archive
Every paper Pith has read. Search by title, abstract, or pith.
506 papers in cs.CE · page 1
-
Language models reconstruct flow fields from under 10% data
Operator Learning for Reconstructing Flow Fields from Sparse Measurements: a Language Model Approach
-
VLM agents output editable figures after 17 edits
LiveFigure: Generating Editable Scientific Illustration with VLM Agents
-
Static contexts make individual dynamics identifiable from single snapshots
Learning Individual Dynamics from Sparse Cross-Sectional Snapshots
-
Reward alignment lifts cyclic peptide closure rates by 20 points
GeoCycler: Reward-Aligned 3D Diffusion for Constraint-Conditioned Cyclic Peptide Design
-
Hawkes-AR-Gumbel model corrects 40% CVaR underestimation at 99.995%
Bayesian Extreme Value Theory with Hawkes-AR-Gumbel Dependence for Extreme CVaR Estimation in Operational Risk
-
Complex DeFi yield strategies add risk without higher returns
DeFi Yield Aggregators: Analysing Investment Strategies and Structural Dependencies
-
Proxy from similar deposits estimates drilling value without ground truth
Effective information gathering for ore estimation, evaluation and perspectives on adaptive sampling
-
Toolkit automates annotation of child-caregiver eye-tracking videos
GazeBehavior Annotation Toolkit (GBAT): AI-powered toolkit for automatic annotation of egocentric eye-tracking and video data of child-caregiver interaction
-
Pretrained model cuts 3D chip thermal error 10x with 20% data
Therm-FM: Foundation Model is ALL YOU NEED for 3D-ICs Thermal Simulation
-
Hypergraphs built from time series without prior structure
Dynamic Hypergraph Representation Learning for Multivariate Time Series without Prior Knowledge
-
Hybrid model cuts NEM electricity price forecast errors by 12%
Hybrid Kolmogorov-Arnold Network and XGBoost Framework for Week-Ahead Price Forecasting in Australia's National Electricity Market
-
Ancestral lineage priors guide family-valid protein generation
LineageFlow: Flow Matching for High-Fidelity Family-Aware Protein Sequence Generation
-
Ancestral priors let flow matching match natural protein family validity
LineageFlow: Flow Matching for High-Fidelity Family-Aware Protein Sequence Generation
-
Sig-Graph GAN better replicates stock log returns with graphs
A Generative Adversarial Graph Neural Network for Synthetic Time Series Data
-
Physics-informed model recovers aerodynamic loads from noisy bridge data
Aerodynamic force reconstruction using physics-informed Gaussian processes
-
SDF method generates post-stent models in 1.5s matching target sizes
SDFStent: Real-time interactive virtual stenting via SDF deformation fields
-
Market maker adapts to new regimes without retraining
Zero-shot adaptation to order book dynamics
-
CNNs classify six PD source types under switching voltage at 96% accuracy
Classification of Single and Mixed Partial Discharges under Switching Voltage Using an AWA-CNN Framework
-
Fortran scientific codes harbor many undefined-behavior-like defects
RSE of a Quantum Transport Code and its Effects
-
GNNs add significant accuracy to financial time series forecasts
The Statistical Significance of the Inclusion of Graph Neural Networks in the Financial Time Series Forecasting Problem
-
Kernel sum of squares speeds Bayesian acquisition search
KSOS-BO: Improving Sampling in Bayesian Optimization via Kernel Sum of Squares
-
Simulated traffic ranks road network datasets via Wasserstein distance
Distance between Road Networks: A Macroscopic Method for Road Network Datasets Comparison Using Traffic-weighted Geographic Distribution
-
Stress shadows control fracture production more than fluid choice
HyFrac.fun: A 3D Hydraulic Fracturing Simulator on Cloud
-
Entry month explains up to 79% of crypto return variance
Modern Portfolio Theory in the Crypto-Wilderness
-
Learned perturbations turn ML interatomic potentials probabilistic
Uncertainty-aware Machine Learning Interatomic Potentials via Learned Functional Perturbations
-
Domain decomposition extends recursive Green's functions to parallel n-diagonal use
