arXiv subjects

Jingshan Li

Publications and source records attributed to Jingshan Li.

3 recordsLinked to original sources

arXiv2026

iScript: A Domain-Adapted Large Language Model and Benchmark for Physical Design Tcl Script Generation

Modern EDA flows rely heavily on Tcl scripting, yet general LLMs perform poorly in this domain due to extreme data scarcity, domain-specific semantics, and the high reliability required in physical design. We present iScript, a domain-adapted Qwen3-8B model for Innovus Tcl script generation, and iScript-Bench, a comprehensive benchmark covering five task categories and three difficulty levels. To overcome the lack of training data, we introduce a multi-stage data synthesis pipeline that integrates command extraction, static linting, requirement back-inference, and Chain-of-Thought generation, producing a 10K-tuple (requirement, CoT, script) dataset. iScript is trained through a two-stage strategy combining domain-adaptive pretraining and supervised fine-tuning. To evaluate script correctness efficiently, we further propose a two-step verification framework consisting of static syntax verification and LLM-based functional evaluation. On our benchmark, iScript shows higher pass@k scores than currently state-of-the-art LLMs on average. These results demonstrate the effectiveness of domain adaptation and data synthesis for EDA scripting tasks.

cs.SE↗

arXiv2021

An Odor Labeling Convolutional Encoder-Decoder for Odor Sensing in Machine Olfaction

Deep learning methods have been widely applied to visual and acoustic technology. In this paper, we proposed an odor labeling convolutional encoder-decoder (OLCE) for odor identification in machine olfaction. OLCE composes a convolutional neural network encoder and decoder where the encoder output is constrained to odor labels. An electronic nose was used for the data collection of gas responses followed by a normative experimental procedure. Several evaluation indexes were calculated to evaluate the algorithm effectiveness: accuracy 92.57%, precision 92.29%, recall rate 92.06%, F1-Score 91.96%, and Kappa coefficient 90.76%. We also compared the model with some algorithms used in machine olfaction. The comparison result demonstrated that OLCE had the best performance among these algorithms. In the paper, some perspectives of machine olfactions have been also discussed.

cs.LG↗

arXiv2013

Growth Patterns of US Children from 1963 to 2012

Anthropometric measurements such as weight, stature (height), and body mass index (BMI) provide reliable indicators of children's growth. The 2000 CDC growth charts are the national standards in the United States for these important measures. But these growth charts were generated using data from 1963-1994. To understand the growth patterns of US children since 1994, we generate weight-for-age, stature-for-age and BMI-for-age percentile curves for both boys and girls aged 2-20 through the methods used to generate the 2000 CDC growth charts. Our datasets are from the National Health and Nutrition Examination Survey (NHANES) for years 1999-2010 and and from NorthShore University HealthSystem's Enterprise Data Warehouse (NS-EDW) for years 2006-2012. The weight and BMI percentile curves generated from NS-EDW and NHANES data differ substantially from the CDC percentile curves, while those for stature do not differ substantially. We conclude that the population weight and BMI values of US children in recent years have increased significantly since 2000 and the 2000 CDC growth charts may no longer be applicable to the current population of US children. Our charts poignantly reveals the increasing obesity of American children.

stat.AP↗