Henry Cerbone
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Henry Cerbone has a total of 66 co-authors. Here are the top co-authors by number of shared publications:
- Rachel H. Parkinson
Department of Biology, University of Oxford, Oxford, UK
3 shared publications - Alasdair D. Wilson
OxRSE, Doctoral Training Centre, University of Oxford, Oxford OX1 3NP, UK
2 shared publications - Andrew F. Brown
Institute of Bee Health, Vetsuisse Faculty, University of Bern, Bern, Switzerland
2 shared publications - Andrew K. Jones
Oxford Brookes University Department of Biological and Medical Sciences Oxford
2 shared publications - Angela Hayward
Centre for Ecology and Conservation, University of Exeter, Penryn, UK
2 shared publications - Ben Lambert
Department of Statistics, University of Oxford, Oxford, UK
2 shared publications
Latest Publications
- Open Access
- 21 May 2026
- Published by Elsevier in Ecological Informatics
- Vol. 96
MetaBeeAI: An AI pipeline for structured evidence extraction from biological literature
- 21 May 2026
- Published by Elsevier in Ecological Informatics
- Vol. 96
Open AccessMetaBeeAI: An AI pipeline for structured evidence extraction from biological literature
- Open Access
An Account of the Ecology of the Parasitic Plant Cistanche phelypaea (L.) Cout. (Orobanchaceae) in the Canary Islands and Implications for Its Conservation
Parasitic plants are ecologically important because they can exert a profound influence on the surrounding ecosystem. Yet the ecology and host specificity of most parasitic plant species remain poorly known or undocumented. Cistanche phelypaea is a local and elusive parasitic plant in the Canary Islands. We carried out the first qualitative assessment of this plant’s ecology on the islands by examining 10 subpopulations over a 7-year period. We examined aspects of the plant’s ecology, distribution, and specificity for eight potential host species. Our observations suggest that four species are hosts: Afrosalsola divaricata, Bassia tomentosa, Suaeda vera, and Traganum moquinii, all of which are shrubby Amaranthaceae; however, host specificity varies across the plant’s range. Afrosalsola divaricata was inferred to be the predominant host and was parasitised wherever it co-occurred with the parasite (50% of sites). Cases of inferred parasitism on more than one host species at a given site were rare. Eight of the ten subpopulations occur in areas of high footfall or close to urbanisation; some disturbance, if managed sensitively, appears to favour recruitment and population dynamics. Based on our observations, we suggest that the integration of species distribution models (SDMs) with targeted surveys would be a promising route for scaling up from site-level observations to island-wide inference. We lay the groundwork for practical recommendations informed by such surveys; together with our long-term observations on host range, this offers a template for parasitic plant conservation more broadly.
Open AccessAn Account of the Ecology of the Parasitic Plant Cistanche phelypaea (L.) Cout. (Orobanchaceae) in the Canary Islands and Implications for Its Conservation
Parasitic plants are ecologically important because they can exert a profound influence on the surrounding ecosystem. Yet the ecology and host specificity of most parasitic plant species remain poorly known or undocumented. Cistanche phelypaea is a local and elusive parasitic plant in the Canary Islands. We carried out the first qualitative assessment of this plant’s ecology on the islands by examining 10 subpopulations over a 7-year period. We examined aspects of the plant’s ecology, distribution, and specificity for eight potential host species. Our observations suggest that four species are hosts: Afrosalsola divaricata, Bassia tomentosa, Suaeda vera, and Traganum moquinii, all of which are shrubby Amaranthaceae; however, host specificity varies across the plant’s range. Afrosalsola divaricata was inferred to be the predominant host and was parasitised wherever it co-occurred with the parasite (50% of sites). Cases of inferred parasitism on more than one host species at a given site were rare. Eight of the ten subpopulations occur in areas of high footfall or close to urbanisation; some disturbance, if managed sensitively, appears to favour recruitment and population dynamics. Based on our observations, we suggest that the integration of species distribution models (SDMs) with targeted surveys would be a promising route for scaling up from site-level observations to island-wide inference. We lay the groundwork for practical recommendations informed by such surveys; together with our long-term observations on host range, this offers a template for parasitic plant conservation more broadly.
