arXiv subjects

Tiantian Chen

Publications and source records attributed to Tiantian Chen.

16 recordsLinked to original sources

arXiv2026

Why Database Manuals Are Not Enough: Efficient and Reliable Configuration Tuning for DBMSs via Code-Driven LLM Agents

Modern database management systems (DBMSs) expose hundreds of configuration knobs that critically influence performance. Existing automated tuning methods either adopt a data-driven paradigm, which incurs substantial overhead, or rely on manual-driven heuristics extracted from database documentation, which are often limited and overly generic. Motivated by the fact that the control logic of configuration knobs is inherently encoded in the DBMS source code, we argue that promising tuning strategies can be mined directly from the code, uncovering fine-grained insights grounded in system internals. To this end, we propose SysInsight, a code-driven database tuning system that automatically extracts fine-grained tuning knowledge from DBMS source code to accelerate and stabilize the tuning process. SysInsight combines static code analysis with LLM-based reasoning to identify knob-controlled execution paths and extract semantic tuning insights. These insights are then transformed into quantitative and verifiable tuning rules via association rule mining grounded in tuning observations. During online tuning, system diagnosis is applied to identify critical knobs, which are adjusted under the rule guidance. Evaluations demonstrate that compared to the SOTA baseline, SysInsight converges to the best configuration on average 7.11X faster while achieving a 19.9% performance improvement.

cs.DB↗

arXiv2026

ES-MemEval: Benchmarking Conversational Agents on Personalized Long-Term Emotional Support

Large Language Models (LLMs) have shown strong potential as conversational agents. Yet, their effectiveness remains limited by deficiencies in robust long-term memory, particularly in complex, long-term web-based services such as online emotional support. However, existing long-term dialogue benchmarks primarily focus on static and explicit fact retrieval, failing to evaluate agents in critical scenarios where user information is dispersed, implicit, and continuously evolving. To address this gap, we introduce ES-MemEval, a comprehensive benchmark that systematically evaluates five core memory capabilities: information extraction, temporal reasoning, conflict detection, abstention, and user modeling, in long-term emotional support settings, covering question answering, summarization, and dialogue generation tasks. To support the benchmark, we also propose EvoEmo, a multi-session dataset for personalized long-term emotional support that captures fragmented, implicit user disclosures and evolving user states. Extensive experiments on open-source long-context, commercial, and retrieval-augmented (RAG) LLMs show that explicit long-term memory is essential for reducing hallucinations and enabling effective personalization. At the same time, RAG improves factual consistency but struggles with temporal dynamics and evolving user states. These findings highlight both the potential and limitations of current paradigms and motivate more robust integration of memory and retrieval for long-term personalized dialogue systems.

cs.CL↗

arXiv2025

RESPOND: Risk-Enhanced Structured Pattern for LLM-driven Online Node-level Decision-making

Current LLM-based driving agents that rely on unstructured plain-text memory suffer from low-precision scene retrieval and inefficient reflection. To address this limitation, we present RESPOND, a structured decision-making framework for LLM-driven agents grounded in explicit risk patterns. RESPOND represents each ego-centric scene using a unified 5 by 3 matrix that encodes spatial topology and road constraints, enabling consistent and reliable retrieval of spatial risk configurations. Based on this representation, a hybrid rule and LLM decision pipeline is developed with a two-tier memory mechanism. In high-risk contexts, exact pattern matching enables rapid and safe reuse of verified actions, while in low-risk contexts, sub-pattern matching supports personalized driving style adaptation. In addition, a pattern-aware reflection mechanism abstracts tactical corrections from crash and near-miss frames to update structured memory, achieving one-crash-to-generalize learning. Extensive experiments demonstrate the effectiveness of RESPOND. In highway-env, RESPOND outperforms state-of-the-art LLM-based and reinforcement learning based driving agents while producing substantially fewer collisions. With step-wise human feedback, the agent acquires a Sporty driving style within approximately 20 decision steps through sub-pattern abstraction. For real-world validation, RESPOND is evaluated on 53 high-risk cut-in scenarios extracted from the HighD dataset. For each event, intervention is applied immediately before the cut-in and RESPOND re-decides the driving action. Compared to recorded human behavior, RESPOND reduces subsequent risk in 84.9 percent of scenarios, demonstrating its practical feasibility under real-world driving conditions. These results highlight RESPONDs potential for autonomous driving, personalized driving assistance, and proactive hazard mitigation.

