arXiv subjects

Tianyi Li

Publications and source records attributed to Tianyi Li.

At least 55 records · Page 3Linked to original sources

arXiv2026

VISTA: Knowledge-Driven Vessel Trajectory Imputation with Repair Provenance

Repairing incomplete trajectory data is essential for downstream spatio-temporal applications. Yet, existing repair methods focus solely on reconstruction without documenting the reasoning behind repair decisions, undermining trust in safety-critical applications where repaired trajectories affect operational decisions, such as in maritime anomaly detection and route planning. We introduce repair provenance - structured, queryable metadata that documents the full reasoning chain behind each repair - which transforms imputation from pure data recovery into a task that supports downstream decision-making. We propose VISTA (knowledge-driven interpretable vessel trajectory imputation), a framework that reliably equips repaired trajectories with repair provenance by grounding LLM reasoning in data-verified knowledge. Specifically, we formalize Structured Data-derived Knowledge (SDK), a knowledge model whose data-verifiable components can be validated against real data and used to anchor and constrain LLM-generated explanations. We organize SDK in a Structured Data-derived Knowledge Graph (SD-KG) and establish a data-knowledge-data loop for extraction, validation, and incremental maintenance over large-scale AIS data. A workflow management layer with parallel scheduling, fault tolerance, and redundancy control ensures consistent and efficient end-to-end processing. Experiments on two large-scale AIS datasets show that VISTA achieves state-of-the-art accuracy, improving over baselines by 5-91% and reducing inference time by 51-93%, while producing repair provenance, whose interpretability is further validated through a case study and an interactive demo system.

cs.DB↗

arXiv2026

Sink-Aware Pruning for Diffusion Language Models

Diffusion Language Models (DLMs) incur high inference cost due to iterative denoising, motivating efficient pruning. Existing pruning heuristics largely inherited from autoregressive (AR) LLMs, typically preserve attention sink tokens because AR sinks serve as stable global anchors. We show that this assumption does not hold for DLMs: the attention-sink position exhibits substantially higher variance over the full generation trajectory (measured by how the dominant sink locations shift across timesteps), indicating that sinks are often transient and less structurally essential than in AR models. Based on this observation, we propose ${\bf \texttt{Sink-Aware Pruning}}$, which automatically identifies and prunes unstable sinks in DLMs (prior studies usually keep sinks for AR LLMs). Without retraining, our method achieves a better quality-efficiency trade-off and outperforms strong prior pruning baselines under matched compute. Our code is available at https://github.com/VILA-Lab/Sink-Aware-Pruning.

cs.CL↗

arXiv2026

MadEvolve: Evolutionary Optimization of Cosmological Algorithms with Large Language Models

We develop a general framework to discover scientific algorithms and apply it to three problems in computational cosmology. Our code, MadEvolve, is similar to Google's AlphaEvolve, but places a stronger emphasis on free parameters and their optimization. Our code starts with a baseline human algorithm implementation, and then optimizes its performance metrics by making iterative changes to its code. As a further convenient feature, MadEvolve automatically generates a report that compares the input algorithm with the evolved algorithm, describes the algorithmic innovations and lists the free parameters and their function. Our code supports both auto-differentiable, gradient-based parameter optimization and gradient-free optimization methods. We apply MadEvolve to the reconstruction of cosmological initial conditions, 21cm foreground contamination reconstruction and effective baryonic physics in N-body simulations. In all cases, we find substantial improvements over the base algorithm. We make MadEvolve and our three tasks publicly available at madevolve.org.

astro-ph.CO↗

arXiv2026

TCRL: Temporal-Coupled Adversarial Training for Robust Constrained Reinforcement Learning in Worst-Case Scenarios

Constrained Reinforcement Learning (CRL) aims to optimize decision-making policies under constraint conditions, making it highly applicable to safety-critical domains such as autonomous driving, robotics, and power grid management. However, existing robust CRL approaches predominantly focus on single-step perturbations and temporally independent adversarial models, lacking explicit modeling of robustness against temporally coupled perturbations. To tackle these challenges, we propose TCRL, a novel temporal-coupled adversarial training framework for robust constrained reinforcement learning (TCRL) in worst-case scenarios. First, TCRL introduces a worst-case-perceived cost constraint function that estimates safety costs under temporally coupled perturbations without the need to explicitly model adversarial attackers. Second, TCRL establishes a dual-constraint defense mechanism on the reward to counter temporally coupled adversaries while maintaining reward unpredictability. Experimental results demonstrate that TCRL consistently outperforms existing methods in terms of robustness against temporally coupled perturbation attacks across a variety of CRL tasks.