- Open Access
- 5 March 2026
- Published by California Digital Library (CDL) in EcoEvoRxiv
Towards ecologically meaningful foundation models
Ecology aims to explain and predict how organisms interact with each other and their environments across space and time. Yet both ecological data and theory are fragmented, leading to models that generalise poorly beyond specific systems or scales. Empirical evidence spans diverse modalities, resolutions and contexts, while theory is distributed across partially overlapping frameworks that are rarely integrated within a single predictive model. We argue that ecological foundation models (ecoFMs), trained on large, multimodal ecological datasets, offer a route toward unifying data and theory within a common framework. Beyond their ecological value, ecoFMs present a challenging and consequential testbed for machine learning, demanding advances in multimodal representation learning, theory-guided modelling, and uncertainty-aware inference. By learning shared representations of organisms, environments and interactions, ecoFMs could improve generalisation, link pattern to process, and enable synthesis across ecological sub-disciplines. We outline a roadmap for developing ecoFMs, including requirements for data infrastructure, model architectures, evaluation strategies and governance, and assess where current machine learning and ecological approaches fall short. If developed responsibly and collaboratively with ecological practitioners and other actors, ecoFMs could enable new modes of analysis and strengthen ecological forecasting, while simultaneously driving advances in machine learning for multimodal integration, theory-guided learning, and generalisation in complex, data-limited systems.
- 5 March 2026
- Published by California Digital Library (CDL) in EcoEvoRxiv
Open AccessTowards ecologically meaningful foundation models
Ecology aims to explain and predict how organisms interact with each other and their environments across space and time. Yet both ecological data and theory are fragmented, leading to models that generalise poorly beyond specific systems or scales. Empirical evidence spans diverse modalities, resolutions and contexts, while theory is distributed across partially overlapping frameworks that are rarely integrated within a single predictive model. We argue that ecological foundation models (ecoFMs), trained on large, multimodal ecological datasets, offer a route toward unifying data and theory within a common framework. Beyond their ecological value, ecoFMs present a challenging and consequential testbed for machine learning, demanding advances in multimodal representation learning, theory-guided modelling, and uncertainty-aware inference. By learning shared representations of organisms, environments and interactions, ecoFMs could improve generalisation, link pattern to process, and enable synthesis across ecological sub-disciplines. We outline a roadmap for developing ecoFMs, including requirements for data infrastructure, model architectures, evaluation strategies and governance, and assess where current machine learning and ecological approaches fall short. If developed responsibly and collaboratively with ecological practitioners and other actors, ecoFMs could enable new modes of analysis and strengthen ecological forecasting, while simultaneously driving advances in machine learning for multimodal integration, theory-guided learning, and generalisation in complex, data-limited systems.
- Open Access
MetaBeeAI: an AI pipeline for structured evidence extraction from biological literature
The volume and complexity of scientific literature are expanding rapidly, making it increasingly difficult to extract and synthesise information across studies. This challenge is particularly acute in the biological sciences, where evidence spans multiple levels of organisation and heterogeneous experimental designs. Large Language Model (LLM) pipelines offer a scalable route to evidence synthesis, but many existing approaches lack transparency, modularity, and effective mechanisms for human oversight. We present MetaBeeAI, an open-source, modular pipeline that integrates established LLM techniques into a coherent, auditable workflow for structured data extraction in biology. MetaBeeAI combines modular prompting, multi-pass extraction, and expert-in-the-loop validation within an interface that presents model outputs alongside source text, enabling inspection, correction, and iterative refinement. The pipeline produces machine-readable records of prompts, configurations, and expert annotations, supporting reproducibility and continuous improvement. We apply MetaBeeAI to 924 research papers on bees and pesticides, extracting structured information on species, compounds, exposure designs, and experimental context. Evaluation demonstrates improved consistency, convergence with expert judgement, and robustness across heterogeneous biological studies, highlighting the value of expert-guided refinement. MetaBeeAI provides a transparent and extensible framework for scalable evidence synthesis, supporting reliable integration of LLMs into biological research workflows.
