AVE-Compass: Towards Holistic Evaluation for Audio-Video Editing Abilities
Lin, Y. et al. (2026). "AVE-Compass: Towards Holistic Evaluation for Audio-Video Editing Abilities." NeurIPS 2026.
CV
Education
- B.S. in Computer Science and Technology, Xi’an Jiaotong University, 2023 – Present
Core Courses: Linear Algebra (99), Advanced Mathematics (95), Data Structures (90), OOP (98), Digital Logic Circuits (90), Probability Theory (89), ICS (96), Computer Graphics (95), Computer Networks (89), Operating Systems (85)
Ranking: Freshman year: 17/193, Sophomore year: 22/187
Scholarships: Second-Class Scholarship (Freshman & Sophomore Year), Excellent Student Award
Research Experience
AVE-Compass: Towards Holistic Evaluation for Audio-Video Editing Abilities (NeurIPS 2026 under review)
- Benchmark Construction: Built a comprehensive audio-video dataset covering 33 fine-grained categories and 190 instructions, addressing the gap in cross-modal physical alignment evaluation.
- Evaluation Design: Implemented a decoupled evaluation pipeline combining “objective physical metrics + modality-separated MLLM checklist” to precisely quantify Instruction Following (IF), Fidelity Preservation (FP), and Realism (R).
- Experimental Analysis: Deployed full evaluation on SOTA models including Wan2.7 and LTX2, revealing core weaknesses such as “cross-modal intention failure” in multimodal LLMs through human-machine consistency and robustness analysis.
RoboCup Robot Electronic Control System Development
- Developed the electronic control system for a multi-degree-of-freedom robotic arm, implementing low-level motor drivers and sensor closed-loop control.
- Applied PID control algorithms for gravity compensation of the robotic arm, effectively eliminating steady-state errors and motion jitter caused by link self-weight, significantly improving end-effector positioning accuracy.
YOLO Deployment and Acceleration on Ascend NPU
- Completed lightweight deployment of YOLO models on Ascend NPU, enabling end-to-end edge inference.
- Improved object detection FPS through model conversion and edge-side acceleration optimization, significantly reducing transmission latency.
Skills
- Programming Languages: C/C++, Python
- Frameworks & Tools: PyTorch, YOLO, Git, Linux/Ubuntu
- Languages: English (CET-4: 619, CET-6: 590)
Honors & Awards
- Meritorious Winner (M奖) — MCM/ICM Mathematical Contest in Modeling
- Third Prize — RoboCup China Competition
- Second-Class Scholarship — Xi’an Jiaotong University
- Excellent Student Award — Xi’an Jiaotong University