cs.LG↗

arXiv2025

Efficient Quantum Simulation of Non-Adiabatic Molecular Dynamics with Precise Electronic Structure

In the study of non-adiabatic chemical processes such as photocatalysis and photosynthesis, non-adiabatic molecular dynamics (NAMD) is an indispensable theoretical tool, which requires precise potential energy surfaces (PESs) of ground and excited states. Quantum computing offers promising potential for calculating PESs that are intractable for classical computers. However, its realistic application poses significant challenges to the development of quantum algorithms that are sufficiently general to enable efficient and precise PES calculations across chemical systems with diverse properties, as well as to seamlessly adapt existing NAMD theories to quantum computing. In this work, we introduce a quantum-adapted extension to the Landau-Zener-Surface-Hopping (LZSH) NAMD. This extension incorporates curvature-driven hopping corrections that protect the population evolution while maintaining the efficiency gained from avoiding the computation of non-adiabatic couplings (NACs), as well as preserving the trajectory independence that enables parallelization. Furthermore, to ensure the high-precision PESs required for surface hopping dynamics, we develop a sub-microhartree-accurate PES calculation protocol. This protocol supports active space selection, enables parallel acceleration either on quantum or classical clusters, and demonstrates adaptability to diverse chemical systems - including the charged H3+ ion and the C2H4 molecule, a prototypical multi-reference benchmark. This work paves the way for practical application of quantum computing in NAMD, showcasing the potential of parallel simulation on quantum-classical heterogeneous clusters for ab-initio computational chemistry.

physics.chem-ph↗

arXiv2025

A Superconducting Qubit-Resonator Quantum Processor with Effective All-to-All Connectivity

In this work we introduce a superconducting quantum processor architecture that uses a transmission-line resonator to implement effective all-to-all connectivity between six transmon qubits. This architecture can be used as a test-bed for algorithms that benefit from high connectivity. We show that the central resonator can be used as a computational element, which offers the flexibility to encode a qubit for quantum computation or to utilize its bosonic modes which further enables quantum simulation of bosonic systems. To operate the quantum processing unit (QPU), we develop and benchmark the qubit-resonator conditional Z gate and the qubit-resonator MOVE operation. The latter allows for transferring a quantum state between one of the peripheral qubits and the computational resonator. We benchmark the QPU performance and achieve a genuinely multi-qubit entangled Greenberger-Horne-Zeilinger (GHZ) state over all six qubits with a readout-error mitigated fidelity of $0.86$.

quant-ph↗

arXiv2025

Beyond Relational: Semantic-Aware Multi-Modal Analytics with LLM-Native Query Optimization

Multi-modal analytical processing has the potential to transform applications in e-commerce, healthcare, entertainment, and beyond. However, real-world adoption remains elusive due to the limited ability of traditional relational query operators to capture query semantics. The emergence of foundation models, particularly the large language models (LLMs), opens up new opportunities to develop flexible, semantic-aware data analytics systems that transcend the relational paradigm. We present Nirvana, a multi-modal data analytics framework that incorporates programmable semantic operators while leveraging both logical and physical query optimization strategies, tailored for LLM-driven semantic query processing. Nirvana addresses two key challenges. First, it features an agentic logical optimizer that uses natural language-specified transformation rules and random-walk-based search to explore vast spaces of semantically equivalent query plans -- far beyond the capabilities of conventional optimizers. Second, it introduces a cost-aware physical optimizer that selects the most effective LLM backend for each operator using a novel improvement-score metric. To further enhance efficiency, Nirvana incorporates computation reuse and evaluation pushdown techniques guided by model capability hypotheses. Experimental evaluations on three real-world benchmarks demonstrate that Nirvana is able to reduce end-to-end runtime by 10%--85% and reduces system processing costs by 76% on average, outperforming state-of-the-art systems at both efficiency and scalability.

