arXiv subjects

Shen Qu

Publications and source records attributed to Shen Qu.

3 recordsLinked to original sources

arXiv2026

HELIX: Hybrid Encoding with Learnable Identity and Cross-dimensional Synthesis for Time Series Imputation

Time series imputation benefits from leveraging cross-feature correlations, yet existing attention-based methods re-discover feature relationships at each layer, lacking persistent anchors to maintain consistent representations. To address this, we propose HELIX, which assigns each feature a learnable feature identity, a persistent embedding that captures intrinsic semantic properties throughout the network. Unlike graph-based methods that rely on predefined topology and assume homogeneous spatial relationships, HELIX learns arbitrary feature dependencies end-to-end from temporal co-variation, naturally handling datasets where features mix spatial locations with semantic variables. Integrated with hybrid temporal-feature attention, HELIX achieves the state-of-the-art performance, surpassing all 16 baselines on 5 public datasets across 21 experimental settings in our evaluation. Furthermore, our mechanistic analysis reveals that HELIX aligns learned feature identities and dependencies with latent physical and semantic structure progressively across layers, demonstrating that it more effectively translates cross-feature structure into imputation accuracy.

cs.LG↗

arXiv2025

Collaborative Lossless LLM Inference Serving with Offloading-based Pipeline Parallelism on Edge Devices

Providing lossless inference services of LLMs on edge devices remains challenging, especially given the extremely tight memory budgets. The existing offloading techniques inevitably introduce numerous loading bubbles, which further inflate the end-to-end latency of the entire inference pipeline. Meanwhile, dynamically fluctuating network bandwidth and diverse user request patterns pose additional obstacles to efficient lossless inference on edge devices. To address this, we propose LOIP, a collaborative lossless LLM inference system that employs an offloading-based interleaved pipeline parallelism to better overlap model offloading with computing and communicating. Specifically, LOIP first constructs an offloading-aware cost model to characterize inference latency and memory overhead under heterogeneous device capabilities and limited bandwidth. Based on this cost model, LOIP develops a fine-grained allocation scheduler that determines latency-efficient layer partitions across devices while explicitly accounting for offloading overhead, along with a unified memory architecture (UMA)-aware loading optimization using customized CUDA operators to reduce runtime loading overhead. LOIP further designs an online memory adaptation strategy to handle the increasing KV cache pressure and dynamic bandwidth fluctuations during inference. We implement LOIP with 2500+ lines of Python and 500+ lines of C++/CUDA code, and deploy it on five heterogeneous NVIDIA Jetson edge devices for lossless collaborative inference of LLaMA3.3-70B-Instruct. Extensive experiments demonstrate that LOIP achieves 8.8$\times$$\sim$20.3$\times$ speedups over the SOTA baselines under different bandwidth conditions and request patterns without compromising model accuracy.

cs.DC↗

arXiv2019

A Review on Energy, Environmental, and Sustainability Implications of Connected and Automated Vehicles

Connected and automated vehicles (CAVs) are poised to reshape transportation and mobility by replacing humans as the driver and service provider. While the primary stated motivation for vehicle automation is to improve safety and convenience of road mobility, this transformation also provides a valuable opportunity to improve vehicle energy efficiency and reduce emissions in the transportation sector. Progress in vehicle efficiency and functionality, however, does not necessarily translate to net positive environmental outcomes. Here we examine the interactions between CAV technology and the environment at four levels of increasing complexity: vehicle, transportation system, urban system, and society. We find that environmental impacts come from CAV-facilitated transformations at all four levels, rather than from CAV technology directly. We anticipate net positive environmental impacts at the vehicle, transportation system, and urban system levels, but expect greater vehicle utilization and shifts in travel patterns at the society level to offset some of these benefits. Focusing on the vehicle-level improvements associated with CAV technology is likely to yield excessively optimistic estimates of environmental benefits. Future research and policy efforts should strive to clarify the extent and possible synergetic effects from a systems level in order to envisage and address concerns regarding the short- and long-term sustainable adoption of CAV technology.

cs.CY↗