cs.HC↗

arXiv2025

SMART: Scalable Multi-Agent Reasoning and Trajectory Planning in Dense Environments

Multi-vehicle trajectory planning is a non-convex problem that becomes increasingly difficult in dense environments due to the rapid growth of collision constraints. Efficient exploration of feasible behaviors and resolution of tight interactions are essential for real-time, large-scale coordination. This paper introduces SMART, Scalable Multi-Agent Reasoning and Trajectory Planning, a hierarchical framework that combines priority-based search with distributed optimization to achieve efficient and feasible multi-vehicle planning. The upper layer explores diverse interaction modes using reinforcement learning-based priority estimation and large-step hybrid A* search, while the lower layer refines solutions via parallelizable convex optimization. By partitioning space among neighboring vehicles and constructing robust feasible corridors, the method decouples the joint non-convex problem into convex subproblems solved efficiently in parallel. This design alleviates the step-size trade-off while ensuring kinematic feasibility and collision avoidance. Experiments show that SMART consistently outperforms baselines. On 50 m x 50 m maps, it sustains over 90% success within 1 s up to 25 vehicles, while baselines often drop below 50%. On 100 m x 100 m maps, SMART achieves above 95% success up to 50 vehicles and remains feasible up to 90 vehicles, with runtimes more than an order of magnitude faster than optimization-only approaches. Built on vehicle-to-everything communication, SMART incorporates vehicle-infrastructure cooperation through roadside sensing and agent coordination, improving scalability and safety. Real-world experiments further validate this design, achieving planning times as low as 0.014 s while preserving cooperative behaviors.

cs.RO↗

arXiv2025

M3HG: Multimodal, Multi-scale, and Multi-type Node Heterogeneous Graph for Emotion Cause Triplet Extraction in Conversations

Emotion Cause Triplet Extraction in Multimodal Conversations (MECTEC) has recently gained significant attention in social media analysis, aiming to extract emotion utterances, cause utterances, and emotion categories simultaneously. However, the scarcity of related datasets, with only one published dataset featuring highly uniform dialogue scenarios, hinders model development in this field. To address this, we introduce MECAD, the first multimodal, multi-scenario MECTEC dataset, comprising 989 conversations from 56 TV series spanning a wide range of dialogue contexts. In addition, existing MECTEC methods fail to explicitly model emotional and causal contexts and neglect the fusion of semantic information at different levels, leading to performance degradation. In this paper, we propose M3HG, a novel model that explicitly captures emotional and causal contexts and effectively fuses contextual information at both inter- and intra-utterance levels via a multimodal heterogeneous graph. Extensive experiments demonstrate the effectiveness of M3HG compared with existing state-of-the-art methods. The codes and dataset are available at https://github.com/redifinition/M3HG.

cs.CL↗

arXiv2024

MetaSSC: Enhancing 3D Semantic Scene Completion for Autonomous Driving through Meta-Learning and Long-sequence Modeling

Semantic scene completion (SSC) is essential for achieving comprehensive perception in autonomous driving systems. However, existing SSC methods often overlook the high deployment costs in real-world applications. Traditional architectures, such as 3D Convolutional Neural Networks (3D CNNs) and self-attention mechanisms, face challenges in efficiently capturing long-range dependencies within 3D voxel grids, limiting their effectiveness. To address these issues, we introduce MetaSSC, a novel meta-learning-based framework for SSC that leverages deformable convolution, large-kernel attention, and the Mamba (D-LKA-M) model. Our approach begins with a voxel-based semantic segmentation (SS) pretraining task, aimed at exploring the semantics and geometry of incomplete regions while acquiring transferable meta-knowledge. Using simulated cooperative perception datasets, we supervise the perception training of a single vehicle using aggregated sensor data from multiple nearby connected autonomous vehicles (CAVs), generating richer and more comprehensive labels. This meta-knowledge is then adapted to the target domain through a dual-phase training strategy that does not add extra model parameters, enabling efficient deployment. To further enhance the model's capability in capturing long-sequence relationships within 3D voxel grids, we integrate Mamba blocks with deformable convolution and large-kernel attention into the backbone network. Extensive experiments demonstrate that MetaSSC achieves state-of-the-art performance, significantly outperforming competing models while also reducing deployment costs.

cs.CV↗

arXiv2024

An Experimental Configuration to Study High-Enthalpy Radiating Flows Under Nonequilibrium De-excitation

