Showing 1–50 of 1,190 results for author: Chung, S

$Υ(nS)$ Production within Jets at the LHC

Authors: Taewook Ha, Hee Sok Chung, Daekyoung Kang, Yunlu Wang, Haixiang Zhu

Abstract: Heavy quarkonium production inside jets offers a sensitive probe of QCD dynamics and bound-state formation mechanisms. While recent studies demonstrate that charmonium-in-jet observables effectively discriminate among competing nonrelativistic QCD (NRQCD) long-distance matrix element (LDME) sets, whether this discriminating power persists in the bottomonium sector remains an open question. Here, w… ▽ More Heavy quarkonium production inside jets offers a sensitive probe of QCD dynamics and bound-state formation mechanisms. While recent studies demonstrate that charmonium-in-jet observables effectively discriminate among competing nonrelativistic QCD (NRQCD) long-distance matrix element (LDME) sets, whether this discriminating power persists in the bottomonium sector remains an open question. Here, we present the first phenomenological study of $Υ(1S)$, $Υ(2S)$, and $Υ(3S)$ production inside jets using the fragmenting jet function (FJF) framework at next-to-leading order (NLO), incorporating DGLAP evolution, threshold resummation, and feeddown contributions from higher bottomonium states. In sharp contrast to charmonium, we find that bottomonium-in-jet momentum-fraction ($z_H$) distributions exhibit a universal shape that is remarkably insensitive to the choice of LDME sets. We show that this universality stems from the strong dominance of the S-wave spin-triplet color-octet ($^3S_1^{[8]}$) production mechanism reinforced by $χ_b$ feeddown transitions. Our predictions capture both the characteristic large-$z_H$ peak and the spectral broadening with increasing jet transverse momentum observed in recent CMS measurements. These results establish a clear physical distinction between charmonium and bottomonium fragmentation inside jets, providing a theoretical benchmark for future high-precision measurements at the LHC. △ Less

Submitted 19 August, 2026; originally announced August 2026.

Comments: 30 pages, 11 figures

UniVerse: Benchmarking and Enhancing LALMs on Culturally Inclusive Low-Resource Music Understanding

Authors: Ziya Zhou, Shangda Wu, Shenyang Xu, Yutong Zheng, Dafang Liang, Suin Chung, Danbinaerin Han, Junyan Jiang, Yongyi Zang, Ruibin Yuan, Rongxiu Zhong, Shilei Zhang, Junlan Feng, Jinglei Liu, Haotian Zhou, Zijin Li, Dasaem Jeong, Wei Xue, Yike Guo

Abstract: Recent advances in large audio-language models (LALMs) have significantly improved performance in tasks such as music captioning, genre classification, and sound event detection. However, limited attention has been paid to improving their adaptability across diverse musical traditions, particularly folk music rooted in distinct cultural contexts. Folk-music traditions are typically resource-scarce… ▽ More Recent advances in large audio-language models (LALMs) have significantly improved performance in tasks such as music captioning, genre classification, and sound event detection. However, limited attention has been paid to improving their adaptability across diverse musical traditions, particularly folk music rooted in distinct cultural contexts. Folk-music traditions are typically resource-scarce, unevenly represented across regions, and poorly documented. Even when such samples appear in large-scale pre-training, LALMs often fail to capture their structural and stylistic characteristics, partly due to the absence of dedicated evaluation protocols and training solutions. To address these limitations, we introduce UniVerse, a reproducible solution for low-resource music understanding. Specifically, we propose UniVerseBench, a benchmark of 5,042 Q&A pairs across more than 38 cultural and linguistic entities, constructed via an expert-guided yet highly automated pipeline. In parallel, we construct a fully automated, model-generated multi-turn dialogue training dataset UniVerseSet. By training LALMs on UniVerseSet, we systematically adapt and investigate representative multimodal imbalance learning strategies across both dense and Mixture-of-Experts (MoE) architectures. Experimental results indicate that fully automated data curation combined with imbalance-aware training yields non-trivial improvements, but models still struggle to capture fine-grained acoustic features, indicating a gap between surface-level alignment and deep musical comprehension. △ Less

Submitted 18 August, 2026; originally announced August 2026.

