arXiv subjects

Cheng Lyu

Publications and source records attributed to Cheng Lyu.

14 recordsLinked to original sources

arXiv2026

The Differential Reasoning Router: Operationalizing Cost-Aware LLM Annotation in E-commerce

Large Language Models (LLMs) are increasingly used to annotate structured product data in e-commerce, but early deployment often begins as a cold-start problem: only limited pre-launch labels are available, the value of expensive reasoning is unknown, and human review is needed before the system can be trusted at scale. This challenge is especially common in rule-based annotation workflows, where each item must satisfy multiple business rules and both model errors and ambiguous rule boundaries affect final decisions. We introduce the Differential Reasoning Router (DRR), a cost-aware framework for cold-start LLM annotation that jointly optimizes model selection and human escalation. Rather than treating a reasoning model as a default fallback, DRR estimates separate success probabilities for a direct model and a reasoning model at both the sample and business-rule levels, enabling adaptive routing: easy cases are handled directly, reasoning is reserved for cases where it is expected to improve the decision, and likely double-failure or rule-disagreement cases are escalated to human annotators. The resulting labels provide targeted ground truth for prompt engineering, supervised fine-tuning, calibration, and rule refinement, enabling a gradual shift from human-heavy cold-start annotation toward high-confidence automated routing. In a production e-commerce workflow, DRR reaches accuracy parity with the strongest confidence-based router while achieving more than 60\% reasoning-token cost savings.

cs.CL↗

arXiv2026

From Pixels to Purchase: Building and Evaluating a Taxonomy-Decoupled Visual Search Engine for Home Goods E-commerce

Visual search is critical for e-commerce, especially in style-driven domains where user intent is subjective and open-ended. Existing industrial systems typically couple object detection with taxonomy-based classification and rely on catalog data for evaluation, which is prone to noise that limits robustness and scalability. We propose a taxonomy-decoupled architecture that uses classification-free region proposals and unified embeddings for similarity retrieval, enabling a more flexible and generalizable visual search. To overcome the evaluation bottleneck, we propose an LLM-as-a-Judge framework that assesses nuanced visual similarity and category relevance for query-result pairs in a zero-shot manner, removing dependence on human annotations or noise-prone catalog data. Deployed at scale on a global home goods platform, our system improves retrieval quality and yields a measurable uplift in customer engagement, while our offline evaluation metrics strongly correlate with real-world outcomes.

cs.CV↗

arXiv2025

A Generalized $\chi_n$-Function

The mapping $\chi_n$ from $\F_{2}^{n}$ to itself defined by $y=\chi_n(x)$ with $y_i=x_i+x_{i+2}(1+x_{i+1})$, where the indices are computed modulo $n$, has been widely studied for its applications in lightweight cryptography. However, $\chi_n $ is bijective on $\F_2^n$ only when $n$ is odd, restricting its use to odd-dimensional vector spaces over $\F_2$. To address this limitation, we introduce and analyze the generalized mapping $\chi_{n, m}$ defined by $y=\chi_{n,m}(x)$ with $y_i=x_i+x_{i+m} (x_{i+m-1}+1)(x_{i+m-2}+1) \cdots (x_{i+1}+1)$, where $m$ is a fixed integer with $m\nmid n$. To investigate such mappings, we further generalize $\chi_{n,m}$ to $\theta_{m, k}$, where $\theta_{m, k}$ is given by $y_i=x_{i+mk} \prod_{\substack{j=1,\,\, m \nmid j}}^{mk-1} \left(x_{i+j}+1\right), \,\,{\rm for }\,\, i\in \{0,1,\ldots,n-1\}$. We prove that these mappings generate an abelian group isomorphic to the group of units in $\F_2[z]/(z^{\lfloor n/m\rfloor +1})$. This structural insight enables us to construct a broad class of permutations over $\F_2^n$ for any positive integer $n$, along with their inverses. We rigorously analyze algebraic properties of these mappings, including their iterations, fixed points, and cycle structures. Additionally, we provide a comprehensive database of the cryptographic properties for iterates of $\chi_{n,m}$ for small values of $n$ and $m$. Finally, we conduct a comparative security and implementation cost analysis among $\chi_{n,m}$, $\chi_n$, $\chi\chi_n$ (EUROCRYPT 2025 \cite{belkheyar2025chi}) and their variants, and prove Conjecture~1 proposed in~\cite{belkheyar2025chi} as a by-product of our study. Our results lead to generalizations of $\chi_n$, providing alternatives to $\chi_n$ and $\chi\chi_n$.

