WetRobo: A Reproducible Robot Kit for Coding Agents in Biological Laboratories
为解决生物实验室自动化问题，研究者开发了WetRobo机器人套件，通过自然语言指令和编程代理适应不同实验室环境，无需特定训练数据或神经网络。
0 citationsRead paper
Selected work
Representative Papers
AutoSchema: Live Schema Grounding for Agentic Text-to-Sparql over Heterogeneous Knowledge Graphs
This study addresses the challenges of schema maintenance and semantic alignment in SPARQL querying over heterogeneous knowledge graphs by proposing a training-free, real-time schema anchoring framework. Through dynamic endpoint probing, entity mapping, and path exploration, this approach enables agents to achieve zero-shot adaptive query construction on unseen graphs without reliance on predefined schemas. Experimental results demonstrate that the framework significantly improves factual accuracy and reduces tool invocation frequency in biomedical KGQA tasks. Furthermore, it successfully transfers to undocumented knowledge graphs, effectively resolving cross-resource semantic alignment and cold-start issues.
0 citationsRead paper
Posterior Inference of Hamiltonian Parameters from RIXS Spectroscopy
This study addresses the challenge of posterior inference for Hamiltonian parameters in Resonant Inelastic X-ray Scattering (RIXS) spectra by proposing the first simulation-based inference framework. Integrating a physics-aware Vision Transformer, truncated marginal neural ratio estimation, and conditional flow matching, this approach enables efficient and accurate full posterior inference for nickel compounds under few-shot conditions. The method not only uncovers critical parameter correlations and yields predicted spectra highly consistent with experimental data but also achieves reliable uncertainty quantification. Consequently, this work establishes a novel paradigm for the spectroscopic analysis of complex quantum materials, overcoming longstanding limitations in extracting precise physical parameters from RIXS measurements through advanced probabilistic modeling and domain-informed deep learning architectures.
0 citationsRead paper
Beyond Structure: Revolutionising Materials Discovery via AI-Driven Synthesis Protocol-Property Relationships
This work addresses the "synthesis gap" in AI-driven materials discovery—the disconnect arising from neglecting synthetic feasibility—by introducing a "synthesis-first" paradigm. It uniquely treats machine-readable synthesis protocols as primary design variables, establishing a causal framework that maps synthesis protocols (P) → structure (X) → performance (y). By integrating generative and inverse design models, closed-loop optimization algorithms, and self-driving laboratory technologies, the approach enables concurrent optimization of synthesis pathways and material properties. This strategy not only bridges the longstanding divide between computational design and experimental realization but also provides a systematic, data-driven methodology for reproducible and sustainable materials discovery.
0 citationsRead paper
Recent publications
Latest Papers
WetRobo: A Reproducible Robot Kit for Coding Agents in Biological Laboratories
为解决生物实验室自动化问题，研究者开发了WetRobo机器人套件，通过自然语言指令和编程代理适应不同实验室环境，无需特定训练数据或神经网络。
0 citationsRead paper
AutoSchema: Live Schema Grounding for Agentic Text-to-Sparql over Heterogeneous Knowledge Graphs
This study addresses the challenges of schema maintenance and semantic alignment in SPARQL querying over heterogeneous knowledge graphs by proposing a training-free, real-time schema anchoring framework. Through dynamic endpoint probing, entity mapping, and path exploration, this approach enables agents to achieve zero-shot adaptive query construction on unseen graphs without reliance on predefined schemas. Experimental results demonstrate that the framework significantly improves factual accuracy and reduces tool invocation frequency in biomedical KGQA tasks. Furthermore, it successfully transfers to undocumented knowledge graphs, effectively resolving cross-resource semantic alignment and cold-start issues.
0 citationsRead paper
Posterior Inference of Hamiltonian Parameters from RIXS Spectroscopy
This study addresses the challenge of posterior inference for Hamiltonian parameters in Resonant Inelastic X-ray Scattering (RIXS) spectra by proposing the first simulation-based inference framework. Integrating a physics-aware Vision Transformer, truncated marginal neural ratio estimation, and conditional flow matching, this approach enables efficient and accurate full posterior inference for nickel compounds under few-shot conditions. The method not only uncovers critical parameter correlations and yields predicted spectra highly consistent with experimental data but also achieves reliable uncertainty quantification. Consequently, this work establishes a novel paradigm for the spectroscopic analysis of complex quantum materials, overcoming longstanding limitations in extracting precise physical parameters from RIXS measurements through advanced probabilistic modeling and domain-informed deep learning architectures.
0 citationsRead paper
Beyond Structure: Revolutionising Materials Discovery via AI-Driven Synthesis Protocol-Property Relationships
This work addresses the "synthesis gap" in AI-driven materials discovery—the disconnect arising from neglecting synthetic feasibility—by introducing a "synthesis-first" paradigm. It uniquely treats machine-readable synthesis protocols as primary design variables, establishing a causal framework that maps synthesis protocols (P) → structure (X) → performance (y). By integrating generative and inverse design models, closed-loop optimization algorithms, and self-driving laboratory technologies, the approach enables concurrent optimization of synthesis pathways and material properties. This strategy not only bridges the longstanding divide between computational design and experimental realization but also provides a systematic, data-driven methodology for reproducible and sustainable materials discovery.
0 citationsRead paper