arXiv subjects

Zehui Jiang

Publications and source records attributed to Zehui Jiang.

3 recordsLinked to original sources

arXiv2024

Neuron Empirical Gradient: Discovering and Quantifying Neurons Global Linear Controllability

While feed-forward neurons in pre-trained language models (PLMs) can encode knowledge, past research targeted a small subset of neurons that heavily influence outputs. This leaves the broader role of neuron activations unclear, limiting progress in areas like knowledge editing. We uncover a global linear relationship between neuron activations and outputs using neuron interventions on a knowledge probing dataset. The gradient of this linear relationship, which we call the neuron empirical gradient (NEG), captures how changes in activations affect predictions. To compute NEG efficiently, we propose NeurGrad, enabling large-scale analysis of neuron behavior in PLMs. We also show that NEG effectively captures language skills across diverse prompts through skill neuron probing. Experiments on MCEval8k, a multi-genre multiple-choice knowledge benchmark, support NEG's ability to represent model knowledge. Further analysis highlights the key properties of NEG-based skill representation: efficiency, robustness, flexibility, and interdependency. The code and data are released.

cs.CL↗

arXiv2018

Anisotropic Characteristic Lengths of Colloidal Monolayers Near a Water-Air Interface

Near-interface colloidal monolayers have often been used as model systems for research on hydrodynamics in biophysics and microfluidic systems. Using optical microscopy and multiparticle tracking techniques, the correlated diffusion of particles is experimentally measured in colloidal monolayers near a water-air interface. It is found that the characteristic lengths X1 and X2 of such a colloidal monolayer are anisotropic in these two perpendicular directions. The former (X1)is equal to the Saffman length of the monolayer and reflects the continuous nature of the system in the longitudinal direction. The latter (X2)is a function of both the Saffman length and the radius of the colloids and reflects the discrete nature of the system in the transverse direction. This discovery demonstrates that the hydrodynamics intrinsically follow different rules in these two directions in this system.

cond-mat.soft↗

arXiv2018

Predicting County Level Corn Yields Using Deep Long Short Term Memory Models

Corn yield prediction is beneficial as it provides valuable information about production and prices prior the harvest. Publicly available high-quality corn yield prediction can help address emergent information asymmetry problems and in doing so improve price efficiency in futures markets. This paper is the first to employ Long Short-Term Memory (LSTM), a special form of Recurrent Neural Network (RNN) method to predict corn yields. A cross sectional time series of county-level corn yield and hourly weather data made the sample space large enough to use deep learning technics. LSTM is efficient in time series prediction with complex inner relations, which makes it suitable for this task. The empirical results from county level data in Iowa show promising predictive power relative to existing survey based methods.

stat.ML↗