arXiv subjects

Tianyi Li

Publications and source records attributed to Tianyi Li.

At least 73 records · Page 4Linked to original sources

arXiv2025

RACER: Rational Artificial Intelligence Car-following-model Enhanced by Reality

This paper introduces RACER, the Rational Artificial Intelligence Car-following model Enhanced by Reality, a cutting-edge deep learning car-following model, that satisfies partial derivative constraints, designed to predict Adaptive Cruise Control (ACC) driving behavior while staying theoretically feasible. Unlike conventional models, RACER effectively integrates Rational Driving Constraints (RDCs), crucial tenets of actual driving, resulting in strikingly accurate and realistic predictions. Against established models like the Optimal Velocity Relative Velocity (OVRV), a car-following Neural Network (NN), and a car-following Physics-Informed Neural Network (PINN), RACER excels across key metrics, such as acceleration, velocity, and spacing. Notably, it displays a perfect adherence to the RDCs, registering zero violations, in stark contrast to other models. This study highlights the immense value of incorporating physical constraints within AI models, especially for augmenting safety measures in transportation. It also paves the way for future research to test these models against human driving data, with the potential to guide safer and more rational driving behavior. The versatility of the proposed model, including its potential to incorporate additional derivative constraints and broader architectural applications, enhances its appeal and broadens its impact within the scientific community.

cs.AI↗

arXiv2025

OneDB: A Distributed Multi-Metric Data Similarity Search System

Increasingly massive volumes of multi-modal data are being accumulated in many {real world} settings, including in health care and e-commerce. This development calls for effective general-purpose data management solutions for multi-modal data. Such a solution must facilitate user-friendly and accurate retrieval of any multi-modal data according to diverse application requirements. Further, such a solution must be capable of efficient and scalable retrieval. To address this need, we present OneDB, a distributed multi-metric data similarity retrieval system. This system exploits the fact that data of diverse modalities, such as text, images, and video, can be represented as metric data. The system thus affords each data modality its own metric space with its own distance function and then uses a multi-metric model to unify multi-modal data. The system features several innovations: (i) an extended Spart SQL query interface; (ii) lightweight means of learning appropriate weights of different modalities when retrieving multi-modal data to enable accurate retrieval; (iii) smart search-space pruning strategies that improve efficiency; (iv) two-layered indexing of data to ensure load-balancing during distributed processing; and (v) end-to-end system parameter autotuning. Experiments on three real-life datasets and two synthetic datasets offer evidence that the system is capable of state-of-the-art performance: (i) efficient and effective weight learning; (ii) retrieval accuracy improvements of 12.63\%--30.75\% over the state-of-the-art vector similarity search system at comparable efficiency; (iii) accelerated search by 2.5--5.75x over state-of-the-art single- or multi-metric solutions; (iv) demonstrated high scalability; and (v) parameter tuning that enables performance improvements of 15+%.

cs.DB↗

arXiv2025

GENPLUGIN: A Plug-and-Play Framework for Long-Tail Generative Recommendation with Exposure Bias Mitigation

Generative recommendation (GenRec) offers LLM integration, reduced embedding costs, and eliminates per-candidate scoring, attracting great attention. Despite its promising performance, this study reveals that it suffers from generation exposure bias and poor long-tail item generalization, two critical limitations overlooked by prior works on GenRec. To address these, we propose GENPLUGIN, a plug-and-play framework featuring a dual-encoder, shared-decoder architecture. During pre-training, it aligns language and ID views via contrastive learning, harmonizing item representations across two complementary views. Besides, GENPLUGIN uses a novel training strategy that probabilistically substitutes ground-truth item ID tokens with predictions from the language-semantics encoder, alleviating exposure bias. To improve long-tail generative recommendation, we propose a retrieval-based data augmentation mechanism. It fine-tunes the decoder of GENPLUGIN to endow GENPLUGIN with the ability to use relevant users w.r.t. contexts or collaborative information to augment the generation of item ID tokens in long-tail recommendation scenarios. We have plugged GENPLUGIN into several representative GenRec models and the extensive experiments demonstrate that GENPLUGIN can notably mitigate generation exposure bias during item ID generation while significantly improving the quality of long-tail item recommendation.