cs.CR↗

arXiv2025

Scalable and Reliable Multi-agent Reinforcement Learning for Traffic Assignment

The evolution of metropolitan cities and the increase in travel demands impose stringent requirements on traffic assignment methods. Multi-agent reinforcement learning (MARL) approaches outperform traditional methods in modeling adaptive routing behavior without requiring explicit system dynamics, which is beneficial for real-world deployment. However, MARL frameworks face challenges in scalability and reliability when managing extensive networks with substantial travel demand, which limiting their practical applicability in solving large-scale traffic assignment problems. To address these challenges, this study introduces MARL-OD-DA, a new MARL framework for the traffic assignment problem, which redefines agents as origin-destination (OD) pair routers rather than individual travelers, significantly enhancing scalability. Additionally, a Dirichlet-based action space with action pruning and a reward function based on the local relative gap are designed to enhance solution reliability and improve convergence efficiency. Experiments demonstrate that the proposed MARL framework effectively handles medium-sized networks with extensive and varied city-level OD demand, surpassing existing MARL methods. When implemented in the SiouxFalls network, MARL-OD-DA achieves better assignment solutions in 10 steps, with a relative gap that is 94.99% lower than that of conventional methods.

cs.LG↗

arXiv2025

Reinforcement Learning-based Sequential Route Recommendation for System-Optimal Traffic Assignment

Modern navigation systems and shared mobility platforms increasingly rely on personalized route recommendations to improve individual travel experience and operational efficiency. However, a key question remains: can such sequential, personalized routing decisions collectively lead to system-optimal (SO) traffic assignment? This paper addresses this question by proposing a learning-based framework that reformulates the static SO traffic assignment problem as a single-agent deep reinforcement learning (RL) task. A central agent sequentially recommends routes to travelers as origin-destination (OD) demands arrive, to minimize total system travel time. To enhance learning efficiency and solution quality, we develop an MSA-guided deep Q-learning algorithm that integrates the iterative structure of traditional traffic assignment methods into the RL training process. The proposed approach is evaluated on both the Braess and Ortuzar-Willumsen (OW) networks. Results show that the RL agent converges to the theoretical SO solution in the Braess network and achieves only a 0.35% deviation in the OW network. Further ablation studies demonstrate that the route action set's design significantly impacts convergence speed and final performance, with SO-informed route sets leading to faster learning and better outcomes. This work provides a theoretically grounded and practically relevant approach to bridging individual routing behavior with system-level efficiency through learning-based sequential assignment.

cs.AI↗

arXiv2024

AI-Driven Day-to-Day Route Choice

Understanding travelers' route choices can help policymakers devise optimal operational and planning strategies for both normal and abnormal circumstances. However, existing choice modeling methods often rely on predefined assumptions and struggle to capture the dynamic and adaptive nature of travel behavior. Recently, Large Language Models (LLMs) have emerged as a promising alternative, demonstrating remarkable ability to replicate human-like behaviors across various fields. Despite this potential, their capacity to accurately simulate human route choice behavior in transportation contexts remains doubtful. To satisfy this curiosity, this paper investigates the potential of LLMs for route choice modeling by introducing an LLM-empowered agent, "LLMTraveler." This agent integrates an LLM as its core, equipped with a memory system that learns from past experiences and makes decisions by balancing retrieved data and personality traits. The study systematically evaluates the LLMTraveler's ability to replicate human-like decision-making through two stages of day-to-day (DTD) congestion games: (1) analyzing its route-switching behavior in single origin-destination (OD) pair scenarios, where it demonstrates patterns that align with laboratory data but cannot be fully explained by traditional models, and (2) testing its capacity to model adaptive learning behaviors in multi-OD scenarios on the Ortuzar and Willumsen (OW) network, producing results comparable to Multinomial Logit (MNL) and Reinforcement Learning (RL) models. These experiments demonstrate that the framework can partially replicate human-like decision-making in route choice while providing natural language explanations for its decisions. This capability offers valuable insights for transportation policymaking, such as simulating traveler responses to new policies or changes in the network.

