arXiv subjects

Yuna Lee

Publications and source records attributed to Yuna Lee.

6 recordsLinked to original sources

arXiv2026

Kinematics and Dynamics of the Open Cluster NGC 2302

Open clusters are ideal observational testbeds to understand the dynamics of stellar systems. We present a dynamical study of the young open cluster NGC 2302. The latest Gaia data and $UBVIJHK_s$ photometric data are used in this study. A total of 117 stars are selected as the genuine members using the Gaia data. This cluster is, on average, reddened by $ = 0.24 \pm 0.06$ (s.d.). The ratio of total-to-selective extinction ($R_V$) in the direction of NGC 2302 is $2.8 \pm 0.1$. The cluster distance is determined to be $1.16 \pm 0.08$ kpc using Gaia parallaxes. Theoretical isochrone fitting for $Z = 0.008$ on color-magnitude diagrams yields an age of $80 \pm 20$ Myr. The relative proper motions of individual members show no significant radial expansion or contraction. NGC 2302 contains a total stellar mass of $333 \pm 48 M_{\odot}$. The one-dimensional velocity dispersion is approximately 0.26 km s$^{-1}$, which is comparable to the viral velocity dispersion of 0.27 km s$^{-1}$ derived from its total mass. Its relaxation time is estimated to be approximately 90 Myr, which is similar to the age of the cluster within the uncertainty in age estimation. Finally, we report a pattern of mass segregation in the radial distribution of stellar masses. Our results suggest that NGC 2302 is virialized and currently approaching a state of dynamical relaxation. However, because no definitive evidence of kinetic energy equipartition is found, the possibility of the in-situ formation of high-mass stars within the central region should be carefully considered.

astro-ph.GA↗

arXiv2026

ERASE: Eliminating Redundant Visual Tokens via Adaptive Two-Stage Token Pruning

Recent advancements in Vision-Language Models (VLMs) enable large language models (LLMs) to process high-resolution images, significantly improving real-world multimodal understanding. However, this capability introduces a large number of vision tokens, resulting in substantial computational overhead. To mitigate this issue, various vision token pruning methods have been proposed. Nevertheless, existing approaches predominantly rely on learned semantic features within the model to capture visual redundancy. Moreover, they lack adaptive mechanisms to adjust pruning strategies according to the complexity of the input image. In this paper, we propose ERASE, a two-stage vision token pruning framework that identifies and retains salient tokens through pruning strategies adaptive to image complexity. Experiment results demonstrate that ERASE significantly reduces vision tokens while preserving accuracy. For Qwen2.5-VL-7B, at a token pruning ratio of 85\%, ERASE retains 89.46% of the original model accuracy, whereas the best prior method retains only 78.1%. Our code is available at https://github.com/Tuna-Luna/ERASE.

cs.CV↗

arXiv2026

AnalogToBi: Device-Level Analog Circuit Topology Generation via Bipartite Graph and Grammar Guided Decoding

Analog circuit design remains highly dependent on expert knowledge due to the complexity of device-level interactions and topology design. Recent transformer-based approaches for device-level topology generation have shown promise, yet they suffer from low electrical validity without human-in-the-loop (HITL) training and severe memorization caused by sequence-based circuit representations. In this work, we propose AnalogToBi, a framework for device-level analog circuit topology generation. AnalogToBi introduces circuit-type conditioning for categorizing heterogeneous multi-type topology datasets, device renaming augmentation to mitigate memorization, a bipartite graph representation for improved structural generalization, and grammar-guided decoding to enforce structural validity during bipartite graph generation. Experimental results demonstrate that AnalogToBi achieves high validity and novelty without HITL training while effectively avoiding memorization of training topologies. Our code is available at https://github.com/Seungmin0825/AnalogToBi.

cs.AR↗

arXiv2025

See, Point, Fly: A Learning-Free VLM Framework for Universal Unmanned Aerial Navigation

We present See, Point, Fly (SPF), a training-free aerial vision-and-language navigation (AVLN) framework built atop vision-language models (VLMs). SPF is capable of navigating to any goal based on any type of free-form instructions in any kind of environment. In contrast to existing VLM-based approaches that treat action prediction as a text generation task, our key insight is to consider action prediction for AVLN as a 2D spatial grounding task. SPF harnesses VLMs to decompose vague language instructions into iterative annotation of 2D waypoints on the input image. Along with the predicted traveling distance, SPF transforms predicted 2D waypoints into 3D displacement vectors as action commands for UAVs. Moreover, SPF also adaptively adjusts the traveling distance to facilitate more efficient navigation. Notably, SPF performs navigation in a closed-loop control manner, enabling UAVs to follow dynamic targets in dynamic environments. SPF sets a new state of the art in DRL simulation benchmark, outperforming the previous best method by an absolute margin of 63%. In extensive real-world evaluations, SPF outperforms strong baselines by a large margin. We also conduct comprehensive ablation studies to highlight the effectiveness of our design choice. Lastly, SPF shows remarkable generalization to different VLMs. Project page: https://spf-web.pages.dev

cs.RO↗

arXiv2025

Formation process of young stellar population in Messier 16 from a kinematic perspective

We present a kinematic study of young stars in Messier 16 (M16) using the Gaia Data Release 3 and high-resolution spectra. A total of 345 stars are selected as genuine members using the published lists of X-ray, infrared sources, and early-type stars as well as the Gaia data. There is severe differential reddening across this region and the reddening law of the intracluster medium appears abnormal. The distance to M16, derived from the parallaxes of the members, is about 1.7 kpc. The ages of members, estimated by comparing their color-magnitude diagram with theoretical isochrones, range from 1 Myr to 4 Myr. This star-forming region is composed of an open cluster (NGC 6611) and a distributed population. This cluster shows a clear pattern of expansion and rotation. Some of the distributed population are spatially associated with the gas pillars located at the ridge of H II bubble. In particular, several stars moving away from the cluster are physically associated with the northeastern pillar. In addition, their younger ages support the idea that the formation of these stars was triggered by the feedback from massive stars in NGC 6611. On the other hand, the other stars do not show systematic radial or stream motions; therefore, they likely formed through spontaneous star formation events. We discuss the formation of young stars in the context of cluster expansion, spontaneous star formation, and feedback-driven star formation, and suggest that all of these mechanisms possibly contributed to their formation.

astro-ph.SR↗

arXiv2025

TTS-CtrlNet: Time varying emotion aligned text-to-speech generation with ControlNet

Recent advances in text-to-speech (TTS) have enabled natural speech synthesis, but fine-grained, time-varying emotion control remains challenging. Existing methods often allow only utterance-level control and require full model fine-tuning with a large emotion speech dataset, which can degrade performance. Inspired by adding conditional control to the existing model in ControlNet (Zhang et al, 2023), we propose the first ControlNet-based approach for controllable flow-matching TTS (TTS-CtrlNet), which freezes the original model and introduces a trainable copy of it to process additional conditions. We show that TTS-CtrlNet can boost the pretrained large TTS model by adding intuitive, scalable, and time-varying emotion control while inheriting the ability of the original model (e.g., zero-shot voice cloning & naturalness). Furthermore, we provide practical recipes for adding emotion control: 1) optimal architecture design choice with block analysis, 2) emotion-specific flow step, and 3) flexible control scale. Experiments show that ours can effectively add an emotion controller to existing TTS, and achieves state-of-the-art performance with emotion similarity scores: Emo-SIM and Aro-Val SIM. The project page is available at: https://curryjung.github.io/ttsctrlnet_project_page

cs.SD↗