Comments: 21 pages, 7 figures, 8 tables

Pitch Contour Tokenization using VQ-VAE and Its Application on Korean Traditional Music Analysis

Authors: Seonguk Ju, Seola Cho, Sooin Chung, Danbinaerin Han, Dasaem Jeong

Abstract: Computational analysis of music often relies on discrete representations, yet many musical traditions are organized around continuous pitch movement that resists segmentation into note-like units. For such traditions, the discrete units that analysis would build on are not given in advance. We address this gap by learning a vocabulary of local pitch-contour patterns directly from unlabeled audio,… ▽ More Computational analysis of music often relies on discrete representations, yet many musical traditions are organized around continuous pitch movement that resists segmentation into note-like units. For such traditions, the discrete units that analysis would build on are not given in advance. We address this gap by learning a vocabulary of local pitch-contour patterns directly from unlabeled audio, using a VQ-VAE that quantizes fixed-length contour segments into a finite codebook. To make the learned tokens stable across segmentation positions and small variations in timing and pitch range, we train the model with a reconstruction objective evaluated under the best alignment among a set of candidate temporal and pitch-domain transformations. Applied to Korean traditional music, the learned tokens recover information about expert-defined sigimsae categories without supervision, and in pansori individual tokens align with the two principal modes, Gyemyeonjo and Ujo, supporting their use as units for corpus-level analysis of contour-centric traditions. △ Less

Submitted 11 August, 2026; originally announced August 2026.

Comments: 8 pages, 3 figures, 2 tables. Accepted at ISMIR 2026

Motif 3: Technical Report

Authors: Junghwan Lim, Joon Son Chung, Sungmin Lee, Wai Ting Cheung, Gihun Cho, Minsu Ha, Sangho Kang, Beomgyu Kim, Dongseok Kim, Jangwoong Kim, Taehyun Kim, Taewhan Kim, Jeesoo Lee, Jeongdoo Lee, Junhyeok Lee, Dongpin Oh, Hyeyeon Cho, Dahye Choi, Jaeheui Her, Hanbin Jung, Changjin Kang, Minjae Kim, Youngrok Kim, Hyukjin Kweon, Hongjoo Lee , et al. (2 additional authors not shown)

Abstract: We introduce Motif 3, a decoder-only Mixture-of-Experts language model with 314 billion total parameters and 13.2 billion activated per token. Each sparse MoE layer contains 384 routed experts, with eight selected per token. This fine-grained sparsity provides substantial expert capacity while limiting computation. Motif 3 is built around Grouped Differential Latent Attention (GDLA), which integra… ▽ More We introduce Motif 3, a decoder-only Mixture-of-Experts language model with 314 billion total parameters and 13.2 billion activated per token. Each sparse MoE layer contains 384 routed experts, with eight selected per token. This fine-grained sparsity provides substantial expert capacity while limiting computation. Motif 3 is built around Grouped Differential Latent Attention (GDLA), which integrates grouped differential attention with the compressed key-value representation of Multi-head Latent Attention. The architecture further incorporates modified manifold-constrained hyper-connections, Expert Specific PolyNorm activations, and multi-token prediction to improve optimization stability, expert specialization, and inference efficiency. We pretrain Motif 3 on approximately 12.5 trillion tokens spanning web documents, STEM, code, mathematics, multilingual content, and domain-specialized corpora. Expert-balancing and numerical-stabilization techniques support stable training at scale, while selective MXFP8 computation and communication, memory-efficient fused kernels, and window-aware context parallelism enable training with context lengths up to 256K tokens. Our post-training pipeline combines general supervised fine-tuning, six specialist teachers trained with reinforcement learning, a software-engineering teacher trained with supervised fine-tuning, and Multi-teacher On-Policy Distillation. The resulting unified model consolidates complementary capabilities in reasoning, coding, tool use, professional work, long-context understanding, calibrated abstention, and instruction following. Across a broad evaluation suite, Motif 3 demonstrates competitive performance against leading open weight models, including strong results on long-horizon agentic tasks, mathematical reasoning, scientific knowledge, and hallucination-sensitive evaluation. △ Less

Submitted 10 August, 2026; originally announced August 2026.

Frame-Level Pansori Mode Classification with Complementary Audio Representations

Authors: Sangheon Park, Seonguk Ju, Suin Chung, Danbinaerin Han, Dasaem Jeong

Abstract: Pansori is a traditional Korean vocal genre whose mode system (jo) is defined not by scale alone but by the entanglement of pitch collection, microtonal ornament (sigimsae), and vocal timbre. In this study, we introduce a 46-hour frame-level pansori mode annotation, expert-labeled across all five canonical batang, and evaluate four complementary input representations (mel spectrogram, F0 contour,… ▽ More Pansori is a traditional Korean vocal genre whose mode system (jo) is defined not by scale alone but by the entanglement of pitch collection, microtonal ornament (sigimsae), and vocal timbre. In this study, we introduce a 46-hour frame-level pansori mode annotation, expert-labeled across all five canonical batang, and evaluate four complementary input representations (mel spectrogram, F0 contour, MIDI piano roll, and a multi-cultural SSL encoder) under two split strategies designed to detect shortcut learning. Across the three well-represented modes, performance degrades by only 2.1--3.6 points of F1 when entire works are held out, indicating that the models learn mode-relevant features rather than memorizing repertoire. Per-class results further show that source separation removes the percussion cue on which changjo depends, and that generic multi-cultural pre-training fails specifically on the Ujo--Gyemyeonjo distinction. Qualitative analysis of cross-modal disagreement recovers musicologically documented phenomena and agrees with published score-based analyses of modern changjak pansori. △ Less

Submitted 6 August, 2026; originally announced August 2026.

