BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Yunpeng Wang

Yunpeng Wang contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Artificial IntelligenceMachine Learningphysics.app-phastro-ph.GAComputer Visioncond-mat.mes-hallcond-mat.mtrl-scieess.IVphysics.comp-phphysics.flu-dynRobotics

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
7works
0followers
11topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

7 published item(s)

preprint2026arXiv

RETO: A Rotary-Enhanced Transformer Operator for High-Fidelity Prediction of Automotive Aerodynamics

Rapid aerodynamic evaluation is crucial for modern vehicle design, yet existing neural operators struggle to capture intricate spatial correlations. We propose the rotary-enhanced transformer operator (RETO), a novel neural solver featuring a dual-stage spatial awareness mechanism: sinusoidal-cosine encodings for global referencing and rotary positional encodings (RoPE) for relative displacements. RoPE encodes spatial relations via unitary rotations, enforcing translation invariance and enhancing local gradient resolution. RETO is validated on ShapeNet and the high-fidelity DrivAerML benchmark. On ShapeNet, RETO achieves a relative $L_2$ error of 0.063, outperforming RegDGCNN at 0.125 and representing a 16\% improvement over the Transolver baseline, which yields an error of 0.075. These performance gains are further amplified on the DrivAerML dataset, where RETO achieves relative $L_2$ errors of 0.089 for surface pressure and 0.097 for velocity. In comparison, Transolver results in errors of 0.116 and 0.121 for the same metrics, indicating that RETO achieves precision enhancements of 23\% and 19\%, respectively. For comprehensive comparison, the surface pressure and velocity errors for AB-UBT are 0.102 and 0.124, while RegDGCNN yields 0.235 and 0.312, respectively. Information-theoretical analysis shows that the entropy peak of RETO at 0.35 is significantly lower than that of Transolver at 0.75 under $10^4$ resolution, indicating a focused attentional mechanism capable of preserving localized gradients against global diffusion.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Example Perplexity

Some examples are easier for humans to classify than others. The same should be true for deep neural networks (DNNs). We use the term example perplexity to refer to the level of difficulty of classifying an example. In this paper, we propose a method to measure the perplexity of an example and investigate what factors contribute to high example perplexity. The related codes and resources are available at https://github.com/vaynexie/Example-Perplexity.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Optical Properties of C$-$rich ($^{12}$C, SiC and FeC) Dust Layered Structure of Massive Stars

The composition and structure of interstellar dust are important and complex for the study of the evolution of stars and the \textbf{interstellar medium} (ISM). However, there is a lack of corresponding experimental data and model theories. By theoretical calculations based on ab-initio method, we have predicted and geometry optimized the structures of Carbon-rich (C-rich) dusts, carbon ($^{12}$C), iron carbide (FeC), silicon carbide (SiC), even silicon ($^{28}$Si), iron ($^{56}$Fe), and investigated the optical absorption coefficients and emission coefficients of these materials in 0D (zero$-$dimensional), 1D, and 2D nanostructures. Comparing the \textbf{nebular spectra} of the supernovae (SN) with the coefficient of dust, we find that the optical absorption coefficient of the 2D $^{12}$C, $^{28}$Si, $^{56}$Fe, SiC and FeC structure corresponds to the absorption peak displayed in the infrared band (5$-$8) $μ$$m$ of the spectrum at 7554 days after the SN1987A explosion. And it also corresponds to the spectrum of 535 days after the explosion of SN2018bsz, when the wavelength in the range of (0.2$-$0.8) and (3$-$10) $μ$$m$. Nevertheless, 2D SiC and FeC corresponds to the spectrum of 844 days after the explosion of SN2010jl, when the wavelength is within (0.08$-$10) $μ$$m$. Therefore, FeC and SiC may be the second type of dust in SN1987A corresponding to infrared band (5$-$8) $μ$$m$ of dust and may be in the ejecta of SN2010jl and SN2018bsz.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Temporally sparse data assimilation for the small-scale reconstruction of turbulence