cs.IR↗

arXiv2025

Test-time Scaling Techniques in Theoretical Physics -- A Comparison of Methods on the TPBench Dataset

Large language models (LLMs) have shown strong capabilities in complex reasoning, and test-time scaling techniques can enhance their performance with comparably low cost. Many of these methods have been developed and evaluated on mathematical reasoning benchmarks such as AIME. This paper investigates whether the lessons learned from these benchmarks generalize to the domain of advanced theoretical physics. We evaluate a range of common test-time scaling methods on the TPBench physics dataset and compare their effectiveness with results on AIME. To better leverage the structure of physics problems, we develop a novel, symbolic weak-verifier framework to improve parallel scaling results. Our empirical results demonstrate that this method significantly outperforms existing test-time scaling approaches on TPBench. We also evaluate our method on AIME, confirming its effectiveness in solving advanced mathematical problems. Our findings highlight the power of step-wise symbolic verification for tackling complex scientific problems.

cs.LG↗

arXiv2025

A Multi-Modal Knowledge-Enhanced Framework for Vessel Trajectory Prediction

Accurate vessel trajectory prediction facilitates improved navigational safety, routing, and environmental protection. However, existing prediction methods are challenged by the irregular sampling time intervals of the vessel tracking data from the global AIS system and the complexity of vessel movement. These aspects render model learning and generalization difficult. To address these challenges and improve vessel trajectory prediction, we propose the multi-modal knowledge-enhanced framework (MAKER) for vessel trajectory prediction. To contend better with the irregular sampling time intervals, MAKER features a Large language model-guided Knowledge Transfer (LKT) module that leverages pre-trained language models to transfer trajectory-specific contextual knowledge effectively. To enhance the ability to learn complex trajectory patterns, MAKER incorporates a Knowledge-based Self-paced Learning (KSL) module. This module employs kinematic knowledge to progressively integrate complex patterns during training, allowing for adaptive learning and enhanced generalization. Experimental results on two vessel trajectory datasets show that MAKER can improve the prediction accuracy of state-of-the-art methods by 12.08%-17.86%.

cs.CV↗

arXiv2025

Neutralizing Bias in LLM Reasoning using Entailment Graphs

LLMs are often claimed to be capable of Natural Language Inference (NLI), which is widely regarded as a cornerstone of more complex forms of reasoning. However, recent works show that LLMs still suffer from hallucinations in NLI due to attestation bias, where LLMs overly rely on propositional memory to build shortcuts. To solve the issue, we design an unsupervised framework to construct counterfactual reasoning data and fine-tune LLMs to reduce attestation bias. To measure bias reduction, we build bias-adversarial variants of NLI datasets with randomly replaced predicates in premises while keeping hypotheses unchanged. Extensive evaluations show that our framework can significantly reduce hallucinations from attestation bias. Then, we further evaluate LLMs fine-tuned with our framework on original NLI datasets and their bias-neutralized versions, where original entities are replaced with randomly sampled ones. Extensive results show that our framework consistently improves inferential performance on both original and bias-neutralized NLI datasets.

cs.CL↗

arXiv2025

Differential absorption ozone Lidar with 4H-SiC single-photon detectors

Differential absorption Lidar (DIAL) in the ultraviolet (UV) region is an effective approach for monitoring tropospheric ozone. 4H-SiC single-photon detectors (SPDs) are emergent devices for UV single-photon detection. Here, we demonstrate a 4H-SiC SPD-based ozone DIAL. We design and fabricate the 4H-SiC single-photon avalanche diode with a beveled mesa structure and optimized layer thickness. An active quenching circuit with a quenching time of 1.03 ns is developed to significantly mitigate the afterpulsing effect while enhancing the maximum count rate. After characterization, the SPD exhibits excellent performance with a photon detection efficiency of 16.6% at 266 nm, a dark count rate of 138 kcps, a maximum count rate of 13 Mcps, and an afterpulse probability of 2.7% at room temperature. Then, we apply two 4H-SiC SPDs in an ozone DIAL. The measured ozone concentrations at altitudes of 1-3.5 km agree well with the results of a commercial ozone DIAL. Our work provides an alternative solution for general UV Lidar applications.