cs.DB↗

arXiv2025

ACTIVE: Continuous Similarity Search for Vessel Trajectories

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., maritime navigation and safety. Existing proposals typically assume an offline setting and focus on finding similarities between complete trajectories. Such proposals are less effective when applied to online scenarios, where similarity comparisons must be performed continuously as new trajectory data arrives and trajectories evolve. We therefore propose a real-time continuous trajectory similarity search method for vessels (ACTIVE). We introduce a novel similarity measure, object-trajectory real-time distance, that emphasizes the anticipated future movement trends of vessels, enabling more predictive and forward-looking comparisons. Next, we propose an efficient continuous similar trajectory search (CSTS) algorithm together with a segment-based vessel trajectory index and a variety of search space pruning strategies that reduce unnecessary computations during the continuous similarity search, thereby further improving efficiency. Extensive experiments on two large real-world AIS datasets offer evidence that ACTIVE is capable of outperforming state-of-the-art methods considerably. ACTIVE significantly reduces index construction costs and index size while achieving a 70% reduction in terms of query time and a 60% increase in terms of hit rate.

cs.DB↗

arXiv2025

MH-GIN: Multi-scale Heterogeneous Graph-based Imputation Network for AIS Data (Extended Version)

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. However, the data suffers from missing values that hamper downstream applications. Imputing the missing values is challenging because the values of different heterogeneous attributes are updated at diverse rates, resulting in the occurrence of multi-scale dependencies among attributes. Existing imputation methods that assume similar update rates across attributes are unable to capture and exploit such dependencies, limiting their imputation accuracy. We propose MH-GIN, a Multi-scale Heterogeneous Graph-based Imputation Network that aims improve imputation accuracy by capturing multi-scale dependencies. Specifically, MH-GIN first extracts multi-scale temporal features for each attribute while preserving their intrinsic heterogeneous characteristics. Then, it constructs a multi-scale heterogeneous graph to explicitly model dependencies between heterogeneous attributes to enable more accurate imputation of missing values through graph propagation. Experimental results on two real-world datasets find that MH-GIN is capable of an average 57% reduction in imputation errors compared to state-of-the-art methods, while maintaining computational efficiency. The source code and implementation details of MH-GIN are publicly available https://github.com/hyLiu1994/MH-GIN.

cs.LG↗

arXiv2025

Estimating the Error of Large Language Models at Pairwise Text Comparison

We measure LLMs' output error at pairwise text comparison, noting the probability of error in their preferences. Our method does not rely on the ground truth and supports two scenarios: (i) uniform error rate regardless of the order of comparison, estimated with two comparisons for each text pair with either text placed first; (ii) binary positional bias assuming distinct error rates for the two orders of comparison, estimated with repeated comparisons between the texts. The Copeland counting constructs a ranking over the compared texts from pairwise preferences; the ranking reveals the poor scalability of LLM-based pairwise comparison and helps yield the estimates for LLMs' error rates. We apply the method to six LLMs (ChatGPT, Claude, DeepSeek, Gemini, Grok, Qwen) with five types of text input and obtain consistent estimates of LLMs' error. In general, the measured two positional bias terms are similar, close to the uniform error. Considering both the error rates and the robustness to the variation of prompts, Claude obtained the most desirable performance in this experiment. Our model outperforms the biased Bradley-Terry model and the commutativity score in indicating LLMs' error at this task.

cs.CL↗

arXiv2025

Moon: A Modality Conversion-based Efficient Multivariate Time Series Anomaly Detection

Multivariate time series (MTS) anomaly detection identifies abnormal patterns where each timestamp contains multiple variables. Existing MTS anomaly detection methods fall into three categories: reconstruction-based, prediction-based, and classifier-based methods. However, these methods face two key challenges: (1) Unsupervised learning methods, such as reconstruction-based and prediction-based methods, rely on error thresholds, which can lead to inaccuracies; (2) Semi-supervised methods mainly model normal data and often underuse anomaly labels, limiting detection of subtle anomalies;(3) Supervised learning methods, such as classifier-based approaches, often fail to capture local relationships, incur high computational costs, and are constrained by the scarcity of labeled data. To address these limitations, we propose Moon, a supervised modality conversion-based multivariate time series anomaly detection framework. Moon enhances the efficiency and accuracy of anomaly detection while providing detailed anomaly analysis reports. First, Moon introduces a novel multivariate Markov Transition Field (MV-MTF) technique to convert numeric time series data into image representations, capturing relationships across variables and timestamps. Since numeric data retains unique patterns that cannot be fully captured by image conversion alone, Moon employs a Multimodal-CNN to integrate numeric and image data through a feature fusion model with parameter sharing, enhancing training efficiency. Finally, a SHAP-based anomaly explainer identifies key variables contributing to anomalies, improving interpretability. Extensive experiments on six real-world MTS datasets demonstrate that Moon outperforms six state-of-the-art methods by up to 93% in efficiency, 4% in accuracy and, 10.8% in interpretation performance.