This paper introduces an experimental configuration, called the Prandtl-Meyer plus duct arrangement (PMD), designed to study high-enthalpy radiating flows undergoing nonequilibrium de-excitation. The original design proposed by Wilson, developed without the benefit of modern computational fluid dynamics (CFD), was inadequate for generating a sufficiently large undisturbed zone or achieving a uniform flow along the centerline, necessitating further refinement. Consequently, significant modifications were implemented to enhance PMD's performance, resulting in an expanded undisturbed zone and a uniform centerline flow that facilitate the measurements of nonequilibrium de-excitation.} A general design method is introduced, combining theoretical analysis and numerical simulations to tailor the flow conditions for various research objectives. The implementation involves considerations of the shock tube conditions, PMD configuration, and the effective measurement zone. The interplay between shock tube conditions and airfoil geometry generates diverse de-excitation patterns. The shock tube test time, transition onset location, and radiance intensity determine the effective measurement zone. An example utilizing N2 as the test gas demonstrates the method, achieving one-dimensional flow with thermal nonequilibrium and chemical freezing along the centerline, validating the method's effectiveness. An effective measurement zone of 200 mm is obtained under this condition, and the primary constraint under high-enthalpy conditions is the limited shock tube test time due to the high shock velocity and low fill pressure.

physics.flu-dyn↗

arXiv2023

ToupleGDD: A Fine-Designed Solution of Influence Maximization by Deep Reinforcement Learning

Aiming at selecting a small subset of nodes with maximum influence on networks, the Influence Maximization (IM) problem has been extensively studied. Since it is #P-hard to compute the influence spread given a seed set, the state-of-the-art methods, including heuristic and approximation algorithms, faced with great difficulties such as theoretical guarantee, time efficiency, generalization, etc. This makes it unable to adapt to large-scale networks and more complex applications. On the other side, with the latest achievements of Deep Reinforcement Learning (DRL) in artificial intelligence and other fields, lots of works have been focused on exploiting DRL to solve combinatorial optimization problems. Inspired by this, we propose a novel end-to-end DRL framework, ToupleGDD, to address the IM problem in this paper, which incorporates three coupled graph neural networks for network embedding and double deep Q-networks for parameters learning. Previous efforts to solve IM problem with DRL trained their models on subgraphs of the whole network, and then tested on the whole graph, which makes the performance of their models unstable among different networks. However, our model is trained on several small randomly generated graphs with a small budget, and tested on completely different networks under various large budgets, which can obtain results very close to IMM and better results than OPIM-C on several datasets, and shows strong generalization ability. Finally, we conduct a large number of experiments on synthetic and realistic datasets, and experimental results prove the effectiveness and superiority of our model.

cs.SI↗

arXiv2022

Graph Representation Learning for Popularity Prediction Problem: A Survey

The online social platforms, like Twitter, Facebook, LinkedIn and WeChat, have grown really fast in last decade and have been one of the most effective platforms for people to communicate and share information with each other. Due to the "word of mouth" effects, information usually can spread rapidly on these social media platforms. Therefore, it is important to study the mechanisms driving the information diffusion and quantify the consequence of information spread. A lot of efforts have been focused on this problem to help us better understand and achieve higher performance in viral marketing and advertising. On the other hand, the development of neural networks has blossomed in the last few years, leading to a large number of graph representation learning (GRL) models. Compared to traditional models, GRL methods are often shown to be more effective. In this paper, we present a comprehensive review for existing works using GRL methods for popularity prediction problem, and categorize related literatures into two big classes, according to their mainly used model and techniques: embedding-based methods and deep learning methods. Deep learning method is further classified into six small classes: convolutional neural networks, graph convolutional networks, graph attention networks, graph neural networks, recurrent neural networks, and reinforcement learning. We compare the performance of these different models and discuss their strengths and limitations. Finally, we outline the challenges and future chances for popularity prediction problem.