Comments: Accepted to the 27th International Society for Music Information Retrieval (ISMIR) Conference, 2026

Journal ref: ISMIR 2026

Rethinking Personalized Reward Modeling for LLMs under Preference Heterogeneity via Group-Debiased Federated Learning

Authors: Seongyoon Kim, Boryeong Cho, Jihwan Oh, Seokhyun Chung, Se-Young Yun

Abstract: Large language models are increasingly aligned to human preferences via reward modeling, but user preference data are sensitive and often cannot be centralized. Federated learning keeps such data local while learning a shared initial reward model, which is later personalized for each client through local fine-tuning. Because users often assign opposite labels to the same pair of responses, existin… ▽ More Large language models are increasingly aligned to human preferences via reward modeling, but user preference data are sensitive and often cannot be centralized. Federated learning keeps such data local while learning a shared initial reward model, which is later personalized for each client through local fine-tuning. Because users often assign opposite labels to the same pair of responses, existing federated methods address preference heterogeneity by clustering similar clients and training one reward model per group, assuming that each group requires its own initialization. We show that this assumption is unnecessary. Under balanced preference groups, a single FedAvg model, despite starting at nearly random accuracy, surpasses reward models trained separately for each ground-truth group after only a few local optimization steps. We attribute this phenomenon to the flatness of the shared initialization: averaging across all clients learns richer shared representations that distinguish responses while canceling conflicting preference directions, leaving the model near a decision boundary that can be rapidly adapted. Group imbalance breaks this effect as the cancellation becomes asymmetric and leaves minority clients too far from the boundary to recover. Motivated by this observation, we propose FedGD (Federated Learning with Group Debiasing), which discovers latent preference groups during federated training and learns a single reward model using group-debiased client sampling. By counteracting the effect of group imbalance, FedGD learns an initialization that remains highly adaptable, enabling effective personalization without prior knowledge of the underlying groups. △ Less

Submitted 2 August, 2026; originally announced August 2026.

KMT-2025-BLG-0975Lb and KMT-2025-BLG-1160Lb: Two Uranus-Mass Planets Beyond the Snow Line Discovered by Microlensing

Authors: Cheongho Han, Chung-Uk Lee, Andrzej Udalski, Andrew Gould, Michael D. Albrow, Sun-Ju Chung, Youn Kil Jung, Kyu-Ha Hwang, Yoon-Hyun Ryu, Yossi Shvartzvald, In-Gu Shin, Jennifer C. Yee, Weicheng Zang, Hongjing Yang, Doeon Kim, Dong-Jin Kim, Byeong-Gon Park, Richard W. Pogge, Przemek Mróz, Michał K. Szymański, Jan Skowron, Radosław Poleskim Igor Soszyński, Paweł Pietrukowicz, Szymon Kozłowski, Krzysztof A. Rybicki , et al. (5 additional authors not shown)

