arXiv subjects

Jia Liang

Publications and source records attributed to Jia Liang.

9 recordsLinked to original sources

arXiv2026

Do Latent-CoT Models Think Step-by-Step? A Mechanistic Study on Sequential Reasoning Tasks

Latent Chain-of-Thought (Latent-CoT) aims to enable step-by-step computation without emitting long rationales, yet its mechanisms remain unclear. We study CODI, a continuous-thought teacher-student distillation model, on strictly sequential polynomial-iteration tasks. Using logit-lens decoding, linear probes, attention analysis, and activation patching, we localize intermediate-state representations and trace their routing to the final readout. On two- and three-hop tasks, CODI forms the full set of bridge states that become decodable across latent-thought positions, while the final input follows a separate near-direct route; predictions arise via late fusion at the end-of-thought boundary. For longer hop lengths, CODI does not reliably execute a full latent rollout, instead exhibiting a partial latent reasoning path that concentrates on late intermediates and fuses them with the last input at the answer readout position. Ablations show that this partial pathway can collapse under regime shifts, including harder optimization. Overall, we delineate when CODI-style latent-CoT yields faithful iterative computation versus compressed or shortcut strategies, and highlight challenges in designing robust latent-CoT objectives for sequential reasoning.

cs.AI↗

arXiv2025

Coincident Learning for Beam-based RF Station Fault Identification Using Phase Information at the SLAC Linac Coherent Light Source

Anomalies in radio-frequency (RF) stations can result in unplanned downtime and performance degradation in linear accelerators such as SLAC's Linac Coherent Light Source (LCLS). Detecting these anomalies is challenging due to the complexity of accelerator systems, high data volume, and scarcity of labeled fault data. Prior work identified faults using beam-based detection, combining RF amplitude and beam-position monitor data. Due to the simplicity of the RF amplitude data, classical methods are sufficient to identify faults, but the recall is constrained by the low-frequency and asynchronous characteristics of the data. In this work, we leverage high-frequency, time-synchronous RF phase data to enhance anomaly detection in the LCLS accelerator. Due to the complexity of phase data, classical methods fail, and we instead train deep neural networks within the Coincident Anomaly Detection (CoAD) framework. We find that applying CoAD to phase data detects nearly three times as many anomalies as when applied to amplitude data, while achieving broader coverage across RF stations. Furthermore, the rich structure of phase data enables us to cluster anomalies into distinct physical categories. Through the integration of auxiliary system status bits, we link clusters to specific fault signatures, providing additional granularity for uncovering the root cause of faults. We also investigate interpretability via Shapley values, confirming that the learned models focus on the most informative regions of the data and providing insight for cases where the model makes mistakes. This work demonstrates that phase-based anomaly detection for RF stations improves both diagnostic coverage and root cause analysis in accelerator systems and that deep neural networks are essential for effective analysis.

physics.acc-ph↗

arXiv2024

Acquire Precise and Comparable Fundus Image Quality Score: FTHNet and FQS Dataset

The retinal fundus images are utilized extensively in the diagnosis, and their quality can directly affect the diagnosis results. However, due to the insufficient dataset and algorithm application, current fundus image quality assessment (FIQA) methods are not powerful enough to meet ophthalmologists` demands. In this paper, we address the limitations of datasets and algorithms in FIQA. First, we establish a new FIQA dataset, Fundus Quality Score(FQS), which includes 2246 fundus images with two labels: a continuous Mean Opinion Score varying from 0 to 100 and a three-level quality label. Then, we propose a FIQA Transformer-based Hypernetwork (FTHNet) to solve these tasks with regression results rather than classification results in conventional FIQA works. The FTHNet is optimized for the FIQA tasks with extensive experiments. Results on our FQS dataset show that the FTHNet can give quality scores for fundus images with PLCC of 0.9423 and SRCC of 0.9488, significantly outperforming other methods with fewer parameters and less computation complexity.We successfully build a dataset and model addressing the problems of current FIQA methods. Furthermore, the model deployment experiments demonstrate its potential in automatic medical image quality control. All experiments are carried out with 10-fold cross-validation to ensure the significance of the results.

eess.IV↗

arXiv2024

Advancing Information Integration through Empirical Likelihood: Selective Reviews and a New Idea

Information integration plays a pivotal role in biomedical studies by facilitating the combination and analysis of independent datasets from multiple studies, thereby uncovering valuable insights that might otherwise remain obscured due to the limited sample size in individual studies. However, sharing raw data from independent studies presents significant challenges, primarily due to the need to safeguard sensitive participant information and the cumbersome paperwork involved in data sharing. In this article, we first provide a selective review of recent methodological developments in information integration via empirical likelihood, wherein only summary information is required, rather than the raw data. Following this, we introduce a new insight and a potentially promising framework that could broaden the application of information integration across a wider spectrum. Furthermore, this new framework offers computational convenience compared to classic empirical likelihood-based methods. We provide numerical evaluations to assess its performance and discuss various extensions in the end.

stat.ME↗

arXiv2024

Integrative data analysis where partial covariates have complex non-linear effects by using summary information from an external data

