10 papers
GAPL: Grounded Action-effect Policy Learning for LLM-Based Trajectory Planning
Zhihong Cui, Hengyu Liu, Zhangkai Wu +5
Trajectory planning for autonomous driving requires both high-level reasoning and precise low-level control. Large Language Models (LLMs) offer semantic-rich planning capabilities,…
Reliable AI Needs to Externalize Implicit Knowledge: A Human-AI Collaboration Perspective
Hengyu Liu, Tianyi Li, Zhihong Cui +5
This position paper argues that reliable AI requires infrastructure for human validation of implicit knowledge. AI learns from both explicit knowledge (papers, documentation, struc…
CLEAR: A Knowledge-Centric Vessel Trajectory Analysis Platform
Hengyu Liu, Tianyi Li, Haoyu Wang +5
Vessel trajectory data from the Automatic Identification System (AIS) is used widely in maritime analytics. Yet, analysis is difficult for non-expert users due to the incompletenes…
VISTA: Knowledge-Driven Vessel Trajectory Imputation with Repair Provenance
Hengyu Liu, Tianyi Li, Haoyu Wang +4
Repairing incomplete trajectory data is essential for downstream spatio-temporal applications. Yet, existing repair methods focus solely on reconstruction without documenting the r…
Towards Practical Zero-Knowledge Proof for PSPACE
Ashwin Karthikeyan, Hengyu Liu, Kuldeep S. Meel +1
Efficient zero-knowledge proofs (ZKPs) have been restricted to NP statements so far, whereas they exist for all statements in PSPACE. This work presents the first practical zero-kn…
MH-GIN: Multi-scale Heterogeneous Graph-based Imputation Network for AIS Data (Extended Version)
Hengyu Liu, Tianyi Li, Yuqiang He +3
Location-tracking data from the Automatic Identification System, much of which is publicly available, plays a key role in a range of maritime safety and monitoring applications. Ho…
ACTIVE: Continuous Similarity Search for Vessel Trajectories
Tiantian Liu, Hengyu Liu, Tianyi Li +2
Publicly available vessel trajectory data is emitted continuously from the global AIS system. Continuous trajectory similarity search on this data has applications in, e.g., mariti…
An Inclusive Theoretical Framework of Robust Supervised Contrastive Loss against Label Noise
Jingyi Cui, Yi-Ge Zhang, Hengyu Liu +1
Learning from noisy labels is a critical challenge in machine learning, with vast implications for numerous real-world scenarios. While supervised contrastive learning has recently…
Deploying Multi-task Online Server with Large Language Model
Yincen Qu, Chao Ma, Xiangying Dai +3
In the industry, numerous tasks are deployed online. Traditional approaches often tackle each task separately by its own network, which leads to excessive costs for developing and…
Contrastive Language Video Time Pre-training
Hengyue Liu, Kyle Min, Hector A. Valdez +1
We introduce LAVITI, a novel approach to learning language, video, and temporal representations in long-form videos via contrastive learning. Different from pre-training on video-t…