cs.LG↗

arXiv2025

A phase-aware AI car-following model for electric vehicles with adaptive cruise control: Development and validation using real-world data

Internal combustion engine (ICE) vehicles and electric vehicles (EVs) exhibit distinct vehicle dynamics. EVs provide rapid acceleration, with electric motors producing peak power across a wider speed range, and achieve swift deceleration through regenerative braking. While existing microscopic models effectively capture the driving behavior of ICE vehicles, a modeling framework that accurately describes the unique car-following dynamics of EVs is lacking. Developing such a model is essential given the increasing presence of EVs in traffic, yet creating an easy-to-use and accurate analytical model remains challenging. To address these gaps, this study develops and validates a Phase-Aware AI (PAAI) car-following model specifically for EVs. The proposed model enhances traditional physics-based frameworks with an AI component that recognizes and adapts to different driving phases, such as rapid acceleration and regenerative braking. Using real-world trajectory data from vehicles equipped with adaptive cruise control (ACC), we conduct comprehensive simulations to validate the model's performance. The numerical results demonstrate that the PAAI model significantly improves prediction accuracy over traditional car-following models, providing an effective tool for accurately representing EV behavior in traffic simulations.

cs.RO↗

arXiv2025

Conceptualizing and Modeling Communication-Based Cyberattacks on Automated Vehicles

Adaptive Cruise Control (ACC) is rapidly proliferating across electric vehicles (EVs) and internal combustion engine (ICE) vehicles, enhancing traffic flow while simultaneously expanding the attack surface for communication-based cyberattacks. Because the two powertrains translate control inputs into motion differently, their cyber-resilience remains unquantified. Therefore, we formalize six novel message-level attack vectors and implement them in a ring-road simulation that systematically varies the ACC market penetration rates (MPRs) and the spatial pattern of compromised vehicles. A three-tier risk taxonomy converts disturbance metrics into actionable defense priorities for practitioners. Across all simulation scenarios, EV platoons exhibit lower velocity standard deviation, reduced spacing oscillations, and faster post-attack recovery compared to ICE counterparts, revealing an inherent stability advantage. These findings clarify how controller-to-powertrain coupling influences vulnerability and offer quantitative guidance for the detection and mitigation of attacks in mixed automated traffic.

eess.SY↗

arXiv2025

Large Language Models for Pedestrian Safety: An Application to Predicting Driver Yielding Behavior at Unsignalized Intersections

Pedestrian safety is a critical component of urban mobility and is strongly influenced by the interactions between pedestrian decision-making and driver yielding behavior at crosswalks. Modeling driver--pedestrian interactions at intersections requires accurately capturing the complexity of these behaviors. Traditional machine learning models often struggle to capture the nuanced and context-dependent reasoning required for these multifactorial interactions, due to their reliance on fixed feature representations and limited interpretability. In contrast, large language models (LLMs) are suited for extracting patterns from heterogeneous traffic data, enabling accurate modeling of driver-pedestrian interactions. Therefore, this paper leverages multimodal LLMs through a novel prompt design that incorporates domain-specific knowledge, structured reasoning, and few-shot prompting, enabling interpretable and context-aware inference of driver yielding behavior, as an example application of modeling pedestrian--driver interaction. We benchmarked state-of-the-art LLMs against traditional classifiers, finding that GPT-4o consistently achieves the highest accuracy and recall, while Deepseek-V3 excels in precision. These findings highlight the critical trade-offs between model performance and computational efficiency, offering practical guidance for deploying LLMs in real-world pedestrian safety systems.