physics.optics↗

arXiv2025

Advancing vision-language models in front-end development via data synthesis

Modern front-end (FE) development, especially when leveraging the unique features of frameworks like React and Vue, presents distinctive challenges. These include managing modular architectures, ensuring synchronization between data and visual outputs for declarative rendering, and adapting reusable components to various scenarios. Such complexities make it particularly difficult for state-of-the-art large vision-language models (VLMs) to generate accurate and functional code directly from design images. To address these challenges, we propose a reflective agentic workflow that synthesizes high-quality image-text data to capture the diverse characteristics of FE development. This workflow automates the extraction of self-contained\footnote{A \textbf{self-contained} code snippet is one that encapsulates all necessary logic, styling, and dependencies, ensuring it functions independently without requiring external imports or context.} code snippets from real-world projects, renders the corresponding visual outputs, and generates detailed descriptions that link design elements to functional code. To further expand the scope and utility of the synthesis, we introduce three data synthesis strategies: Evolution-based synthesis, which enables scalable and diverse dataset expansion; Waterfall-Model-based synthesis, which generates logically coherent code derived from system requirements; and Additive Development synthesis, which iteratively increases the complexity of human-authored components. We build a large vision-language model, Flame, trained on the synthesized datasets and demonstrate its effectiveness in generating React code via the $\text{pass}@k$ metric. Our results suggest that a code VLM trained to interpret images before code generation may achieve better performance.

cs.CV↗

arXiv2025

Quantifying Point Contributions: A Lightweight Framework for Efficient and Effective Query-Driven Trajectory Simplification

As large volumes of trajectory data accumulate, simplifying trajectories to reduce storage and querying costs is increasingly studied. Existing proposals face three main problems. First, they require numerous iterations to decide which GPS points to delete. Second, they focus only on the relationships between neighboring points (local information) while neglecting the overall structure (global information), reducing the global similarity between the simplified and original trajectories and making it difficult to maintain consistency in query results, especially for similarity-based queries. Finally, they fail to differentiate the importance of points with similar features, leading to suboptimal selection of points to retain the original trajectory information. We propose MLSimp, a novel Mutual Learning query-driven trajectory simplification framework that integrates two distinct models: GNN-TS, based on graph neural networks, and Diff-TS, based on diffusion models. GNN-TS evaluates the importance of a point according to its globality, capturing its correlation with the entire trajectory, and its uniqueness, capturing its differences from neighboring points. It also incorporates attention mechanisms in the GNN layers, enabling simultaneous data integration from all points within the same trajectory and refining representations, thus avoiding iterative processes. Diff-TS generates amplified signals to enable the retention of the most important points at low compression rates. Experiments involving eight baselines on three databases show that MLSimp reduces the simplification time by 42%--70% and improves query accuracy over simplified trajectories by up to 34.6%.

cs.DB↗

arXiv2025

Theoretical Physics Benchmark (TPBench) -- a Dataset and Study of AI Reasoning Capabilities in Theoretical Physics

We introduce a benchmark to evaluate the capability of AI to solve problems in theoretical physics, focusing on high-energy theory and cosmology. The first iteration of our benchmark consists of 57 problems of varying difficulty, from undergraduate to research level. These problems are novel in the sense that they do not come from public problem collections. We evaluate our data set on various open and closed language models, including o3-mini, o1, DeepSeek-R1, GPT-4o and versions of Llama and Qwen. While we find impressive progress in model performance with the most recent models, our research-level difficulty problems are mostly unsolved. We address challenges of auto-verifiability and grading, and discuss common failure modes. While currently state-of-the art models are still of limited use for researchers, our results show that AI assisted theoretical physics research may become possible in the near future. We discuss the main obstacles towards this goal and possible strategies to overcome them. The public problems and solutions, results for various models, and updates to the data set and score distribution, are available on the website of the dataset tpbench.org.

cs.LG↗

arXiv2025

Generation of cosmic ray trajectories by a Diffusion Model trained on test particles in 3D magnetohydrodynamic turbulence

Models for the transport of high energy charged particles through strong magnetic turbulence play a key role in space and astrophysical studies, such as describing the propagation of solar energetic particles and high energy cosmic rays. Inspired by the recent advances in high-performance machine learning techniques, we investigate the application of generative diffusion models to synthesizing test particle trajectories obtained from a turbulent magnetohydrodynamics simulation. We consider velocity increment, spatial transport and curvature statistics, and find excellent agreement with the baseline trajectories for fixed particle energies. Additionally, we consider two synthetic turbulence models for comparison. Finally, challenges towards an application-ready transport model based on our approach are discussed.