cs.LG↗

arXiv2024

Squat: Quant Small Language Models on the Edge

A growing trend has emerged in designing high-quality Small Language Models (SLMs) with a few million parameters. This trend is driven by the increasing concerns over cloud costs, privacy, and latency. Considering that full parameter training is feasible for SLMs on mobile devices, Quantization-Aware Training (QAT) is employed to improve efficiency by reducing computational overhead and memory footprint. However, previous QAT works adopt fine-grained quantization methods to compress models with billions of parameters on GPUs, incompatible with current commodity hardware, such as mobile and edge devices, which relies on Single Instruction Multiple Data (SIMD) instructions. Thus, the generalization of these methods to SLMs on mobile devices is limited. In this paper, we propose Squat method, an effective QAT framework with deployable quantization for SLMs on mobile devices. Specifically, we propose entropy-guided and distribution-aligned distillation to mitigate the distortion of attention information from quantization. Besides, we employ sub-8-bit token adaptive quantization, assigning varying bit widths to different tokens based on their importance. Furthermore, we develop a SIMD-based Multi-Kernel Mixed-Precision (MKMP) multiplier to support sub-8-bit mixed-precision MAC on mobile devices. Our extensive experiments verify the substantial improvements of our method compared to other QAT methods across various datasets. Furthermore, we achieve an on-device speedup of up to 2.37x compared with its FP16 counterparts, signaling a great advancement. Code: https://github.com/shawnricecake/squant

cs.LG↗

arXiv2023

Traffic4cast at NeurIPS 2022 -- Predict Dynamics along Graph Edges from Sparse Node Data: Whole City Traffic and ETA from Stationary Vehicle Detectors

The global trends of urbanization and increased personal mobility force us to rethink the way we live and use urban space. The Traffic4cast competition series tackles this problem in a data-driven way, advancing the latest methods in machine learning for modeling complex spatial systems over time. In this edition, our dynamic road graph data combine information from road maps, $10^{12}$ probe data points, and stationary vehicle detectors in three cities over the span of two years. While stationary vehicle detectors are the most accurate way to capture traffic volume, they are only available in few locations. Traffic4cast 2022 explores models that have the ability to generalize loosely related temporal vertex data on just a few nodes to predict dynamic future traffic states on the edges of the entire road graph. In the core challenge, participants are invited to predict the likelihoods of three congestion classes derived from the speed levels in the GPS data for the entire road graph in three cities 15 min into the future. We only provide vehicle count data from spatially sparse stationary vehicle detectors in these three cities as model input for this task. The data are aggregated in 15 min time bins for one hour prior to the prediction time. For the extended challenge, participants are tasked to predict the average travel times on super-segments 15 min into the future - super-segments are longer sequences of road segments in the graph. The competition results provide an important advance in the prediction of complex city-wide traffic states just from publicly available sparse vehicle data and without the need for large amounts of real-time floating vehicle data.

cs.LG↗

arXiv2022

Privileged Prior Information Distillation for Image Matting

Performance of trimap-free image matting methods is limited when trying to decouple the deterministic and undetermined regions, especially in the scenes where foregrounds are semantically ambiguous, chromaless, or high transmittance. In this paper, we propose a novel framework named Privileged Prior Information Distillation for Image Matting (PPID-IM) that can effectively transfer privileged prior environment-aware information to improve the performance of students in solving hard foregrounds. The prior information of trimap regulates only the teacher model during the training stage, while not being fed into the student network during actual inference. In order to achieve effective privileged cross-modality (i.e. trimap and RGB) information distillation, we introduce a Cross-Level Semantic Distillation (CLSD) module that reinforces the trimap-free students with more knowledgeable semantic representations and environment-aware information. We also propose an Attention-Guided Local Distillation module that efficiently transfers privileged local attributes from the trimap-based teacher to trimap-free students for the guidance of local-region optimization. Extensive experiments demonstrate the effectiveness and superiority of our PPID framework on the task of image matting. In addition, our trimap-free IndexNet-PPID surpasses the other competing state-of-the-art methods by a large margin, especially in scenarios with chromaless, weak texture, or irregular objects.

cs.CV↗

arXiv2022

Similarity-based Feature Extraction for Large-scale Sparse Traffic Forecasting

Short-term traffic forecasting is an extensively studied topic in the field of intelligent transportation system. However, most existing forecasting systems are limited by the requirement of real-time probe vehicle data because of their formulation as a time series forecasting problem. Towards this issue, the NeurIPS 2022 Traffic4cast challenge is dedicated to predicting the citywide traffic states with publicly available sparse loop count data. This technical report introduces our second-place winning solution to the extended challenge of ETA prediction. We present a similarity-based feature extraction method using multiple nearest neighbor (NN) filters. Similarity-based features, static features, node flow features and combined features of segments are extracted for training the gradient boosting decision tree model. Experimental results on three cities (including London, Madrid and Melbourne) demonstrate the strong predictive performance of our approach, which outperforms a number of graph-neural-network-based solutions in the task of travel time estimation. The source code is available at \url{https://github.com/c-lyu/Traffic4Cast2022-TSE}.

