead to better long-range interaction modeling. Besides, the flexibility of point convolutions allows our approach to generalize to sparsely sampled points from mesh vertices and dynamically generate features on important interaction points on mesh faces. Experimental results demonstrate that our approach significantly improves the state-of-the-art, especially in scenarios that require accurate gravity or collision reasoning.
Open AccessObject Dynamics Modeling with Hierarchical Point Cloud-based Representations
Modeling object dynamics with a neural network is an important problem with numerous applications. Most recent work has been based on graph neural networks. However, physics happens in 3D space, where geometric information potentially plays an important role in modeling physical phenomena. In this work, we propose a novel U-net architecture based on continuous point convolution which naturally embeds information from 3D coordinates and allows for multi-scale feature representations with established downsampling and upsampling procedures. Bottleneck layers in the downsampled point clouds lead to better long-range interaction modeling. Besides, the flexibility of point convolutions allows our approach to generalize to sparsely sampled points from mesh vertices and dynamically generate features on important interaction points on mesh faces. Experimental results demonstrate that our approach significantly improves the state-of-the-art, especially in scenarios that require accurate gravity or collision reasoning.
- 30 January 2024
- Published by Elsevier in Computer Vision and Image Understanding
- Vol. 241
Space–time recurrent memory network
- 30 January 2024
- Published by Elsevier in Computer Vision and Image Understanding
- Vol. 241
Space–time recurrent memory network
- Open Access
Diverse Shape Completion via Style Modulated Generative Adversarial Networks
Shape completion aims to recover the full 3D geometry of an object from a partial observation. This problem is inherently multi-modal since there can be many ways to plausibly complete the missing regions of a shape. Such diversity would be indicative of the underlying uncertainty of the shape and could be preferable for downstream tasks such as planning. In this paper, we propose a novel conditional generative adversarial network that can produce many diverse plausible completions of a partially observed point cloud. To enable our network to produce multiple completions for the same partial input, we introduce stochasticity into our network via style modulation. By extracting style codes from complete shapes during training, and learning a distribution over them, our style codes can explicitly carry shape category information leading to better completions. We further introduce diversity penalties and discriminators at multiple scales to prevent conditional mode collapse and to train without the need for multiple ground truth completions for each partial input. Evaluations across several synthetic and real datasets demonstrate that our method achieves significant improvements in respecting the partial observations while obtaining greater diversity in completions.
Open AccessDiverse Shape Completion via Style Modulated Generative Adversarial Networks
Shape completion aims to recover the full 3D geometry of an object from a partial observation. This problem is inherently multi-modal since there can be many ways to plausibly complete the missing regions of a shape. Such diversity would be indicative of the underlying uncertainty of the shape and could be preferable for downstream tasks such as planning. In this paper, we propose a novel conditional generative adversarial network that can produce many diverse plausible completions of a partially observed point cloud. To enable our network to produce multiple completions for the same partial input, we introduce stochasticity into our network via style modulation. By extracting style codes from complete shapes during training, and learning a distribution over them, our style codes can explicitly carry shape category information leading to better completions. We further introduce diversity penalties and discriminators at multiple scales to prevent conditional mode collapse and to train without the need for multiple ground truth completions for each partial input. Evaluations across several synthetic and real datasets demonstrate that our method achieves significant improvements in respecting the partial observations while obtaining greater diversity in completions.
- Open Access
- 14 June 2023
- Published by Oxford University Press (OUP) in Horticulture Research
- Vol. 10 (8) , uhad125
GWAS identifies candidate genes controlling adventitious rooting in Populus trichocarpa
Adventitious rooting (AR) is critical to the propagation, breeding, and genetic engineering of trees. The capacity for plants to undergo this process is highly heritable and of a polygenic nature; however, the basis of its genetic variation is largely uncharacterized. To identify genetic regulators of AR, we performed a genome-wide association study (GWAS) using 1148 genotypes of Populus trichocarpa. GWASs are often limited by the abilities of researchers to collect precise phenotype data on a high-throughput scale; to help overcome this limitation, we developed a computer vision system to measure an array of traits related to adventitious root development in poplar, including temporal measures of lateral and basal root length and area. GWAS was performed using multiple methods and significance thresholds to handle non-normal phenotype statistics and to gain statistical power. These analyses yielded a total of 277 unique associations, suggesting that genes that control rooting include regulators of hormone signaling, cell division and structure, reactive oxygen species signaling, and other processes with known roles in root development. Numerous genes with uncharacterized functions and/or cryptic roles were also identified. These candidates provide targets for functional analysis, including physiological and epistatic analyses, to better characterize the complex polygenic regulation of AR.
