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

Youngwook Kim

Youngwook Kim contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Artificial IntelligenceMachine LearningComputation and LanguageComputer Visioncond-mat.mes-hallcond-mat.mtrl-scicond-mat.str-elphysics.ins-detquant-ph

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
8works
0followers
9topics
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

8 published item(s)

preprint2026arXiv

FreeTimeGS++: Secrets of Dynamic Gaussian Splatting and Their Principles

The recent surge in 4D Gaussian Splatting (4DGS) has achieved impressive dynamic scene reconstruction. While these methods demonstrate remarkable performance, the specific drivers behind such gains remain less explored, making a systematic understanding of the underlying principles challenging. In this paper, we perform a comprehensive analysis of these hidden factors to provide a clearer perspective on the 4DGS framework. We first establish a controlled baseline, FreeTimeGS_ours, by formalizing and reproducing the heuristics of the state-of-the-art FreeTimeGS. Using this framework, we dissect 4DGS along its fundamental axes and uncover key secrets, including the emergent temporal partitioning driven by Gaussian durations and the discrepancy between photometric fidelity and spatiotemporal consistency. Based on these insights, we propose FreeTimeGS++, a principled method that employs gated marginalization and neural velocity fields to achieve superior stability and robust dynamic representations. Our approach yields reproducible results with reduced run-to-run variance. We will release our implementation to provide a reliable foundation for future 4DGS research.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Dual Task Framework for Improving Persona-grounded Dialogue Dataset

This paper introduces a simple yet effective data-centric approach for the task of improving persona-conditioned dialogue agents. Prior model-centric approaches unquestioningly depend on the raw crowdsourced benchmark datasets such as Persona-Chat. In contrast, we aim to fix annotation artifacts in benchmarking, which is orthogonally applicable to any dialogue model. Specifically, we augment relevant personas to improve dialogue dataset/agent, by leveraging the primal-dual structure of the two tasks, predicting dialogue responses and personas based on each other. Experiments on Persona-Chat show that our approach outperforms pre-trained LMs by an 11.7 point gain in terms of accuracy.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Large Loss Matters in Weakly Supervised Multi-Label Classification

Weakly supervised multi-label classification (WSML) task, which is to learn a multi-label classification using partially observed labels per image, is becoming increasingly important due to its huge annotation cost. In this work, we first regard unobserved labels as negative labels, casting the WSML task into noisy multi-label classification. From this point of view, we empirically observe that memorization effect, which was first discovered in a noisy multi-class setting, also occurs in a multi-label setting. That is, the model first learns the representation of clean labels, and then starts memorizing noisy labels. Based on this finding, we propose novel methods for WSML which reject or correct the large loss samples to prevent model from memorizing the noisy label. Without heavy and complex components, our proposed methods outperform previous state-of-the-art WSML methods on several partial label settings including Pascal VOC 2012, MS COCO, NUSWIDE, CUB, and OpenImages V3 datasets. Various analysis also show that our methodology actually works well, validating that treating large loss properly matters in a weakly supervised multi-label classification. Our code is available at https://github.com/snucml/LargeLossMatters.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Modularized Transfer Learning with Multiple Knowledge Graphs for Zero-shot Commonsense Reasoning

Commonsense reasoning systems should be able to generalize to diverse reasoning cases. However, most state-of-the-art approaches depend on expensive data annotations and overfit to a specific benchmark without learning how to perform general semantic reasoning. To overcome these drawbacks, zero-shot QA systems have shown promise as a robust learning scheme by transforming a commonsense knowledge graph (KG) into synthetic QA-form samples for model training. Considering the increasing type of different commonsense KGs, this paper aims to extend the zero-shot transfer learning scenario into multiple-source settings, where different KGs can be utilized synergetically. Towards this goal, we propose to mitigate the loss of knowledge from the interference among the different knowledge sources, by developing a modular variant of the knowledge aggregation as a new zero-shot commonsense reasoning framework. Results on five commonsense reasoning benchmarks demonstrate the efficacy of our framework, improving the performance with multiple KGs.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Preparation of large Cu3Sn single crystal by Czochralski method