physics.flu-dyn↗

arXiv2025

Detecting subtle cyberattacks on adaptive cruise control vehicles: A machine learning approach

With the advent of vehicles equipped with advanced driver-assistance systems, such as adaptive cruise control (ACC) and other automated driving features, the potential for cyberattacks on these automated vehicles (AVs) has emerged. While overt attacks that force vehicles to collide may be easily identified, more insidious attacks, which only slightly alter driving behavior, can result in network-wide increases in congestion, fuel consumption, and even crash risk without being easily detected. To address the detection of such attacks, we first present a traffic model framework for three types of potential cyberattacks: malicious manipulation of vehicle control commands, false data injection attacks on sensor measurements, and denial-of-service (DoS) attacks. We then investigate the impacts of these attacks at both the individual vehicle (micro) and traffic flow (macro) levels. A novel generative adversarial network (GAN)-based anomaly detection model is proposed for real-time identification of such attacks using vehicle trajectory data. We provide numerical evidence {to demonstrate} the efficacy of our machine learning approach in detecting cyberattacks on ACC-equipped vehicles. The proposed method is compared against some recently proposed neural network models and observed to have higher accuracy in identifying anomalous driving behaviors of ACC vehicles.

cs.MA↗

arXiv2024

Open-Sora: Democratizing Efficient Video Production for All

Vision and language are the two foundational senses for humans, and they build up our cognitive ability and intelligence. While significant breakthroughs have been made in AI language ability, artificial visual intelligence, especially the ability to generate and simulate the world we see, is far lagging behind. To facilitate the development and accessibility of artificial visual intelligence, we created Open-Sora, an open-source video generation model designed to produce high-fidelity video content. Open-Sora supports a wide spectrum of visual generation tasks, including text-to-image generation, text-to-video generation, and image-to-video generation. The model leverages advanced deep learning architectures and training/inference techniques to enable flexible video synthesis, which could generate video content of up to 15 seconds, up to 720p resolution, and arbitrary aspect ratios. Specifically, we introduce Spatial-Temporal Diffusion Transformer (STDiT), an efficient diffusion framework for videos that decouples spatial and temporal attention. We also introduce a highly compressive 3D autoencoder to make representations compact and further accelerate training with an ad hoc training strategy. Through this initiative, we aim to foster innovation, creativity, and inclusivity within the community of AI content creation. By embracing the open-source principle, Open-Sora democratizes full access to all the training/inference/data preparation codes as well as model weights. All resources are publicly available at: https://github.com/hpcaitech/Open-Sora.

cs.CV↗

arXiv2024

A theoretical study of the upper bound of surface elevation variance in the Phillips initial stage during wind-wave generation

The resonance mechanism in the initial stage of wind-wave generation proposed by Phillips (J. Fluid Mech., vol. 2, 1957, 417$\unicode{x2013}$445) is a foundation of wind-wave generation theory, but a precise theoretical quantification of wave energy growth in this initial stage has not been obtained yet after more than six decades of research. In this study, we aim to address this knowledge gap by developing an analytical approach based on a novel complex analysis method to theoretically investigate the temporal evolution of the wave energy in the Phillips initial stage. We quantitatively derive and analyse the growth behaviour of the surface wave energy and obtain an analytical solution for its upper bound. Our result highlights the crucial effects of surface tension. Because the phase velocity of gravity$\unicode{x2013}$capillary waves has a minimal value at a critical wavenumber, gravity$\unicode{x2013}$capillary waves and gravity waves (which neglect surface tension) exhibit distinct resonance curve properties and wave energy growth behaviours. For gravity waves, the resonance curve extends indefinitely; for gravity$\unicode{x2013}$capillary waves, it either forms a finite-length curve or does not exist, depending on the wind speed. The leading-order term of the upper-bound solution of the energy of gravity waves increases linearly over time, while for gravity$\unicode{x2013}$capillary waves, the term increases linearly over time under strong wind conditions but remains finite under weak wind conditions. This theoretical study provides an analytical framework for the generation of wind-waves in the Phillips initial stage, which may inspire further theoretical, numerical, and experimental research.

