Jiale Kong

Jiale Kong

(he/him)

Incoming Ph.D. Student

SIA, UCAS

Biography

Hi! I’m Jiale Kong(孔嘉乐), a fourth-year undergraduate student at Central China Normal University (CCNU), graduating soon. I have been admitted through the recommendation (exam-exempt) program to the Ph.D. program at the Shenyang Institute of Automation (SIA), University of Chinese Academy of Sciences (UCAS), where I am currently working with Prof. Hao Liu and Prof. Yongming Yang.

My research focuses on embodied intelligence for flexible robots, including soft and continuum robots. I am interested in modeling and control of continuum robots, learning-based control with sim-to-real transfer, shape, force and tactile sensing for compliant bodies, vision-based perception and navigation for flexible endoscopic and surgical robots, and vision-language-action models for compliant manipulation.

I am actively looking for research collaborations on flexible robots, embodied intelligence, and their applications in surgical and medical robotics. If my background fits your project, please feel free to reach out by email. I am happy to share my CV and chat.

GitHub Activity

Last 53 weeks

Interests

  • Large Language Models
  • Computer Vision
  • Reinforcement Learning
  • AI Ethics

Education

  • Ph.D. in Mechatronic Information Engineering, 2027–2032 (expected)

    Shenyang Institute of Automation, University of Chinese Academy of Sciences

  • B.E. in Computer Science and Technology (Huawei Intelligent Base Class), 2023–2027 (expected)

    Central China Normal University

Prizes & Talks

  • National Scholarship

    Ministry of Education of China

    2026
  • National Scholarship

    Ministry of Education of China

    2025
Recent Publications
CPGR-Mem: Cross-Platform Graph-based Routing Memory for Multimodal Intelligent Agents
Jiale Kong, Yuqi Zhao† († corresponding author)
IEEE International Joint Conference on Neural Networks (IJCNN) 2026ConferenceCCF-C

News

Milestones

  • Admitted through the recommendation (exam-exempt) program to the Ph.D. program at SIA, UCAS.
  • Our work CPGR-Mem was accepted by IEEE IJCNN.

Experience

  1. Student Developer

    Open Source Summer (OSPP) 2026 · llama.cpp RVV Operator Optimization

    2026 – Present · Remote

    • Optimizing the RISC-V backend of llama.cpp/GGML with RVV (RISC-V Vector Extension) kernels, developed and validated on a VLEN=1024 K3 platform.
    • Targets - vectorized SSM scan and Gated-Delta-Net operators, VLEN=1024 quantized dot-product kernels, tinyBLAS LMUL tuning, and perf/llama-bench profiling, aiming for at least 1.10× geometric-mean speedup and upstream Pull Requests.
  2. LFX Mentorship Mentee

    Volcano (CNCF)

    Jun 2026 – Aug 2026 · Remote

    • Worked on namespace-scoped Queue support for the Volcano batch scheduler, introducing a NamespaceQueue CRD derived from the cluster Queue with the same semantics, so tenants can manage queues without cluster-admin rights.
    • Made PodGroup and Job able to reference a NamespaceQueue while staying compatible with the scheduling.volcano.sh/queue-name annotation, with Ginkgo E2E tests and documentation. Stack - Go, Kubernetes (CRD, controller, RBAC).