Back to Projects
Research PaperCompleted
TARS: SOICT 2025
Co-first authored a research paper accepted at SOICT 2025 presenting TARS (Temporal Alignment Retrieval System), a training-free order-aware framework for multi-segment video event retrieval using monotonic dynamic programming alignment over vision-language encoders.
Key Details
Query decomposed into ordered sub-event sequences embedded by complementary vision-language encoders.
Monotonic DP alignment finds the best ordered path on the frame-subevent similarity matrix with O(nm) time and O(m) memory.
Training-free design requires no additional dataset-specific training beyond base encoders, ensuring robustness under domain shift.
Integrates cleanly with standard two-stage candidate retrieval and re-ranking pipelines.
Demonstrated 93.15% accuracy on the Ho Chi Minh City AI Challenge 2025 benchmark.
Highlights
- •Accepted at SOICT 2025
- •Training-free temporal reasoning at inference time
- •Monotonic DP with O(nm) time, O(m) memory
- •93.15% accuracy on competition benchmark
Technologies
PythonPyTorchCLIPFAISSNumPyGoogle Gemini