Revisiting recursive methods for Dyson and Keldysh in NEGF: Part I
-
Neural optimization designs gel-elastomer composites for programmed morphing
GELATO: Multi-Material Topology Optimization of Programmable Gel-Elastomer Structures
-
RefineGPT lifts consistency of automated refinery diagrams
RefiningGPT: Specialized language Models for Automated Refinery Unit-level Process Diagram Synthesis
-
Dual KANs speed structural and fluid topology optimization
A Dual Physics-Informed Kolmogorov-Arnold Neural Network Framework for Continuum Topology Optimization
-
Finite element model matches theory for wall sound transmission
Building Acoustics 01: Finite Element Model of an Building Acoustics Test Facility to Predict the Sound Transmission Loss Based on DIN EN ISO 10140
-
Protein Thoughts ranks true binders at mean position 11.2
Protein Thoughts: Interpretable Reasoning with Tree of Thoughts and Embedding-Space Flow Matching for Protein-Protein Interaction Discovery
-
Collateral trigger makes hybrid AMM the Nash choice in duopoly
Bounding LVR in AMMs via Secant-Tangent Divergence and Collateralized Liquidity Scaling
-
Amortizing options let DeFi share tail risk without central clearing
Designing On-Chain Options: Amortizing Perpetual Options
-
Mixed bandit strategy taps data-center flexibility without internal schedules
Robust Restless Multi-Armed Bandit for Data Center Flexibility Services Through Virtual Machine Scheduling
-
LLM proposals yield 25 feasible states in stress topology tests
IterSIMP-{\sigma}: Evaluating LLM-Assisted Spatial Interventions in Stress-Aware Topology Optimization
-
Python tool computes minimal repairs for Boolean network models
pyModRev: a Python Tool for Model Revision of Boolean Networks
-
Water dosage shifts produce fixable shape errors in concrete 3D prints
Quantifying water-driven geometric uncertainties in powder bed concrete printing using high-resolution 3D modeling
-
SRM algorithm scales to 10 million particles for composites
Efficient generation of large-scale non-equilibrium distributions of particles
-
Buffer parameters let ML predict signal integrity across IC technologies
Buffer-Parameterized Machine Learning Surrogate Models for Cross-Technology Signal Integrity Analysis and Optimization
-
Age contacts speed urban epidemics while distance barely matters
Epidemics in a Synthetic Urban Population with Multiple Levels of Mixing
-
No multimodal model tops 65 on financial document reasoning benchmark
FinDocMRE: A Benchmark for Document-Level Financial Multimodal Reasoning Evaluation
-
50% equity cap keeps pension decumulation nearly optimal
Numerical methods for optimal decumulation of a defined contribution pension plan
-
Bayesian-Monte Carlo method refines construction schedule forecasts
Bayesian-Monte Carlo Schedule Updating for Construction Digital Twins: A Probabilistic Framework for Dynamic Project Forecasting
-
Weight tying across dilations yields scale-equivariant time-series generators
Scale-Equivariant Generative Forecasting: Weight-Tied Dilated Convolutions, Wavelet Scattering Inputs, and Spectral-Consistency Training for Self-Similar Time Series
-
Hybrid optimizer improves spatial packaging by over 10 percent
A Hybrid Optimization Framework for Spatial Packaging of Interconnected Systems
-
Optimizer packs components into non-convex custom shapes
Spatial Optimization of Interconnected Systems in Non-Convex Design Spaces
-
Reweighted flow matching stabilizes digital-twin estimates
Weighted Flow Matching and Physics-Informed Nonlinear Filtering for Parameter Estimation in Digital Twins
-
Federated models detect PV fraud by fusing irradiance data locally
Privacy-Preserving Generation Fraud Detection for Distributed Photovoltaic Systems: A Solar Irradiance-Fused Federated Learning Framework
-
LLM trading alpha is not deployment evidence
The Alpha Illusion: Reported Alpha from LLM Trading Agents Should Not Be Treated as Deployment Evidence
-
Emulator finds 181 maize traits that hold up under future climate
From Simulation to Discovery: AI Enabled Probabilistic Emulation of Mechanistic Crop Systems