Open AccessMetaBeeAI: an AI pipeline for structured evidence extraction from biological literature
The volume and complexity of scientific literature are expanding rapidly, making it increasingly difficult to extract and synthesise information across studies. This challenge is particularly acute in the biological sciences, where evidence spans multiple levels of organisation and heterogeneous experimental designs. Large Language Model (LLM) pipelines offer a scalable route to evidence synthesis, but many existing approaches lack transparency, modularity, and effective mechanisms for human oversight. We present MetaBeeAI, an open-source, modular pipeline that integrates established LLM techniques into a coherent, auditable workflow for structured data extraction in biology. MetaBeeAI combines modular prompting, multi-pass extraction, and expert-in-the-loop validation within an interface that presents model outputs alongside source text, enabling inspection, correction, and iterative refinement. The pipeline produces machine-readable records of prompts, configurations, and expert annotations, supporting reproducibility and continuous improvement. We apply MetaBeeAI to 924 research papers on bees and pesticides, extracting structured information on species, compounds, exposure designs, and experimental context. Evaluation demonstrates improved consistency, convergence with expert judgement, and robustness across heterogeneous biological studies, highlighting the value of expert-guided refinement. MetaBeeAI provides a transparent and extensible framework for scalable evidence synthesis, supporting reliable integration of LLMs into biological research workflows.
- 30 September 2025
- Published by The Company of Biologists in Journal of Experimental Biology
- Vol. 228 (19)
Extreme insect fashion is electric
- 30 September 2025
- Published by The Company of Biologists in Journal of Experimental Biology
- Vol. 228 (19)
Extreme insect fashion is electric
- Open Access
- 30 September 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-12
Optimized user-guided motion control of modular robots
Transferring motion instructions from a user enables robots to perform new and unforeseen operations. Robot collectives, in particular, offer greater adaptability to changing tasks and environments. However, effectively transferring motion instructions becomes challenging as the collective’s shape and size evolve. These changes often require additional system constraints to maintain robust motion control, which typically depends on pre-programmed knowledge of new tasks, ultimately limiting the collective’s adaptability. To overcome the above challenges, we propose a physical and computational platform for user-guided control of self-reconfigurable modular robots. This platform consists of an optimization scheme for online processing of user commands, which prevents any modular robot actions that would violate system or environment constraints. The second component consists of Joint-space Joysticks, which match the robot’s morphology, enabling the user to control diverse and dynamically changing modular robot structures through direct physical interaction. We present a platform that enables users to safely control modular, shape-changing robots through a physical interface. We demonstrate the platform’s efficacy and generalizability across a diverse set of modular robot morphologies using two independent robotic systems — Mori3 and Roombots — performing a range of tasks including pick-and-place, human assistance, legged locomotion, and workspace expansion.
- 30 September 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-12
Open AccessOptimized user-guided motion control of modular robots
Transferring motion instructions from a user enables robots to perform new and unforeseen operations. Robot collectives, in particular, offer greater adaptability to changing tasks and environments. However, effectively transferring motion instructions becomes challenging as the collective’s shape and size evolve. These changes often require additional system constraints to maintain robust motion control, which typically depends on pre-programmed knowledge of new tasks, ultimately limiting the collective’s adaptability. To overcome the above challenges, we propose a physical and computational platform for user-guided control of self-reconfigurable modular robots. This platform consists of an optimization scheme for online processing of user commands, which prevents any modular robot actions that would violate system or environment constraints. The second component consists of Joint-space Joysticks, which match the robot’s morphology, enabling the user to control diverse and dynamically changing modular robot structures through direct physical interaction. We present a platform that enables users to safely control modular, shape-changing robots through a physical interface. We demonstrate the platform’s efficacy and generalizability across a diverse set of modular robot morphologies using two independent robotic systems — Mori3 and Roombots — performing a range of tasks including pick-and-place, human assistance, legged locomotion, and workspace expansion.