Previous works have shown that the small-scale information of incompressible homogeneous isotropic turbulence (HIT) is fully recoverable as long as sufficient large-scale structures are continuously enforced through temporally continuous data assimilation (TCDA). In the current work, we show that the assimilation time step can be relaxed to values about 1 $\sim$ 2 orders larger than that for TCDA, using a temporally sparse data assimilation (TSDA) strategy, while the accuracy is still maintained or even slightly better in the presence of non-negligible large-scale errors. The one-step data assimilation (ODA) is examined to unravel the mechanism of TSDA. It is shown that the relaxation effect for errors above the assimilation wavenumber $k_a$ is responsible for the error decay in ODA. Meanwhile, The errors contained in the large scales can propagate into small scales and make the high-wavenumber ($k>k_a$) error noise decay slower with TCDA than TSDA. This mechanism is further confirmed by incorporating different levels of errors in the large scales of the reference flow field. The advantage of TSDA is found to grow with the magnitude of the incorporated errors. Thus, it is potentially more beneficial to adopt TSDA if the reference data contains non-negligible errors. Finally, an outstanding issue raised in previous works regarding the possibility of recovering the dynamics of sub-Kolmogorov scales using direct numerical simulation (DNS) data at Kolmogorov scale resolution is also discussed.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

TrajGen: Generating Realistic and Diverse Trajectories with Reactive and Feasible Agent Behaviors for Autonomous Driving

Realistic and diverse simulation scenarios with reactive and feasible agent behaviors can be used for validation and verification of self-driving system performance without relying on expensive and time-consuming real-world testing. Existing simulators rely on heuristic-based behavior models for background vehicles, which cannot capture the complex interactive behaviors in real-world scenarios. To bridge the gap between simulation and the real world, we propose TrajGen, a two-stage trajectory generation framework, which can capture more realistic behaviors directly from human demonstration. In particular, TrajGen consists of the multi-modal trajectory prediction stage and the reinforcement learning based trajectory modification stage. In the first stage, we propose a novel auxiliary RouteLoss for the trajectory prediction model to generate multi-modal diverse trajectories in the drivable area. In the second stage, reinforcement learning is used to track the predicted trajectories while avoiding collisions, which can improve the feasibility of generated trajectories. In addition, we develop a data-driven simulator I-Sim that can be used to train reinforcement learning models in parallel based on naturalistic driving data. The vehicle model in I-Sim can guarantee that the generated trajectories by TrajGen satisfy vehicle kinematic constraints. Finally, we give comprehensive metrics to evaluate generated trajectories for simulation scenarios, which shows that TrajGen outperforms either trajectory prediction or inverse reinforcement learning in terms of fidelity, reactivity, feasibility, and diversity.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Microscopic observation of carrier-transport dynamics in quantum-structure solar cells using a time-of-flight technique

In this study, we propose a carrier time-of-flight technique to evaluate the carrier transport time across a quantum structure in an active region of solar cells. By observing the time-resolved photoluminescence signal with a quantum-well probe inserted under the quantum structure at forward bias, the carrier transport time can be efficiently determined at room temperature. The averaged drift velocity shows linear dependence on the internal field, allowing us to estimate the quantum structure as a quasi-bulk material with low effective mobility containing the information of carrier dynamics. We show that this direct and real-time observation is more sensitive to carrier transport than other conventional techniques, providing better insights into microscopic carrier transport dynamics to overcome a device design difficulty.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Prediction of tubular solar still performance by machine learning integrated with Bayesian optimization algorithm

Presented is a new generation prediction model of a tubular solar still (TSS) productivity utilizing two machine learning (ML) techniques, namely:Random forest (RF) and Artificial neural network (ANN). Prediction models were conducted based on experimental data recorded under Egyptian climate. Meteorological and operational thermal parameters were utilized as input layers. Moreover, Bayesian optimization algorithm (BOA) was used to obtain the optimal performance of RF and ANN models. In addition, these models results were compared to those of a multilinear regression (MLR) model. As resulted, experimentally, the average value accumulated productivity was 4.3 L/(m2day). For models results, RF was less sensitive to hyper parameters than ANN as ANN performance could be significantly improved by BOA more than RF. In addition, RF achieved better prediction performance of TSS on the current dataset. The determination coefficients (R2) of RF and ANN were 0.9964 and 0.9977, respectively, which were much higher than MLR models, 0.9431. Based on the robustness performance and high accuracy, RF is recommended as a stable method for predicting the productivity of TSS.

0 citations0 reviewsNo code linkedOpen access