arXiv subjects

Fuyuan Ma

Publications and source records attributed to Fuyuan Ma.

3 recordsLinked to original sources

arXiv2026

A Common Language for Dynamic Graphs: Interaction Tokenization Enables Foundation Models

Dynamic graphs capture evolving interactions in Web systems, including online communication, user engagement, and recommendation. Graph foundation models aim to reuse predictive capabilities across networks, but extending this capability to dynamic graphs faces two challenges. First, node identities are graph-specific, and similar interaction patterns can appear at different temporal and activity scales, making histories difficult to represent consistently across networks. Second, the predictive roles of historical patterns vary across interaction dynamics, limiting transfer from a narrow range of training processes. In this paper, we show that a common interaction language, combined with learning across diverse interaction processes, offers a basis for dynamic graph foundation modeling. We propose CL-DGFM, a Common Language Dynamic Graph Foundation Model for temporal link prediction. Its Dynamic Interaction Tokenizer encodes interaction memory, historical connectivity, and ordered destination transitions into calibrated tabular inputs, making dynamic histories accessible to existing predictors without an additional trainable graph encoder. Its Dynamic Interaction Prior trains a pretrained tabular foundation predictor across diverse synthetic dynamics. Labeled historical context guides prediction, with a Contextual Value Adapter modulating attention values. The model transfers to real graphs without target-graph parameter updates. Experiments on TGB and TGB-Seq demonstrate competitive performance. Further comparisons show that the common language supports basic machine learning and neural models, and enables direct use of frozen tabular foundation models on dynamic graphs.

cs.LG↗

arXiv2026

JoyAI-Sim: A Simulation-Enabled Interconversion Toolchain for the Embodied Data Pyramid

Generalist robot policies require trustworthy evaluation and robot-usable training data, but both are difficult to scale with physical robots alone. Real-robot trials and demonstrations remain the most faithful source of deployment signals, yet they are slow, costly, and hard to reproduce. We present JoyAI-Sim, a simulation-enabled interconversion toolchain for human-robot aligned model evaluation and data generation, denoted as Robot $\rightleftharpoons$ Simulation $\rightleftharpoons$ Human. On the one hand, the Robot $\rightarrow$ Simulation $\rightarrow$ Human pathway supports human-robot aligned model evaluation by reconstructing real-robot tabletop organization tasks as calibrated digital twins for scalable evaluation, while using human embodied feedback to inspect and refine the naturalness of simulated motions. On the other hand, the Human $\rightarrow$ Simulation $\rightarrow$ Robot pathway supports human-robot aligned data generation: it lifts ego-centric human demonstrations into simulation, checks them under robot physical constraints, and converts them into robot-centered trajectories, annotations, and visual observations. Together, these pathways use the JoySim simulator as both a scalable evaluation layer and a physical consistency filter for robot data generation. We further package the core reconstruction, simulation, rendering, and realism-augmentation modules as cloud services on JD Cloud, turning the system into a reusable and scalable infrastructure for robot data generation and model evaluation.

cs.RO↗

arXiv2020

Explicit gain equations for hybrid graphene-quantum-dot photodetectors

Graphene is an attractive material for broadband photodetection but suffers from weak light absorption. Coating graphene with quantum dots can significantly enhance light absorption and create extraordinarily high photo gain. This high gain is often explained by the classical gain theory which is unfortunately an implicit function and may even be questionable. In this work, we managed to derive explicit gain equations for hybrid graphene-quantum-dot photodetectors. Due to the work function mismatch, lead sulfide (PbS) quantum dots coated on graphene will form a surface depletion region near the interface of quantum dots and graphene. Light illumination narrows down the surface depletion region, creating a photovoltage that gates the graphene. As a result, high photo gain in graphene is observed. The explicit gain equations are derived from the theoretical gate transfer characteristics of graphene and the correlation of the photovoltage with the light illumination intensity. The derived explicit gain equations fit well with the experimental data, from which physical parameters are extracted.

cond-mat.mes-hall↗