cs.SI↗

arXiv2022

Blockchain Driven Privacy Preserving Contact Tracing Framework in Pandemics

Contact tracing has been proven an effective approach to control the virus spread in pandemics like COVID-19 pandemic. As an emerging powerful decentralized technique, blockchain has been explored to ensure data privacy and security in contact tracing processes. However, existing works are mostly high-level designs with no sufficient demonstration and treat blockchain as separate storage system assisting third-party central servers, ignoring the importance and capability of consensus mechanism and incentive mechanism. In this paper, we propose a light-weight and fully third-party free Blockchain-Driven Contact Tracing framework (BDCT) to bridge the gap. In the BDCT framework, RSA encryption based transaction verification method (RSA-TVM) is proposed to ensure contact tracing correctness, which can achieve more than 96\% contact cases recording accuracy even each person has 60\% probability of failing to verify the contact information. Reputation Corrected Delegated Proof of Stake (RC-DPoS) consensus mechanism is proposed together with the incentive mechanism, which can ensure timeliness of reporting contact cases and keep blockchain decentralized. A novel contact tracing simulation environment is created, which considers three different contact scenarios based on population density. The simulation results demonstrate the effectiveness, robustness and attack resistance of RSA-TVM and RC-DPoS in the proposed BDCT.

cs.CR↗

arXiv2022

Adaptive Multi-Feature Budgeted Profit Maximization in Social Networks

Online social network has been one of the most important platforms for viral marketing. Most of existing researches about diffusion of adoptions of new products on networks are about one diffusion. That is, only one piece of information about the product is spread on the network. However, in fact, one product may have multiple features and the information about different features may spread independently in social network. When a user would like to purchase the product, he would consider all of the features of the product comprehensively not just consider one. Based on this, we propose a novel problem, multi-feature budgeted profit maximization (MBPM) problem, which first considers budgeted profit maximization under multiple features propagation of one product. Given a social network with each node having an activation cost and a profit, MBPM problem seeks for a seed set with expected cost no more than the budget to make the total expected profit as large as possible. We consider MBPM problem under the adaptive setting, where seeds are chosen iteratively and next seed is selected according to current diffusion results. We study adaptive MBPM problem under two models, oracle model and noise model. The oracle model assumes conditional expected marginal profit of any node could be obtained in O(1) time and a (1-1/e) expected approximation policy is proposed. Under the noise model, we estimate conditional expected marginal profit of a node by modifying the EPIC algorithm and propose an efficient policy, which could return a (1-exp(ε-1)) expected approximation ratio. Several experiments are conducted on six realistic datasets to compare our proposed policies with their corresponding non-adaptive algorithms and some heuristic adaptive policies. Experimental results show efficiencies and superiorities of our policies.

cs.SI↗

arXiv2021

Urban traffic dynamic rerouting framework: A DRL-based model with fog-cloud architecture

Past research and practice have demonstrated that dynamic rerouting framework is effective in mitigating urban traffic congestion and thereby improve urban travel efficiency. It has been suggested that dynamic rerouting could be facilitated using emerging technologies such as fog-computing which offer advantages of low-latency capabilities and information exchange between vehicles and roadway infrastructure. To address this question, this study proposes a two-stage model that combines GAQ (Graph Attention Network - Deep Q Learning) and EBkSP (Entropy Based k Shortest Path) using a fog-cloud architecture, to reroute vehicles in a dynamic urban environment and therefore to improve travel efficiency in terms of travel speed. First, GAQ analyzes the traffic conditions on each road and for each fog area, and then assigns a road index based on the information attention from both local and neighboring areas. Second, EBkSP assigns the route for each vehicle based on the vehicle priority and route popularity. A case study experiment is carried out to investigate the efficacy of the proposed model. At the model training stage, different methods are used to establish the vehicle priorities, and their impact on the results is assessed. Also, the proposed model is tested under various scenarios with different ratios of rerouting and background (non-rerouting) vehicles. The results demonstrate that vehicle rerouting using the proposed model can help attain higher speed and reduces possibility of severe congestion. This result suggests that the proposed model can be deployed by urban transportation agencies for dynamic rerouting and ultimately, to reduce urban traffic congestion.

cs.AI↗

arXiv2021

Development and testing of an image transformer for explainable autonomous driving systems

In the last decade, deep learning (DL) approaches have been used successfully in computer vision (CV) applications. However, DL-based CV models are generally considered to be black boxes due to their lack of interpretability. This black box behavior has exacerbated user distrust and therefore has prevented widespread deployment DLCV models in autonomous driving tasks even though some of these models exhibit superiority over human performance. For this reason, it is essential to develop explainable DL models for autonomous driving task. Explainable DL models can not only boost user trust in autonomy but also serve as a diagnostic approach to identify anydefects and weaknesses of the model during the system development phase. In this paper, we propose an explainable end-to-end autonomous driving system based on "Transformer", a state-of-the-art (SOTA) self-attention based model, to map visual features from images collected by onboard cameras to guide potential driving actions with corresponding explanations. The model achieves a soft attention over the global features of the image. The results demonstrate the efficacy of our proposed model as it exhibits superior performance (in terms of correct prediction of actions and explanations) compared to the benchmark model by a significant margin with lower computational cost.