Abstract: We present the analysis of two planetary microlensing events, KMT-2025-BLG-0975 and KMT-2025-BLG-1160, discovered during the 2025 Galactic bulge microlensing season through high-cadence survey observations. In both events, short-duration anomalies near the peaks of the lensing light curves reveal the presence of planetary companions. Light-curve modeling yields planet-to-host mass ratios of… ▽ More We present the analysis of two planetary microlensing events, KMT-2025-BLG-0975 and KMT-2025-BLG-1160, discovered during the 2025 Galactic bulge microlensing season through high-cadence survey observations. In both events, short-duration anomalies near the peaks of the lensing light curves reveal the presence of planetary companions. Light-curve modeling yields planet-to-host mass ratios of $q = 8.6 \times 10^{-4}$ for KMT-2025-BLG-0975 and $1.3 \times 10^{-4}$ for KMT-2025-BLG-1160. For KMT-2025-BLG-0975, finite-source effects are detected, enabling a measurement of the angular Einstein radius, whereas only a lower limit on this quantity is obtained for KMT-2025-BLG-1160. We estimate the physical parameters of the lens systems through Bayesian analyses constrained by the measured microlensing observables. The results indicate that the planetary companions have masses of $M_{\rm p}=29.8^{+50.5}_{-16.0}~M_\oplus$ for KMT-2025-BLG-0975Lb and $25.4^{+15.5}_{-14.1}~M_\oplus$ for KMT-2025-BLG-1160Lb. Both planets have masses comparable to that of Uranus. The host stars are inferred to be a low-mass M dwarf with a mass of $M_{\rm h}=0.10^{+0.18}_{-0.06}~M_\odot$ for KMT-2025-BLG-0975L and a late K dwarf with a mass of $M_{\rm h}=0.58^{+0.35}_{-0.32}~M_\odot$ for KMT-2025-BLG-1160L. The projected planet--host separations are $a_\perp=0.81^{+0.10}_{-0.11}$~au for KMT-2025-BLG-0975Lb and $a_\perp=2.56^{+0.48}_{-0.71}$~au and $3.29^{+0.61}_{-0.92}$~au for the inner and wide solutions, respectively, of KMT-2025-BLG-1160Lb. In both systems, the planets are located beyond the expected snow-line distances of their hosts, placing them in the cold ice-giant regime. △ Less

Submitted 28 July, 2026; originally announced July 2026.

Comments: 11 pages, 7 figures

Out of Sight, Still in Mind: Token Compression for Omni-LLMs

Authors: Suho Yoo, Youngjoon Jang, Hyebin Cho, Joon Son Chung

Abstract: The goal of this paper is to reduce the input token cost of Omni-modal large language models (Omni-LLMs) at inference time. Omni-LLMs reason jointly over audio, video and text, but the cost of the three streams is highly unbalanced: visual tokens account for the vast majority of the input, and are highly redundant. In this paper, we propose ReMo, a training-free framework that compresses visual to… ▽ More The goal of this paper is to reduce the input token cost of Omni-modal large language models (Omni-LLMs) at inference time. Omni-LLMs reason jointly over audio, video and text, but the cost of the three streams is highly unbalanced: visual tokens account for the vast majority of the input, and are highly redundant. In this paper, we propose ReMo, a training-free framework that compresses visual tokens by redistributing their information across modalities: a visual token is kept only if its information appears nowhere else. ReMo achieves this in two ways: (i) it aligns audio and video in a common embedding space, and removes visual tokens already explained by the audio or by other visual tokens; and (ii) it replaces object-level visual tokens with compact text proxies, short descriptions of each object and its location, conveying the same content in far fewer tokens. On Qwen2.5-Omni at two model scales, ReMo removes 54% of the input tokens with no loss in accuracy. Indeed, it slightly exceeds the full-token model, reaching 101.2% and 101.3% of its average accuracy over five audio-visual benchmarks. △ Less

Submitted 23 July, 2026; originally announced July 2026.

Comments: Preprint

First Observation of an Exotic Reggeon

Authors: G. D. Alexeev, M. G. Alexeev, C. Alice, A. Amoroso, V. Andrieux, V. Anosov, K. Augsten, W. Augustyniak, C. D. R. Azevedo, B. Badelek, R. Beck, J. Beckers, Y. Bedfer, V. Benesova, J. Bernhard, F. Bradamante, A. Bressan, W. -C. Chang, C. Chatterjee, M. Chiosso, S. -U. Chung, A. Cicuttin, M. L. Crespo, D. D'Ago, S. Dalla Torre , et al. (159 additional authors not shown)

Abstract: We present new \compass high-statistics measurements of the peripheral production of $ηπ^-$ and $η^\prime π^-$ pairs in the reactions $π^- p \to η^{(\prime)}π^- p$. For the first time, we perform an unbinned analysis of the high-mass region of the $ηπ^-$ and $η^\primeπ^-$ systems, which allows us to disentangle the exchange mechanisms governing their production. We report the first observation in… ▽ More We present new \compass high-statistics measurements of the peripheral production of $ηπ^-$ and $η^\prime π^-$ pairs in the reactions $π^- p \to η^{(\prime)}π^- p$. For the first time, we perform an unbinned analysis of the high-mass region of the $ηπ^-$ and $η^\primeπ^-$ systems, which allows us to disentangle the exchange mechanisms governing their production. We report the first observation in high-energy scattering of an exotic Reggeon with high significance exceeding $5\,σ$ for both channels, which is, most likely, related to the exotic $π_1(1600)$.8 △ Less

Submitted 23 July, 2026; v1 submitted 20 July, 2026; originally announced July 2026.

