GIST Scholar
Not a member yet
30271 research outputs found
Sort by
Electron Acceleration in Ionizing Gas by Self-Divergent Amplitude-Modulated Laser Pulses
No full textWe investigate electron acceleration driven by a self diverging laser filament propagating through a tunnel-ionizing gas. When a high-intensity, amplitude-modulated laser interacts with the gas, it ionizes the medium and generates plasma. Due to local plasma density gradients, the laser filament diverges as it propagates, altering the dynamics of electron acceleration. Electrons can gain energy from the rising edge of the laser pulse, while the subsequent divergence of the laser helps mitigate deceleration by reducing the overlap between the trailing fields and the electron trajectory. Consequently, a significant portion of the acquired energy can be retained. To optimize the acceleration process, the spatial alignment between the laser pulse peak and the electron position is crucial. We develop a theoretical model to describe the divergence behavior of the laser filament and its impact on electron dynamics. Using this model, we estimate the achievable electron energy gain and derive scaling laws to guide the optimization and control of the acceleration process. This work suggests a novel mechanism for enhancing electron energy gain through the interaction of amplitude-modulated laser filaments with gas jets. © 1973-2012 IEEE.sciescopu
Thermochromic nanoreflectors for battery safety
No full textWe present thermochromic nanophotonic reflectors as a rapid, real-time visual sensing platform for temperature monitoring on battery surfaces. These devices consist of a nanoscale tellurium (Te) layer sandwiched between a metallic mirror and a transparent protective layer. Using the solid-to-quasi-liquid phase transition of Te below 80 °C, the reflectors exhibit reversible, sub-second optical changes in response to thermal stimuli. When exposed to open flame, mimicking thermal runaway scenarios, the reflectors effectively track heat propagation across the battery surface. Furthermore, we demonstrate practical applicability by visualizing temperature dynamics during standard charge–discharge cycles of commercial lithium-ion batteries and smartphones. This work highlights the potential of Te-based thermochromic nanophotonic reflectors as compact, wireless, and intuitive thermal sensors for enhancing battery safety in real-world electronic systems
Input-Bounded Displacement-Style Formation Control
No full textThis article introduces input-bounded displacement-style formation control methods for multiagent systems modeled by single- and double-integrator dynamics with input saturation. Displacement-based formation control, which relies on relative positions between agents, is extended by incorporating orientation estimation. The proposed control laws ensure that the agents achieve the desired formations while maintaining bounded inputs to avoid input saturation. Under the condition of a strongly connected and detailed balanced graph, we prove that the multiagent systems globally asymptotically converge to the desired formation up to the translation and rotation with bounded input when the orientation estimations reach consensus. The simulation results of the formation control have been provided to validate the theoretical results.FALSEsciescopu
Pulse Compression in a Gradient-Density Plasma Combined with Plasma Gratings
No full textBy combining grating structures with a density gradient, plasma-based optical pulse compression is made more feasible by considerably lowering the density compared to using only a gradient and by compensating for density fluctuations. © 2025 Optica
Exploring Planning Capability of Large Language Model Using Abstraction and Reasoning Corpus Benchmark Woochang Sim Gwangju Institute of Science and Technology