physics.flu-dyn↗

arXiv2024

HalluCana: Fixing LLM Hallucination with A Canary Lookahead

In this paper, we present HalluCana, a canary lookahead to detect and correct factuality hallucinations of Large Language Models (LLMs) in long-form generation. HalluCana detects and intervenes as soon as traces of hallucination emerge, during and even before generation. To support timely detection, we exploit the internal factuality representation in the LLM hidden space, where we investigate various proxies to the LLMs' factuality self-assessment, and discuss its relation to the models' context familiarity from their pre-training. On biography generation, our method improves generation quality by up to 2.5x, while consuming over 6 times less compute.

cs.CL↗

arXiv2024

Digital Twin-Empowered Voltage Control for Power Systems

Emerging digital twin technology has the potential to revolutionize voltage control in power systems. However, the state-of-the-art digital twin method suffers from low computational and sampling efficiency, which hinders its applications. To address this issue, we propose a Gumbel-Consistency Digital Twin (GC-DT) method that enhances voltage control with improved computational and sampling efficiency. First, the proposed method incorporates a Gumbel-based strategy improvement that leverages the Gumbel-top trick to enhance non-repetitive sampling actions and reduce the reliance on Monte Carlo Tree Search simulations, thereby improving computational efficiency. Second, a consistency loss function aligns predicted hidden states with actual hidden states in the latent space, which increases both prediction accuracy and sampling efficiency. Experiments on IEEE 123-bus, 34-bus, and 13-bus systems demonstrate that the proposed GC-DT outperforms the state-of-the-art DT method in both computational and sampling efficiency.

cs.LG↗

arXiv2024

UniTE: A Survey and Unified Pipeline for Pre-training Spatiotemporal Trajectory Embeddings

Spatiotemporal trajectories are sequences of timestamped locations, which enable a variety of analyses that in turn enable important real-world applications. It is common to map trajectories to vectors, called embeddings, before subsequent analyses. Thus, the qualities of embeddings are very important. Methods for pre-training embeddings, which leverage unlabeled trajectories for training universal embeddings, have shown promising applicability across different tasks, thus attracting considerable interest. However, research progress on this topic faces two key challenges: a lack of a comprehensive overview of existing methods, resulting in several related methods not being well-recognized, and the absence of a unified pipeline, complicating the development of new methods and the analysis of methods. We present UniTE, a survey and a unified pipeline for this domain. In doing so, we present a comprehensive list of existing methods for pre-training trajectory embeddings, which includes methods that either explicitly or implicitly employ pre-training techniques. Further, we present a unified and modular pipeline with publicly available underlying code, simplifying the process of constructing and evaluating methods for pre-training trajectory embeddings. Additionally, we contribute a selection of experimental results using the proposed pipeline on real-world datasets. Implementation of the pipeline is publicly available at https://github.com/Logan-Lin/UniTE.

cs.LG↗

arXiv2024

Stochastic Reconstruction of Gappy Lagrangian Turbulent Signals by Conditional Diffusion Models

We present a stochastic method for reconstructing missing spatial and velocity data along the trajectories of small objects passively advected by turbulent flows with a wide range of temporal or spatial scales, such as small balloons in the atmosphere or drifters in the ocean. Our approach makes use of conditional generative diffusion models, a recently proposed data-driven machine learning technique. We solve the problem for two paradigmatic open problems, the case of 3D tracers in homogeneous and isotropic turbulence, and 2D trajectories from the NOAA-funded Global Drifter Program. We show that for both cases, our method is able to reconstruct velocity signals retaining non-trivial scale-by-scale properties that are highly non-Gaussian and intermittent. A key feature of our method is its flexibility in dealing with the location and shape of data gaps, as well as its ability to naturally exploit correlations between different components, leading to superior accuracy, with respect to Gaussian process regressions, for both pointwise reconstruction and statistical expressivity. Our method shows promising applications also to a wide range of other Lagrangian problems, including multi-particle dispersion in turbulence, dynamics of charged particles in astrophysics and plasma physics, and pedestrian dynamics.

physics.flu-dyn↗