Comments: 8 pages, 7 figures + supplement material

A study of neutrinoless double electron capture in $^{40}$Ca from the AMoRE experiment

Authors: AMoRE Collaboration, A. Agrawal, V. V. Alenkov, P. Aryal, J. Beyer, B. Bhandari, R. S. Boiko, K. Boonin, O. Buzanov, C. R. Byeon, N. Chanthima, M. K. Cheoun, J. S. Choe, Seonho Choi, S. Choudhury, J. S. Chung, F. A. Danevich, M. Djamal, D. Drung, C. Enss, A. Fleischmann, A. M. Gangapshev, L. Gastaldo, Y. M. Gavrilyuk, A. M. Gezhaev , et al. (85 additional authors not shown)

Abstract: The search for neutrinoless double electron capture ($0ν\mathrm{2EC}$) provides a sensitive probe of lepton-number violation and the Majorana nature of neutrinos. We investigate the $0ν\mathrm{2EC}$ decay of $^{40}$Ca using cryogenic detectors equipped with metallic magnetic calorimeters in the AMoRE-I experiment. The analysis is based on a physics dataset corresponding to a total exposure of 7.32… ▽ More The search for neutrinoless double electron capture ($0ν\mathrm{2EC}$) provides a sensitive probe of lepton-number violation and the Majorana nature of neutrinos. We investigate the $0ν\mathrm{2EC}$ decay of $^{40}$Ca using cryogenic detectors equipped with metallic magnetic calorimeters in the AMoRE-I experiment. The analysis is based on a physics dataset corresponding to a total exposure of 7.32 kg$\cdot$yr from thirteen $^{40}$Ca$^{100}$MoO$_4$ crystals. No significant excess is observed, and a lower limit on the half-life is obtained as $T^{0ν}_{1/2} > 1.7 \times 10^{22}$ yr at 90$\%$ confidence level. An improved sensitivity is expected for the upcoming AMoRE-II experiment. These results demonstrate the potential of CaMoO$_4$ detectors to explore rare decay processes beyond the primary $^{100}$Mo $0νββ$ search program. △ Less

Submitted 8 July, 2026; originally announced July 2026.

Comments: 7 pages, 3 figures

From Voting to Agent Collaboration: Answer-Type-Aware LLM Pipelines for BioASQ 14b

Authors: Taeyun Roh, Eunha Lee, Wonjune Jang, Sohyun Chung, Junha Jung, Jaewoo Kang

Abstract: Biomedical question answering requires not only accurate extraction of information from scientific literature but also reliable integration of evidence across multiple documents. This study presents a question-type-specific large language model (LLM) framework for BioASQ 14b Task B, designed to improve answer robustness and evidence grounding in biomedical question answering. Rather than applying… ▽ More Biomedical question answering requires not only accurate extraction of information from scientific literature but also reliable integration of evidence across multiple documents. This study presents a question-type-specific large language model (LLM) framework for BioASQ 14b Task B, designed to improve answer robustness and evidence grounding in biomedical question answering. Rather than applying a single prompting strategy to all questions, the framework selects different inference procedures for yes/no, factoid, and list questions according to their distinct reasoning and evaluation requirements. For yes/no questions, snippet shuffling and self-reflection are used to reduce sensitivity to evidence ordering and improve decision stability. For factoid questions, full-snippet input is combined with chain-of-thought-based in-context learning to support accurate biomedical entity identification. For list questions, a multi-agent architecture is employed, in which evidence extraction, candidate generation, answer verification, and final aggregation are handled collaboratively. Preliminary experiments on BioASQ 13b were used to identify effective inference strategies for each question type, and the resulting framework was subsequently evaluated in the official BioASQ 14b Task B challenge. In the official evaluation, our framework showed competitive performance across multiple batches and achieved first place in the factoid subtask of Batch 4. These results demonstrate the effectiveness of combining question-type-specific inference, ensemble prediction, and agent-based verification for reliable biomedical question answering. △ Less

Submitted 7 July, 2026; originally announced July 2026.

