arXiv subjects

Haoan Wang

Publications and source records attributed to Haoan Wang.

7 recordsLinked to original sources

arXiv2026

Demand Transfer Estimation at Scale via Restricted Logit Modeling

Item demand forecasting is an integral component of store assortment optimization. Existing literature focuses on learning a suitable customer choice model and using this model to determine the value of an objective function (i.e. expected demand) with respect to an assortment proposal. However, for large item universe with many categories, this approach can prove inefficient, needing a separate demand forecast for every possible item assortment. An alternate approach exists whereby we combine the efficiency of forecasting item demand independently, while at the same time applying adjustments to the independent forecasts that account for the relations between item demand and the availability of other similar items on the shelf. Central to this approach is the estimation of Demand Transfer (DT) coefficients. These DT coefficients represent the percent of a particular target item's (item that the customer walked in the store to buy) demand that is redirected to each other item in the universe should the target item be removed from the shelf. We introduce an approach that allows us to compute these DT coefficients on large item universes (assortments having 1 million+ items). Experiments on data as well as historical transaction data for multiple locations within categories demonstrate that when certain reasonable assumptions about substitution behavior are satisfied, our procedure is able to accurately estimate underlying DT coefficients and lead to improvements in demand forecasting.

cs.LG↗

arXiv2026

Beyond Static RAG: An Adaptive, Tri-Metric Routing Framework for Efficient Long-Context Inference on Commodity GPUs

Deploying retrieval-augmented generation (RAG) on commodity GPUs such as the NVIDIA T4 (16 GB VRAM) exposes a practical failure mode we call the Compression Paradox: neural prompt compression can add key-value (KV) cache contention and preprocessing latency that outweigh generation-time savings, while skipping compression can cause out-of-memory (OOM) failures on long contexts. We identify two distinct failure mechanisms when a vLLM-served LLM and a PyTorch-based compressor are co-deployed under tight memory budgets, and introduce the Tri-Metric Router, a deterministic, training-free policy that selects among Raw, Neural (LLMLingua-2), and Lexical (BM25) pipelines. The router uses three CPU-side signals: spatial complexity ($L$), syntactic density ($ρ_{key}$), and type-token ratio (TTR). Unlike prior semantic-only adaptation, our dispatch signal is hardware-physical, based on VRAM headroom and a latency crossover point. Thresholds are calibrated from profiling on LongBench qasper, yielding an operating crossover near 4,332 words on T4; our contribution is this calibration methodology rather than a hardware-specific constant. On out-of-distribution holdouts, the method achieves 0% OOM failures, 88.5 $\pm$ 4.4% oracle alignment, and 49.3% Combined F1, improving over always-on lexical compression by 5.2 points without additional VRAM or training cost.

cs.LG↗

arXiv2026

LEAP: Layer-wise Exit-Aware Pretraining for Efficient Transformer Inference

Layer-aligned distillation and convergence-based early exit represent two predominant computational efficiency paradigms for transformer inference; yet we establish that they exhibit systematic incompatibility under standard deployment conditions for convergence-based early exit. Distillation objectives that align intermediate student layers to teacher representations suppress the representational convergence that early-exit mechanisms exploit, rendering such mechanisms ineffective on distilled models. We introduce LEAP (Layer-wise Exit-Aware Pretraining), an auxiliary training objective that reconciles this incompatibility. LEAP requires no architectural modifications; it augments standard distillation with a single constraint ensuring intermediate layers approximate final-layer representations. LEAP-MiniLM achieves 1.61$\times$ measured wall-clock speedup (batch=1, NVIDIA L4) at $θ$=0.95, with 91.9% of samples exiting by layer 7 and 1.80$\times$ theoretical layer reduction, where standard distilled models achieve zero effective speedup. We validate across sentence similarity (STS-B: 0.760 $\pm$ 0.006) and retrieval benchmarks (BEIR), providing operational guidance including latency measurements, decision thresholds, and deployment criteria.

cs.LG↗

arXiv2024

Neural Spatial-Temporal Tensor Representation for Infrared Small Target Detection