No full text최근 대규모 언어 모델은 거대한 파라미터에 많은 데이터와 GPU 자원을 기반으로 학습하여 다양한 벤치마크에서 좋은 성능을 발휘하고 있다. 이러한 대규모 언어 모델이 더 발전하기 위해서는 처음 직면한 문제에 대해서 패턴을 파악하여 문제 해결을 위한 목표를 설정하고 이를 해결할 전략을 수립할 줄 알아야 한다. 즉, 대규모 언어 모델이 한 걸음 더 나아가기 위해서는 계획을 세우는 능력이 필요하다. 본 논문에서는 계획 능력을 목표 설정과 전략 수립 등의 2가지 하위 능력의 집합으로 정의하였다. 본 논문에서는 크게 2가지 종류의 실험을 진행하였다. 하나의 실험은 대규모 언어 모델의 계획 능력을 확인하는 실험이다. 즉, 대규모 언어 모델의 목표 설정 능력과 전략 수립 능력을 평가하 는 실험이다. 다른 하나는 계획 능력을 평가하기 최적의 실험을 할 수 있도록 프롬프트 셋팅을 찾는 실험을 진행하였다. 실험 결과를 통해 현재 대규모 언어 모델의 계획 능력 은 다른 벤치마크에 비해 부족하다는 것을 알 수 있었다. Feedback loop나 메타 개념의 계층화 등의 방법을 사용한다면 계획 능력 더 향상될 것으로 보인다.|Large language models show strong performance on benchmarks, trained on vast data with enormous parameters using substantial GPU resources. For these models to advance further, they must be able to identify patterns, set objectives when en- countering problems, and establish solving strategies. In short, large language models need planning abilities for advancement. In this paper, planning ability is defined as a set of two sub-capabilities: objective setting and strategy formulation. The paper conducted two main experiments. One verified the planning ability of large language models by evaluating their objective setting and strategy formulation. The other fo- cused on finding optimal prompt settings to evaluate planning abilities. Results showed that current LLMs’ planning abilities are insufficient compared to other benchmarks. Planning abilities could improve through feedback loops and hierarchical meta-concept organization.MasterAbstract (English) i
Abstract (Korean) ii
List of Contents iii
List of Tables v
List of Figures vii
1 Introduction 1
2 Background 4
2.1 Planning Capability 4
2.2 Benchmark 6
2.2.1 Introduction to the ARC Benchmark 6
2.2.2 DSLs in ARC 8
2.2.3 Suitability of ARC for Evaluating LLMs Planning Capabilities . 9
2.3 Learning Methods 10
2.4 Prompt Methods 14
2.4.1 Chain of Thought 15
2.4.2 Least to Most 15
2.4.3 Tree of Thoughts 16
3 Preliminary Experiments 19
3.1 Optimizing Prompt Configuration for LLMs in ARC 19
3.1.1 Optimizing Grid Representation for ARC 20
3.1.2 Comparative Analysis of Prompting Methods for LLMs in ARC 26
3.2 Assessing the Importance of Planning Capability 33
3.2.1 Motivation 33
3.2.2 Method 34
3.2.3 Experiment and Result 35
3.2.4 Analysis 36
3.2.5 Summary 37
– iii –
4 Evaluating Planning Capability of LLMs 38
4.1 Evaluating Objective Setting 38
4.1.1 Motivation 38
4.1.2 Method 39
4.1.3 Experiment and Result 41
4.1.4 Analysis 41
4.1.5 Summary 42
4.2 Evaluating Strategy Formulation 43
4.2.1 Motivation 43
4.2.2 Method 44
4.2.3 Experiment and Result 46
4.2.4 Analysis 46
4.2.5 Summary 49
5 Discussion 50
5.1 Is the Planning Capability of LLMs Really Poor 50
5.2 How Can the Planning Capability of LLMs Be Improved 52
6 Conclusions 55
Summary 56
References 57
A Detailed Methodology for Experiments with Constraint Alleviation 64
A.1 Measuring Human Performance in Strategic Formulation 64
– iv
Enrichment of gut-derived metabolites in a Parkinson's disease subtype with REM sleep behavior disorder
No full textRecent reports indicate that Parkinson's disease (PD)-like changes in gut microbial signatures develop in idiopathic REM sleep behavior disorder (iRBD), a prodrome of alpha-synucleinopathies. However, <50% of PD cases exhibit RBD at onset, underscoring PD's heterogeneity. Using untargeted metabolomics, plasma samples from PD patients with and without RBD, iRBD subjects, and healthy controls were analyzed to characterize the metabolic differences in PD patients with and without RBD. Metabolomic analysis revealed the enrichment of gut microbial-origin metabolites, such as secondary bile acids and p-cresol sulfate, in PD patients with RBD, while metabolites linked to neuropsychiatric diseases were elevated in PD patients without RBD. Additionally, our prediction that p-cresol sulfate, enriched in the RBD groups with a gut microbial origin, crosses the blood-brain barrier, suggests a potential mechanistic link between gut microbial dysbiosis, iRBD, and PD with RBD. Our study elucidates the heterogeneous nature of PD subtypes.TRUEsciescopu
Garlic mustard (Alliaria petiolata) Invasion Reshapes Soil Microbial Communities and Network Structures in South Korea
No full textGarlic mustard (Alliaria petiolata ) is a recently introduced invasive species in South Korea, yet its ecological impacts remain unclear. We assessed how garlic mustard invasion modifies soil microbial communities and chemical properties as part of potential plant-soil feedback mechanisms. Soil samples were collected from invaded and adjacent uninvaded sites within three regions―Samcheok, Suwon, and Incheon―where garlic mustard has been reported. We analyzed soil bacterial and fungal communities and measured water-soluble ion concentrations to evaluate associated biotic and abiotic changes. Microbial diversity was significantly altered by invasion across all sites, but functional shifts in microbial communities were observed only at the Samcheok sites. Specifically, the abundance of chemoheterotrophic bacteria