A full parametric and linear specification may be insufficient to capture complicated patterns in studies exploring complex features, such as those investigating age-related changes in brain functional abilities. Alternatively, a partially linear model (PLM) consisting of both parametric and non-parametric elements may have a better fit. This model has been widely applied in economics, environmental science, and biomedical studies. In this paper, we introduce a novel statistical inference framework that equips PLM with high estimation efficiency by effectively synthesizing summary information from external data into the main analysis. Such an integrative scheme is versatile in assimilating various types of reduced models from the external study. The proposed method is shown to be theoretically valid and numerically convenient, and it ensures a high-efficiency gain compared to classic methods in PLM. Our method is further validated using two data applications by evaluating the risk factors of brain imaging measures and blood pressure.

stat.ME↗

arXiv2023

AdaSelection: Accelerating Deep Learning Training through Data Subsampling

In this paper, we introduce AdaSelection, an adaptive sub-sampling method to identify the most informative sub-samples within each minibatch to speed up the training of large-scale deep learning models without sacrificing model performance. Our method is able to flexibly combines an arbitrary number of baseline sub-sampling methods incorporating the method-level importance and intra-method sample-level importance at each iteration. The standard practice of ad-hoc sampling often leads to continuous training with vast amounts of data from production environments. To improve the selection of data instances during forward and backward passes, we propose recording a constant amount of information per instance from these passes. We demonstrate the effectiveness of our method by testing it across various types of inputs and tasks, including the classification tasks on both image and language datasets, as well as regression tasks. Compared with industry-standard baselines, AdaSelection consistently displays superior performance.

cs.LG↗

arXiv2021

Linear Predictive Coding for Acute Stress Prediction from Computer Mouse Movements

Prior work demonstrated the potential of using the Linear Predictive Coding (LPC) filter to approximate muscle stiffness and damping from computer mouse movements to predict acute stress levels of users. Theoretically, muscle stiffness and damping in the arm can be estimated using a mass-spring-damper (MSD) biomechanical model. However, the damping frequency (i.e., stiffness) and damping ratio values derived using LPC were not yet compared with those from a theoretical MSD model. This work demonstrates that the damping frequency and damping ratio from LPC are significantly correlated with those from an MSD model, thus confirming the validity of using LPC to infer muscle stiffness and damping. We also compare the stress level binary classification performance using the values from LPC and MSD with each other and with neural network-based baselines. We found comparable performance across all conditions demonstrating LPC and MSD model-based stress prediction efficacy, especially for longer mouse trajectories. Clinical relevance: This work demonstrates the validity of the LPC filter to approximate muscle stiffness and damping and predict acute stress from computer mouse movements.

eess.SP↗

arXiv2020

Bone Suppression on Chest Radiographs With Adversarial Learning

Dual-energy (DE) chest radiography provides the capability of selectively imaging two clinically relevant materials, namely soft tissues, and osseous structures, to better characterize a wide variety of thoracic pathology and potentially improve diagnosis in posteroanterior (PA) chest radiographs. However, DE imaging requires specialized hardware and a higher radiation dose than conventional radiography, and motion artifacts sometimes happen due to involuntary patient motion. In this work, we learn the mapping between conventional radiographs and bone suppressed radiographs. Specifically, we propose to utilize two variations of generative adversarial networks (GANs) for image-to-image translation between conventional and bone suppressed radiographs obtained by DE imaging technique. We compare the effectiveness of training with patient-wisely paired and unpaired radiographs. Experiments show both training strategies yield "radio-realistic'' radiographs with suppressed bony structures and few motion artifacts on a hold-out test set. While training with paired images yields slightly better performance than that of unpaired images when measuring with two objective image quality metrics, namely Structural Similarity Index (SSIM) and Peak Signal-to-Noise Ratio (PSNR), training with unpaired images demonstrates better generalization ability on unseen anteroposterior (AP) radiographs than paired training.

eess.IV↗

arXiv2017

Relating Complexity-theoretic Parameters with SAT Solver Performance

Over the years complexity theorists have proposed many structural parameters to explain the surprising efficiency of conflict-driven clause-learning (CDCL) SAT solvers on a wide variety of large industrial Boolean instances. While some of these parameters have been studied empirically, until now there has not been a unified comparative study of their explanatory power on a comprehensive benchmark. We correct this state of affairs by conducting a large-scale empirical evaluation of CDCL SAT solver performance on nearly 7000 industrial and crafted formulas against several structural parameters such as backdoors, treewidth, backbones, and community structure. Our study led us to several results. First, we show that while such parameters only weakly correlate with CDCL solving time, certain combinations of them yield much better regression models. Second, we show how some parameters can be used as a "lens" to better understand the efficiency of different solving heuristics. Finally, we propose a new complexity-theoretic parameter, which we call learning-sensitive with restarts (LSR) backdoors, that extends the notion of learning-sensitive (LS) backdoors to incorporate restarts and discuss algorithms to compute them. We mathematically prove that for certain class of instances minimal LSR-backdoors are exponentially smaller than minimal-LS backdoors.

cs.AI↗