Optimization-based approaches dominate infrared small target detection as they leverage infrared imagery's intrinsic low-rankness and sparsity. While effective for single-frame images, they struggle with dynamic changes in multi-frame scenarios as traditional spatial-temporal representations often fail to adapt. To address these challenges, we introduce a Neural-represented Spatial-Temporal Tensor (NeurSTT) model. This framework employs nonlinear networks to enhance spatial-temporal feature correlations in background approximation, thereby supporting target detection in an unsupervised manner. Specifically, we employ neural layers to approximate sequential backgrounds within a low-rank informed deep scheme. A neural three-dimensional total variation is developed to refine background smoothness while reducing static target-like clusters in sequences. Traditional sparsity constraints are incorporated into the loss functions to preserve potential targets. By replacing complex solvers with a deep updating strategy, NeurSTT simplifies the optimization process in a domain-awareness way. Visual and numerical results across various datasets demonstrate that our method outperforms detection challenges. Notably, it has 16.6$\times$ fewer parameters and averaged 19.19\% higher in $IoU$ compared to the suboptimal method on $256 \times 256$ sequences.

cs.CV↗

arXiv2023

Use of Robust DOB/CDOB Compensation to Improve Autonomous Vehicle Path Following Performance in the Presence of Model Uncertainty, CAN Bus Delays and External Disturbances

A path tracking control system is chosen as the proof-of-concept demonstration application in this paper. A disturbance observer (DOB) is embedded within the steering to path error automated driving loop to handle uncertain parameters such as vehicle mass, vehicle velocities and road friction coefficient and to reject yaw moment disturbances. The compensation of vehicle model with the embedded disturbance observer forces it to behave like its nominal model within the bandwidth of the disturbance observer. A parameter space approach based steering controller is then used to optimize performance. The proposed method demonstrates good disturbance rejection and achieves stability robustness. The variable time delay from the steer-by-wire system in an actual vehicle can also lead to stability issues since it adds large negative phase angle to the plant frequency response and tends to destabilize it. A communication disturbance observer (CDOB) based time delay compensation approach that does not require exact knowledge of this time delay is embedded into the steering actuation loop to handle this problem. Stability analysis of both DOB and CDOB compensation system are presented in this paper. Extensive model-in-the-loop simulations were performed to test the designed disturbance observer and CDOB systems and show reduced path following errors in the presence of uncertainty, disturbances and time delay. A validated model of our 2017 Ford Fusion Hybrid research autonomous vehicle is used in the simulation analyses. Simulation results verify the performance enhancement of the vehicle path following control with proposed DOB and CDOB structure. A HiL simulator that uses a validated CarSim model with sensors and traffic will be used later to verify the real time capability of our approach.

eess.SY↗

arXiv2023

Discrete-time Robust PD Controlled System with DOB/CDOB Compensation for High Speed Autonomous Vehicle Path Following

Autonomous vehicle path following performance is one of significant consideration. This paper presents discrete time design of robust PD controlled system with disturbance observer (DOB) and communication disturbance observer (CDOB) compensation to enhance autonomous vehicle path following performance. Although always implemented on digital devices, DOB and CDOB structure are usually designed in continuous time in the literature and also in our previous work. However, it requires high sampling rate for continuous-time design block diagram to automatically convert to corresponding discrete-time controller using rapid controller prototyping systems. In this paper, direct discrete time design is carried out. Digital PD feedback controller is designed based on the nominal plant using the proposed parameter space approach. Zero order hold method is applied to discretize the nominal plant, DOB and CDOB structure in continuous domain. Discrete time DOB is embedded into the steering to path following error loop for model regulation in the presence of uncertainty in vehicle parameters such as vehicle mass, vehicle speed and road-tire friction coefficient and rejecting external disturbance like crosswind force. On the other hand, time delay from CAN bus based sensor and actuator command interfaces results in degradation of system performance since large negative phase angles are added to the plant frequency response. Discrete time CDOB compensated control system can be used for time delay compensation where the accurate knowledge of delay time value is not necessary. A validated model of our lab Ford Fusion hybrid automated driving research vehicle is used for the simulation analysis while the vehicle is driving at high speed. Simulation results successfully demonstrate the improvement of autonomous vehicle path following performance with the proposed discrete time DOB and CDOB structure.

cs.RO↗

arXiv2021

Multi-Objective Digital PID Controller Design in Parameter Space

This paper presents a multi-objective digital PID controller design method using the parameter space approach of robust control. Absolute stability is treated first by finding the digital PID controller gain parameter space corresponding to closed loop poles being inside the unit circle. Phase margin, gain margin and a mixed sensitivity bound are treated as frequency domain constraints. Determination of digital PID controller parameter space regions satisfying these constraints is presented. All of these regions are superimposed to obtain a multi-objective digital PID controller gain parameter space solution region. The path following controller design of an automated driving vehicle is used as an example to illustrate the method.

eess.SY↗