Comments: 15 pages

Network Interdependency-Informed Power System Dynamic Trajectory Prediction Utilizing Black-Box Modeling of Multiple Inverter-Based Resources

Authors: Sungjoo Chung, Ying Zhang, Meng Yue, Hantao Cui

Abstract: Black-box modeling of inverter-based resources (IBRs) has attracted growing interest for real-time grid operation and control in the presence of proprietary electronic control architectures. Existing machine learning (ML)-based online dynamic trajectory prediction approaches using IBR black-box models either significantly accumulate prediction errors when multiple surrogates are simultaneously use… ▽ More Black-box modeling of inverter-based resources (IBRs) has attracted growing interest for real-time grid operation and control in the presence of proprietary electronic control architectures. Existing machine learning (ML)-based online dynamic trajectory prediction approaches using IBR black-box models either significantly accumulate prediction errors when multiple surrogates are simultaneously used or ignore measurement errors, limiting their deployment in practical grids. To address these limitations, this paper proposes a novel network interdependency-informed ML algorithm for online dynamic trajectory prediction in IBR-integrated power systems. A modular spatiotemporal attention network (STAN)-based predictor for the black-box modeling of each IBR unit is first proposed. Utilizing past measurements, the proposed STAN can effectively capture and predict the spatiotemporal dynamics of IBRs by employing an attention mechanism to attend to the most pertinent features for trajectory prediction. Furthermore, a novel hybrid physics-informed loss function that integrates a decoupled linearized AC power flow formulation is proposed. The proposed loss function effectively ensures physical consistency of predictions within network operation while avoiding the computational complexity of iterative power flow solving, thereby enabling efficient gradient backpropagation and overall improved prediction accuracy. Case studies on the IEEE 14- and WECC 179-bus systems demonstrate that the proposed method achieves significant accuracy enhancement and robustness against measurement errors, outperforming recent ML-based trajectory prediction methods. △ Less

Submitted 21 July, 2026; v1 submitted 7 July, 2026; originally announced July 2026.

Four Cold Giant Planets Discovered by High-Cadence Microlensing Surveys

Authors: Cheongho Han, Chung-Uk Lee, Andrzej Udalski, Ian A. Bond, Michael D. Albrow, Sun-Ju Chung, Andrew Gould, Youn Kil Jung, Kyu-Ha Hwang, Yoon-Hyun Ryu, Yossi Shvartzvald, In-Gu Shin, Jennifer C. Yee, Weicheng Zang, Hongjing Yang, Doeon Kim, Dong-Jin Kim, Seung-Lee Kim, Dong-Joo Lee, Sang-Mok Cha, Yongseok Lee, Byeong-Gon Park, Richard W. Pogge, Przemek Mróz, Michał K. Szymański , et al. (30 additional authors not shown)

Abstract: We report the discovery of four cold giant planets identified through the analysis of microlensing events detected by high-cadence surveys: OGLE-2016-BLG-0261, KMT-2025-BLG-0026, KMT-2025-BLG-0030, and KMT-2025-BLG-2272. The planetary signals appear as short-duration anomalies in the light curves and are well described by binary-lens single-source models with mass ratios between the lens component… ▽ More We report the discovery of four cold giant planets identified through the analysis of microlensing events detected by high-cadence surveys: OGLE-2016-BLG-0261, KMT-2025-BLG-0026, KMT-2025-BLG-0030, and KMT-2025-BLG-2272. The planetary signals appear as short-duration anomalies in the light curves and are well described by binary-lens single-source models with mass ratios between the lens components of order $q \sim 10^{-3}$. Finite-source effects are securely measured in three out of four events, enabling determinations of the angular Einstein radius. A Bayesian analysis incorporating the measured event timescale and angular Einstein radius yields host masses of $\sim 0.07$--$0.6~M_\odot$ and companion masses of $\sim 0.2$--$2.5~M_{\rm J}$, confirming that all companions lie in the giant-planet regime. The projected separations are ~ 0.7--6 au, placing all planets at or beyond the snow lines of their host stars. The inferred lens distances span $\sim 6.6$--$7.9$ kpc, with all systems consistent with bulge lenses. These detections expand the sample of cold giant planets from homogeneous high-cadence surveys and highlight the sensitivity of microlensing to planetary systems beyond the snow line, providing further constraints on the occurrence and properties of giant planets around low-mass stars. △ Less

Submitted 5 July, 2026; originally announced July 2026.

Comments: 14 pages, 7 figures

Sampling-Based Coordination-Informed Multi-Objective Multi-Robot Reinforcement Learning

Authors: Antonio Marino, Esteban Restrepo, Soon-jo Chung, Paolo Robuffo Giordano, Claudio Pacchierotti