cs.LG↗

arXiv2022

Efficient Meta-Tuning for Content-aware Neural Video Delivery

Recently, Deep Neural Networks (DNNs) are utilized to reduce the bandwidth and improve the quality of Internet video delivery. Existing methods train corresponding content-aware super-resolution (SR) model for each video chunk on the server, and stream low-resolution (LR) video chunks along with SR models to the client. Although they achieve promising results, the huge computational cost of network training limits their practical applications. In this paper, we present a method named Efficient Meta-Tuning (EMT) to reduce the computational cost. Instead of training from scratch, EMT adapts a meta-learned model to the first chunk of the input video. As for the following chunks, it fine-tunes the partial parameters selected by gradient masking of previous adapted model. In order to achieve further speedup for EMT, we propose a novel sampling strategy to extract the most challenging patches from video frames. The proposed strategy is highly efficient and brings negligible additional cost. Our method significantly reduces the computational cost and achieves even better performance, paving the way for applying neural video delivery techniques to practical applications. We conduct extensive experiments based on various efficient SR architectures, including ESPCN, SRCNN, FSRCNN and EDSR-1, demonstrating the generalization ability of our work. The code is released at \url{https://github.com/Neural-video-delivery/EMT-Pytorch-ECCV2022}.

cs.CV↗

arXiv2021

Peer-to-Peer Energy Cooperation in Building Community over A Lossy Network

Energy management of buildings is of vital importance for the urban low-carbon transition. This paper proposes a sustainable energy cooperation framework for the building community by communication-efficient peer-to-peer transaction. Firstly, the energy cooperation of buildings is formulated as a social welfare maximization problem, in which buildings may directly trade energy with neighbors. In addition, considering privacy concerns and communication losses arisen in peer-to-peer energy trading, a communication-failure-robust distributed algorithm is developed to achieve the optimal energy dispatch solutions. Finally, simulation results show that the proposed framework substantially reduces the total cost of the building community and the algorithm is robust to communication losses in the network when only part of links (even one link) are active during iterations.

eess.SY↗

arXiv2021

Real-time Operation Optimization of Microgrids with Battery Energy Storage System: A Tube-based Model Predictive Control Approach

Battery energy storage systems (ESS) are widely used in microgrids to complement high renewables. However, the real-time energy management of microgrids with battery ESS is challenging in two aspects: 1) the evolution process of battery energy level is across-time coupled; 2) uncertainties unavoidably arise in the forecasting process for renewable generation. In this paper, a tube-based model predictive control (MPC) approach is innovatively proposed in accommodating the real-time energy management of microgrids with battery ESS. Firstly, a real-time operation model of battery, including the degradation cost and time-aware SoC range, is proposed for the battery ESS. In particular, the battery feature shallower-cheaper is depicted and the terminal SoC requirement is achieved. Secondly, two cascaded MPC controllers are designed in the proposed tube-based MPC, in which reference trajectories are generated by the nominal MPC without uncertainties, and then the ancillary MPC steers the actual trajectories to the nominal ones upon the realization of uncertainties. Specifically, in this paper, the battery SoC is viewed as the state variable of the system, while the generator power output and exchange power with the utility are seen as control variables. Lastly, numerous case studies demonstrate the effectiveness of the proposed approach, including both the low and high penetration level of renewables. Additional Monte Carlo simulations of consecutive 365 days show that the competitive ratio of the proposed approach is excellently below 1.10.

eess.SY↗

arXiv2020

Beyond Application End-Point Results: Quantifying Statistical Robustness of MCMC Accelerators

Statistical machine learning often uses probabilistic algorithms, such as Markov Chain Monte Carlo (MCMC), to solve a wide range of problems. Probabilistic computations, often considered too slow on conventional processors, can be accelerated with specialized hardware by exploiting parallelism and optimizing the design using various approximation techniques. Current methodologies for evaluating correctness of probabilistic accelerators are often incomplete, mostly focusing only on end-point result quality ("accuracy"). It is important for hardware designers and domain experts to look beyond end-point "accuracy" and be aware of the hardware optimizations impact on other statistical properties. This work takes a first step towards defining metrics and a methodology for quantitatively evaluating correctness of probabilistic accelerators beyond end-point result quality. We propose three pillars of statistical robustness: 1) sampling quality, 2) convergence diagnostic, and 3) goodness of fit. We apply our framework to a representative MCMC accelerator and surface design issues that cannot be exposed using only application end-point result quality. Applying the framework to guide design space exploration shows that statistical robustness comparable to floating-point software can be achieved by slightly increasing the bit representation, without floating-point hardware requirements.

eess.SP↗