increased, while phototrophic and nitrogen-fixing groups declined. Fungal communities in Samcheok also exhibited a decrease in ectomycorrhizal fungi and an increase in pathogenic fungi. Constrained ordination (db-RDA and CCA) revealed that Samcheok was the only region where microbial communities were strongly correlated with soil chemistry. pH emerged as a key driver of both bacterial and fungal communities, with NH4+ and NO3- further contributing to the differentiation between invaded and uninvaded bacterial communities. These patterns suggest that altered nitrogen availability and acidity may underlie the compositional and functional changes observed in this region. Based on the 100 most abundant taxa, network analysis consistently showed greater connectivity and density in bacterial communities of invaded soils across all regions, indicating enhanced structural stability. In contrast, the fungal network remained unchanged. These findings suggest that garlic mustard invasion alters microbial community structure and interactions, with consistent shifts in bacterial network structure and region-specific environmental associations, possibly reflecting differences in invasion stage.
Acknowledgements : This work was supported by the National Research Foundation of Korea grant funded by the Korea government (Ministry of Science and ICT) (No. NRF-RS-2024-00456189)
Improved Charge Balance in Quantum Dot Light-Emitting Diodes with Sputtered Li-doped NiO and a Dual-function Self-Assembled Monolayer
No full textAchieving high-performance quantum dot light-emitting diodes (QLEDs), which are promising candidates for next-generation displays, requires balanced charge injection in terms of electrons and holes while minimizing nonradiative recombination. In this study, to improve charge balance in QLEDs, we enhanced the electrical conductivity of the hole injection layer by depositing NiO:Li thin films via radio frequency sputtering under varying oxygen partial pressures. Furthermore, a self-assembled monolayer (SAM) made of 4-trifluoromethyl benzoic acid was introduced on the NiO:Li films to reduce the hole injection barrier and facilitate efficient hole transport. This SAM was also applied on top of the quantum dot layer to act as an electron blocking layer by increasing the electron injection barrier, thereby suppressing excessive electron injection. The optimized QLEDs exhibited a maximum external quantum efficiency of 9.41% and a peak luminance of 50,199 cd/m2. These findings demonstrate a viable strategy to enhance QLED performance via the synergistic modulation of charge injection and recombination dynamics.FALSEsciescopu
Arbitrary 3D Organic Mixed Ionic-Electronic Conductor Architectures via Self-Fusion of PEDOT:PSS Microfibers
No full textIn this research, a first-of-its-kind fabricating strategy is reported that assembles arbitrary 3D organic mixed ionic-electronic conductor (OMIEC) architectures using poly(3,4-ethylenedioxythiophene):polystyrene sulfonate (PEDOT:PSS) microfiber building blocks. This approach exploits a water-assisted self-fusion process, in which adhesion can be modulated as reversible (PSS-rich) or irreversible (PEDOT-rich) self-fusion depending on the post-treatment condition of building blocks. Phenomenological characterization and structural analyses reveal that hydration-induced swelling of hydrophilic PSS chains and crystalline pi-pi-stacked PEDOT domains govern interfacial bonding. Using PEDOT:PSS microfibers as modular units, structures ranging from 2D mesh electrodes to centimeter-scale free-standing 3D architectures are demonstrated. The resulting microfiber network structures are mechanically robust under bending and folding in aqueous environments and exhibit a high volumetric capacitance. Furthermore, hydration reduces the elastic modulus by approximate to 80%, enabling soft, conformal adhesion onto wet and irregular surfaces without additional adhesives. Finally, "cut-and-stick" PEDOT:PSS mesh electrodes are fabricated as a proof-of-concept and employed for recording in vivo cardiac activities from rodent hearts with minimal motion artifacts, outperforming conventional rigid platinum electrodes. This self-fusion strategy establishes a simple and scalable route for the first-time construction of arbitrary 3D OMIEC architectures, opening new opportunities for multifunctional OMIEC platforms in bioelectronics and energy-storage applications.TRUEsciescopu