- 14 June 2023
- Published by Oxford University Press (OUP) in Horticulture Research
- Vol. 10 (8) , uhad125
Open AccessGWAS identifies candidate genes controlling adventitious rooting in Populus trichocarpa
Adventitious rooting (AR) is critical to the propagation, breeding, and genetic engineering of trees. The capacity for plants to undergo this process is highly heritable and of a polygenic nature; however, the basis of its genetic variation is largely uncharacterized. To identify genetic regulators of AR, we performed a genome-wide association study (GWAS) using 1148 genotypes of Populus trichocarpa. GWASs are often limited by the abilities of researchers to collect precise phenotype data on a high-throughput scale; to help overcome this limitation, we developed a computer vision system to measure an array of traits related to adventitious root development in poplar, including temporal measures of lateral and basal root length and area. GWAS was performed using multiple methods and significance thresholds to handle non-normal phenotype statistics and to gain statistical power. These analyses yielded a total of 277 unique associations, suggesting that genes that control rooting include regulators of hormone signaling, cell division and structure, reactive oxygen species signaling, and other processes with known roles in root development. Numerous genes with uncharacterized functions and/or cryptic roles were also identified. These candidates provide targets for functional analysis, including physiological and epistatic analyses, to better characterize the complex polygenic regulation of AR.
- Open Access
Maximal Cliques on Multi-Frame Proposal Graph for Unsupervised Video Object Segmentation
Unsupervised Video Object Segmentation (UVOS) aims at discovering objects and tracking them through videos. For accurate UVOS, we observe if one can locate precise segment proposals on key frames, subsequent processes are much simpler. Hence, we propose to reason about key frame proposals using a graph built with the object probability masks initially generated from multiple frames around the key frame and then propagated to the key frame. On this graph, we compute maximal cliques, with each clique representing one candidate object. By making multiple proposals in the clique to vote for the key frame proposal, we obtain refined key frame proposals that could be better than any of the single-frame proposals. A semi-supervised VOS algorithm subsequently tracks these key frame proposals to the entire video. Our algorithm is modular and hence can be used with any instance segmentation and semi-supervised VOS algorithm. We achieve state-of-the-art performance on the DAVIS-2017 validation and test-dev dataset. On the related problem of video instance segmentation, our method shows competitive performance with the previous best algorithm that requires joint training with the VOS algorithm.
Open AccessMaximal Cliques on Multi-Frame Proposal Graph for Unsupervised Video Object Segmentation
Unsupervised Video Object Segmentation (UVOS) aims at discovering objects and tracking them through videos. For accurate UVOS, we observe if one can locate precise segment proposals on key frames, subsequent processes are much simpler. Hence, we propose to reason about key frame proposals using a graph built with the object probability masks initially generated from multiple frames around the key frame and then propagated to the key frame. On this graph, we compute maximal cliques, with each clique representing one candidate object. By making multiple proposals in the clique to vote for the key frame proposal, we obtain refined key frame proposals that could be better than any of the single-frame proposals. A semi-supervised VOS algorithm subsequently tracks these key frame proposals to the entire video. Our algorithm is modular and hence can be used with any instance segmentation and semi-supervised VOS algorithm. We achieve state-of-the-art performance on the DAVIS-2017 validation and test-dev dataset. On the related problem of video instance segmentation, our method shows competitive performance with the previous best algorithm that requires joint training with the VOS algorithm.
- 22 October 2022
- Published by Springer Nature
- p. 612-629
BATMAN: Bilateral Attention Transformer in Motion-Appearance Neighboring Space for Video Object Segmentation
- 22 October 2022
- Published by Springer Nature
- p. 612-629
BATMAN: Bilateral Attention Transformer in Motion-Appearance Neighboring Space for Video Object Segmentation