Abstract: Multi-robot systems must simultaneously optimize competing objectives while maintaining coordinated behavior. Existing multi-agent reinforcement learning approaches often rely on fixed or centralized coordination, which limits adaptability and violates distributed constraints. This work introduces the Coordination-Informed Multi-Objective Reinforcement Learning (CIMORL) framework, integrating a di… ▽ More Multi-robot systems must simultaneously optimize competing objectives while maintaining coordinated behavior. Existing multi-agent reinforcement learning approaches often rely on fixed or centralized coordination, which limits adaptability and violates distributed constraints. This work introduces the Coordination-Informed Multi-Objective Reinforcement Learning (CIMORL) framework, integrating a distributed weight prediction mechanism, a privileged expert training strategy, and theoretical guarantees for Pareto-optimal solutions. We present the base CIMORL method alongside two sampling-based variants, CIMORL-TS (Tree Search) and CIMORL-MPPI (MPPI), which leverage privileged global information during training to enable fully decentralized deployment. Experimental validation in cooperative and adversarial scenarios demonstrates a $21.2\%$ hypervolume improvement and superior policy stability compared to state-of-the-art baselines. Real-world experiments with Crazyflie drones further validate the framework's robustness in resource allocation and multi-attacker multi-defend scenarios under partial observability. △ Less

Submitted 14 August, 2026; v1 submitted 29 June, 2026; originally announced June 2026.

Comments: 20 pages, 11 figures, 4 tables

KMT-2025-BLG-2093: Free-Floating Planet Candidate Near the Shore of the Einstein Desert

Authors: Yoon-Hyun Ryu, Andrew Gould, Kyu-Ha Hwang, Qiyue Qian, Michael D. Albrow, Sun-Ju Chung, Cheongho Han, Youn Kil Jung, Zhixing Li, Shude Mao, In-Gu Shin, Yossi Shvartzvald, Hongjing Yang, Jennifer C. Yee, Weicheng Zang, Dong-Jin Kim, Chung-Uk Lee, Byeong-Gon Park, Richard W. Pogge

Abstract: We analyze KMT-2025-BLG-2093, with angular Einstein radius $θ_{\rm E}=13.1\pm 2.8\,μ{\rm as}$, which makes it the second isolated microlens that lies in the ``Einstein Desert'' ($9\,μ{\rm as}<θ_{\rm E}<25\,μ{\rm as}$) between free-floating planets (FFPs) on one side and brown dwarfs and stars on the other. We discuss how its characteristics may give clues to future exploration of FFPs, especially… ▽ More We analyze KMT-2025-BLG-2093, with angular Einstein radius $θ_{\rm E}=13.1\pm 2.8\,μ{\rm as}$, which makes it the second isolated microlens that lies in the ``Einstein Desert'' ($9\,μ{\rm as}<θ_{\rm E}<25\,μ{\rm as}$) between free-floating planets (FFPs) on one side and brown dwarfs and stars on the other. We discuss how its characteristics may give clues to future exploration of FFPs, especially in the era of satellite missions that have a major FFP focus, including Earth 2.0 and Roman. △ Less

Submitted 6 July, 2026; v1 submitted 26 June, 2026; originally announced June 2026.

Comments: 16 pages, 4 figures, submitted to AAS

See & Sniff: Learning Visuo-Olfactory Representations

Authors: Seongyu Kim, Seungwoo Lee, Hyeonggon Ryu, Joon Son Chung, Arda Senocak

Abstract: While modern multimodal models integrate vision with language, audio, or touch, olfaction remains largely unexplored due to the lack of paired visuo-olfactory data. We introduce SmellNet-V, a scalable visuo-olfactory dataset built on the insight that odor identity is largely invariant to visual transformations within a semantic category. This allows us to synthetically pair smell-only samples with… ▽ More While modern multimodal models integrate vision with language, audio, or touch, olfaction remains largely unexplored due to the lack of paired visuo-olfactory data. We introduce SmellNet-V, a scalable visuo-olfactory dataset built on the insight that odor identity is largely invariant to visual transformations within a semantic category. This allows us to synthetically pair smell-only samples with semantically aligned in-the-wild web images, converting a unimodal olfactory dataset into a cross-modal benchmark without costly co-collection. Building on this dataset, we propose See & Sniff, a self-supervised framework that learns joint visuo-olfactory representations via dense local alignment and naturally produces smell saliency maps for spatial grounding of odor sources. We further introduce pixel-level smell localization task and a benchmark for evaluation. Our method surpasses smell-only baselines by 7% in smell classification from smell alone and generalizes to cross-modal retrieval and smell localization, establishing visuo-olfactory learning as a new direction in multimodal perception. △ Less

Submitted 25 June, 2026; originally announced June 2026.