- 27 June 2025
- Published by The Company of Biologists in Journal of Experimental Biology
- Vol. 228 (13)
The (virtual) flight of the fruit fly
- 27 June 2025
- Published by The Company of Biologists in Journal of Experimental Biology
- Vol. 228 (13)
The (virtual) flight of the fruit fly
- Open Access
- 26 June 2025
- Published by California Digital Library (CDL) in EcoEvoRxiv
Democratizing 3D ecology: Mobile neural radiance field for scalable ecosystem mapping in change detection
High-resolution, three-dimensional monitoring is increasingly essential for capturing ecological dynamics, yet conventional approaches such as terrestrial laser scanning (TLS) and photogrammetry remain limited by cost, accessibility, and technical barriers. Here, we introduce and evaluate the application of mobile neural radiance field (NeRF) methods for ecological research. Leveraging consumer-grade smartphones and open-source platforms (e.g. Luma AI), we demonstrate that mobile NeRFs can reconstruct detailed 3D structures of vegetation with accuracy comparable to TLS in open-canopy environments. We assess the strengths and limitations of NeRFs across habitat types, showing that while performance declines under occlusion (e.g. dense canopies), these methods excel at capturing understory complexity, making them particularly valuable for savannas, grasslands, and urban systems. We further explore the potential of radiance fields to integrate hyperspectral and robotic data streams, expanding their utility for dynamic ecosystem monitoring. By reducing hardware requirements and broadening participation, mobile NeRFs offer a promising avenue for democratising ecological data collection and advancing scalable environmental surveillance.
- 26 June 2025
- Published by California Digital Library (CDL) in EcoEvoRxiv
Open AccessDemocratizing 3D ecology: Mobile neural radiance field for scalable ecosystem mapping in change detection
High-resolution, three-dimensional monitoring is increasingly essential for capturing ecological dynamics, yet conventional approaches such as terrestrial laser scanning (TLS) and photogrammetry remain limited by cost, accessibility, and technical barriers. Here, we introduce and evaluate the application of mobile neural radiance field (NeRF) methods for ecological research. Leveraging consumer-grade smartphones and open-source platforms (e.g. Luma AI), we demonstrate that mobile NeRFs can reconstruct detailed 3D structures of vegetation with accuracy comparable to TLS in open-canopy environments. We assess the strengths and limitations of NeRFs across habitat types, showing that while performance declines under occlusion (e.g. dense canopies), these methods excel at capturing understory complexity, making them particularly valuable for savannas, grasslands, and urban systems. We further explore the potential of radiance fields to integrate hyperspectral and robotic data streams, expanding their utility for dynamic ecosystem monitoring. By reducing hardware requirements and broadening participation, mobile NeRFs offer a promising avenue for democratising ecological data collection and advancing scalable environmental surveillance.
- 3 April 2025
- Published by Cambridge University Press (CUP) in Behavioral and Brain Sciences
- Vol. 48, e59
A nation by any other name: A failure to focus on function
Moffett's interdisciplinary definition of society seeks to distinguish itself from the prevalent, political understanding of the term. Through engagement with international relations literature, we outline how Moffett's proposed “society” results in a recapitulation of the definition of a nation-state. We suggest that this tension could be addressed by adopting a functional, rather than identity-based, approach.
- 3 April 2025
- Published by Cambridge University Press (CUP) in Behavioral and Brain Sciences
- Vol. 48, e59
A nation by any other name: A failure to focus on function
Moffett's interdisciplinary definition of society seeks to distinguish itself from the prevalent, political understanding of the term. Through engagement with international relations literature, we outline how Moffett's proposed “society” results in a recapitulation of the definition of a nation-state. We suggest that this tension could be addressed by adopting a functional, rather than identity-based, approach.
- 31 March 2025
- Published by The Company of Biologists in Journal of Experimental Biology
- Vol. 228 (7)
Ferrets move their heads to reduce image blur during the chase
- 31 March 2025
- Published by The Company of Biologists in Journal of Experimental Biology
- Vol. 228 (7)
Ferrets move their heads to reduce image blur during the chase