cs.CL↗

arXiv2025

Advancing Knowledge Tracing by Exploring Follow-up Performance Trends

Intelligent Tutoring Systems (ITS), such as Massive Open Online Courses, offer new opportunities for human learning. At the core of such systems, knowledge tracing (KT) predicts students' future performance by analyzing their historical learning activities, enabling an accurate evaluation of students' knowledge states over time. We show that existing KT methods often encounter correlation conflicts when analyzing the relationships between historical learning sequences and future performance. To address such conflicts, we propose to extract so-called Follow-up Performance Trends (FPTs) from historical ITS data and to incorporate them into KT. We propose a method called Forward-Looking Knowledge Tracing (FINER) that combines historical learning sequences with FPTs to enhance student performance prediction accuracy. FINER constructs learning patterns that facilitate the retrieval of FPTs from historical ITS data in linear time; FINER includes a novel similarity-aware attention mechanism that aggregates FPTs based on both frequency and contextual similarity; and FINER offers means of combining FPTs and historical learning sequences to enable more accurate prediction of student future performance. Experiments on six real-world datasets show that FINER can outperform ten state-of-the-art KT methods, increasing accuracy by 8.74% to 84.85%.

cs.CY↗

arXiv2025

A Multi-Task Evaluation of LLMs' Processing of Academic Text Input

How much large language models (LLMs) can aid scientific discovery, notably in assisting academic peer review, is in heated debate. Between a literature digest and a human-comparable research assistant lies their practical application potential. We organize individual tasks that computer science studies employ in separate terms into a guided and robust workflow to evaluate LLMs' processing of academic text input. We employ four tasks in the assessment: content reproduction/comparison/scoring/reflection, each demanding a specific role of the LLM (oracle/judgmental arbiter/knowledgeable arbiter/collaborator) in assisting scholarly works, and altogether testing LLMs with questions that increasingly require intellectual capabilities towards a solid understanding of scientific texts to yield desirable solutions. We exemplify a rigorous performance evaluation with detailed instructions on the prompts. Adopting first-rate Information Systems articles at three top journals as the input texts and an abundant set of text metrics, we record a compromised performance of the leading LLM - Google's Gemini: its summary and paraphrase of academic text is acceptably reliable; using it to rank texts through pairwise text comparison is faintly scalable; asking it to grade academic texts is prone to poor discrimination; its qualitative reflection on the text is self-consistent yet hardly insightful to inspire meaningful research. This evidence against an endorsement of LLMs' text-processing capabilities is consistent across metric-based internal (linguistic assessment), external (comparing to the ground truth), and human evaluation, and is robust to the variations of the prompt. Overall, we do not recommend an unchecked use of LLMs in constructing peer reviews.

cs.CL↗

arXiv2025

User-Centered Design with AI in the Loop: A Case Study of Rapid User Interface Prototyping with "Vibe Coding"

We present a case study of using generative user interfaces, or ``vibe coding,'' a method leveraging large language models (LLMs) for generating code via natural language prompts, to support rapid prototyping in user-centered design (UCD). Extending traditional UCD practices, we propose an AI-in-the-loop ideate-prototyping process. We share insights from an empirical experience integrating this process to develop an interactive data analytics interface for highway traffic engineers to effectively retrieve and analyze historical traffic data. With generative UIs, the team was able to elicit rich user feedback and test multiple alternative design ideas from user evaluation interviews and real-time collaborative sessions with domain experts. We discuss the advantages and pitfalls of vibe coding for bridging the gaps between design expertise and domain-specific expertise.

cs.HC↗

arXiv2025

Deterministic diffusion models for Lagrangian turbulence: robustness and encoding of extreme events

Modeling Lagrangian turbulence remains a fundamental challenge due to its multiscale, intermittent, and non-Gaussian nature. Recent advances in data-driven diffusion models have enabled the generation of realistic Lagrangian velocity trajectories that accurately reproduce statistical properties across scales and capture rare extreme events. This study investigates three key aspects of diffusion-based modeling for Lagrangian turbulence. First, we assess architectural robustness by comparing a U-Net backbone with a transformer-based alternative, finding strong consistency in generated trajectories, with only minor discrepancies at small scales. Second, leveraging a deterministic variant of diffusion model formulation, namely the deterministic denoising diffusion implicit model (DDIM), we identify structured features in the initial latent noise that align consistently with extreme acceleration events. Third, we explore accelerated generation by reducing the number of diffusion steps, and find that DDIM enables substantial speedups with minimal loss of statistical fidelity. These findings highlight the robustness of diffusion models and their potential for interpretable, scalable modeling of complex turbulent systems.

physics.flu-dyn↗