Cu3Sn was recently predicted to host topological Dirac fermions, but related research is still in its infancy. The growth of large and high-quality Cu3Sn single crystals is, therefore, highly desired to investigate the possible topological properties. In this work, we report the single crystal growth of Cu3Sn by Czochralski (CZ) method. Crystal structure, chemical composition, and transport properties of Cu3Sn single crystals were analyzed to verify the crystal quality. Notably, compared to the mm-sized crystals from a molten Sn-flux, the cm-sized crystals obtained by the CZ method are free from contamination from flux materials, paving the way for the follow-up works.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

TrustAL: Trustworthy Active Learning using Knowledge Distillation

Active learning can be defined as iterations of data labeling, model training, and data acquisition, until sufficient labels are acquired. A traditional view of data acquisition is that, through iterations, knowledge from human labels and models is implicitly distilled to monotonically increase the accuracy and label consistency. Under this assumption, the most recently trained model is a good surrogate for the current labeled data, from which data acquisition is requested based on uncertainty/diversity. Our contribution is debunking this myth and proposing a new objective for distillation. First, we found example forgetting, which indicates the loss of knowledge learned across iterations. Second, for this reason, the last model is no longer the best teacher -- For mitigating such forgotten knowledge, we select one of its predecessor models as a teacher, by our proposed notion of "consistency". We show that this novel distillation is distinctive in the following three aspects; First, consistency ensures to avoid forgetting labels. Second, consistency improves both uncertainty/diversity of labeled data. Lastly, consistency redeems defective labels produced by human annotators.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Rectification by hydrodynamic flow in an encapsulated graphene Tesla valve

Systems in which interparticle interactions prevail can be described by hydrodynamics. This regime is typically difficult to access in the solid state for electrons. However, the high purity of encapsulated graphene combined with its advantageous phonon properties make it possible, and hydrodynamic corrections to the conductivity of graphene have been observed. Examples include electron whirlpools, enhanced flow through constrictions as well as a Poiseuille flow profile. An electronic device relying specifically on viscous behaviour and acting as a viscometer has however been lacking. Here, we implement the analogue of the Tesla valve. It exhibits nonreciprocal transport and can be regarded as an electronic viscous diode. Rectification occurs at carrier densities and temperatures consistent with the hydrodynamic regime, and disappears both in the ballistic and diffusive transport regimes. In a device in which the electrons are exposed to a Moiré superlattice, the Lifshitz transition when crossing the Van Hove singularity is observed in the rectifying behaviour.

0 citations0 reviewsNo code linkedOpen access
preprint2019arXiv

Nanoscale imaging of equilibrium quantum Hall edge currents and of the magnetic monopole response in graphene

The recently predicted topological magnetoelectric effect and the response to an electric charge that mimics an induced mirror magnetic monopole are fundamental attributes of topological states of matter with broken time reversal symmetry. Using a SQUID-on-tip, acting simultaneously as a tunable scanning electric charge and as ultrasensitive nanoscale magnetometer, we induce and directly image the microscopic currents generating the magnetic monopole response in a graphene quantum Hall electron system. We find a rich and complex nonlinear behavior governed by coexistence of topological and nontopological equilibrium currents that is not captured by the monopole models. Furthermore, by utilizing a tuning fork that induces nanoscale vibrations of the SQUID-on-tip, we directly image the equilibrium currents of individual quantum Hall edge states for the first time. We reveal that the edge states that are commonly assumed to carry only a chiral downstream current, in fact carry a pair of counterpropagating currents, in which the topological downstream current in the incompressible region is always counterbalanced by heretofore unobserved nontopological upstream current flowing in the adjacent compressible region. The intricate patterns of the counterpropagating equilibrium-state orbital currents provide new insights into the microscopic origins of the topological and nontopological charge and energy flow in quantum Hall systems.

0 citations0 reviewsNo code linkedOpen access