cs.CV↗

arXiv2021

A Random Algorithm for Profit Maximization with Multiple Adoptions in Online Social Networks

Online social networks have been one of the most effective platforms for marketing and advertising. Through "word of mouth" effects, information or product adoption could spread from some influential individuals to millions of users in social networks. Given a social network $G$ and a constant $k$, the influence maximization problem seeks for $k$ nodes in $G$ that can influence the largest number of nodes. This problem has found important applications, and a large amount of works have been devoted to identifying the few most influential users. But most of existing works only focus on the diffusion of a single idea or product in social networks. However, in reality, one company may produce multiple kinds of products and one user may also have multiple adoptions. Given multiple kinds of different products with different activation costs and profits, it is crucial for the company to distribute the limited budget among multiple products in order to achieve profit maximization. Profit Maximization with Multiple Adoptions (PM$^{2}$A) problem aims to seek for a seed set within the budget to maximize the overall profit. In this paper, a Randomized Modified Greedy (RMG) algorithm based on the Reverse Influence Sampling (RIS) technique is presented for the PM$^{2}$A problem, which could achieve a $(1-1/e-\varepsilon)$-approximate solution with high probability. Compared with the algorithm proposed in [16] that achieves a $\frac{1}{2}(1-1/e^{2})$-approximate solution, our algorithm provides a better performance ratio which is also the best performance ratio of the PM$^{2}$A problem. Comprehensive experiments on three real-world social networks are conducted, and the results demonstrate that our RMG algorithm outperforms the algorithm proposed in [16] and other heuristics in terms of profit maximization, and could better allocate the budget.

cs.SI↗

arXiv2020

Continuous Activity Maximization in Online Social Networks

Activity maximization is a task of seeking a small subset of users in a given social network that makes the expected total activity benefit maximized. This is a generalization of many real applications. In this paper, we extend activity maximization problem to that under the general marketing strategy $\vec{x}$, which is a $d$-dimensional vector from a lattice space and has probability $h_u(\vec{x})$ to activate a node $u$ as a seed. Based on that, we propose the continuous activity maximization (CAM) problem, where the domain is continuous and the seed set we select conforms to a certain probability distribution. It is a new topic to study the problem about information diffusion under the lattice constraint, thus, we address the problem systematically here. First, we analyze the hardness of CAM and how to compute the objective function of CAM accurately and effectively. We prove this objective function is monotone, but not DR-submodular and not DR-supermodular. Then, we develop a monotone and DR-submodular lower bound and upper bound of CAM, and apply sampling techniques to design three unbiased estimators for CAM, its lower bound and upper bound. Next, adapted from IMM algorithm and sandwich approximation framework, we obtain a data-dependent approximation ratio. This process can be considered as a general method to solve those maximization problem on lattice but not DR-submodular. Last, we conduct experiments on three real-world datasets to evaluate the correctness and effectiveness of our proposed algorithms.

cs.SI↗

arXiv2019

A Multi-Feature Diffusion Model: Rumor Blocking in Social Networks

Online social networks provide a convenient platform for the spread of rumors, which could lead to serious aftermaths such as economic losses and public panic. The classical rumor blocking problem aims to launch a set of nodes as a positive cascade to compete with misinformation in order to limit the spread of rumors. However, most of the related researches were based on one-dimensional diffusion model. In reality, there are more than one feature associated with an object. The user's impression on this object is determined not just by one feature but by his/her overall evaluation on all of these features. Thus, the influence spread of this object can be decomposed into the spread of multiple features. Based on that, we propose a Multi-Feature diffusion model (MF-model) in this paper, and a novel problem, Multi-Feature Rumor Blocking (MFRB), is formulated on a multi-layer network structure according to this model. To solve MFRB, we design a creative sampling method, called Multi-Sampling, which can be applied to a multi-layer network structure. Inspired by martingale analysis, the Revised-IMM algorithm is proposed, and returns a satisfactory approximate solution to MFRB. Finally, we evaluate our proposed algorithm by conducting experiments on real datasets, and show the effectiveness and accuracy of the Revised-IMM algorithm and significantly outperforms other baseline algorithms.

cs.SI↗