Comments: ECCV 2026. Project Page: https://mm.kaist.ac.kr/projects/SeeandSniff/

Production and installation of wavelength-shifting reflective light enhancers for the Short-Baseline Near Detector

Authors: R. Acciarri, L. Aliaga-Soplin, R. Alvarez-Garrote, D. Andrade Aldana, C. Andreopoulos, A. Antonakis, S. Balasubramanian, A. Barnard, V. Basque, J. Bateman, M. C. Bazetto, A. Beever, E. Belchior, M. Betancourt, A. Bhat, M. Bishai, A. Blake, B. Bogart, D. Brailsford, A. Brandt, S. Brickner, M. B. Brunetti, L. Camilleri, D. Caratelli, D. Carber , et al. (172 additional authors not shown)

Abstract: We report on the design, production, and installation of a wavelength-shifting reflective system on the cathode of the Short-Baseline Near Detector (SBND), a liquid argon time projection chamber located along the Fermilab Booster Neutrino Beam. To increase and homogenize scintillation-light collection, 64 double-sided plates were fabricated from FR4, laminated with specular reflector film and coat… ▽ More We report on the design, production, and installation of a wavelength-shifting reflective system on the cathode of the Short-Baseline Near Detector (SBND), a liquid argon time projection chamber located along the Fermilab Booster Neutrino Beam. To increase and homogenize scintillation-light collection, 64 double-sided plates were fabricated from FR4, laminated with specular reflector film and coated with 300 $μ$g/cm$^2$ of tetraphenyl butadiene (TPB) wavelength shifter using controlled physical vapor deposition. The coating uniformity was validated through dedicated measurements of deposited mass and profilometry studies. Because exposure to ambient blue/UV light could degrade the TPB, protective filtering and controlled storage conditions were implemented during handling and installation. The coated plates were assembled between conductive meshes for high-voltage compatibility and installed in situ during detector integration. This system constitutes the largest TPB-coated area deployed in a neutrino detector. It operates in conjunction with SBND's photon detection system, which consists of photomultiplier tubes and X-ARAPUCAs. Early light-collection measurements show high uniformity and light response across the detector, supporting improved triggering, calorimetry, and position reconstruction in SBND. △ Less

Submitted 24 June, 2026; originally announced June 2026.

ProsoCodec: Prosody-Oriented Speech Codec for Voice Conversion

Authors: Jeongsoo Choi, Ji-Hoon Kim, Shujie Hu, Joon Son Chung

Abstract: Neural speech codecs efficiently compress speech and have become a foundation for speech generation, but they are typically learned as holistic representations that intertwine linguistic content, speaker identity, and prosody. While this design is effective for zero-shot voice cloning, it hinders downstream tasks that require prosody preservation or transfer, such as voice conversion. To address t… ▽ More Neural speech codecs efficiently compress speech and have become a foundation for speech generation, but they are typically learned as holistic representations that intertwine linguistic content, speaker identity, and prosody. While this design is effective for zero-shot voice cloning, it hinders downstream tasks that require prosody preservation or transfer, such as voice conversion. To address this, we introduce ProsoCodec, a prosody-oriented speech codec that models prosody as a conditional residual rather than as a disentangled stream. Specifically, by conditioning both the encoder and decoder on text and speaker embeddings as prefix tokens, the discrete bottleneck is encouraged to capture prosodic variation not explained by content and speaker. To further preserve prosody, we use the low-frequency mel band and train the model on paired same-speaker utterances. Experiments on voice conversion show improved prosody preservation and reduced source-timbre leakage. △ Less

Submitted 20 June, 2026; originally announced June 2026.

Comments: Interspeech 2026

MixProLAP: Mixture-Induced Uncertainty Modeling for Probabilistic Language-Audio Pretraining

Authors: Yu Nakagome, Jaesong Lee, Soo-Whan Chung

Abstract: Acoustic environments often contain multiple overlapping sound events, and the same acoustic scene can be described using diverse textual expressions, making audio-text alignment inherently ambiguous. This paper proposes a probabilistic audio-language pretraining framework to model many-to-many correspondence ambiguity in audio-text alignment. Unlike conventional contrastive methods that learn det… ▽ More Acoustic environments often contain multiple overlapping sound events, and the same acoustic scene can be described using diverse textual expressions, making audio-text alignment inherently ambiguous. This paper proposes a probabilistic audio-language pretraining framework to model many-to-many correspondence ambiguity in audio-text alignment. Unlike conventional contrastive methods that learn deterministic point embeddings, our approach represents each modality as a distribution and learns uncertainty-aware cross-modal alignment. Rather than relying on masking-based uncertainty simulation, we mix audio-text pairs to create overlapping sounds that better reflect real acoustic mixtures and capture semantic inclusion relations among sound events. We further introduce a multi-level inclusion loss to enforce representations consistent with these relations. Experiments on audio-text retrieval benchmarks show that the proposed method outperforms deterministic baselines. △ Less

Submitted 18 June, 2026; originally announced June 2026.

Comments: Accepted to Interspeech 2026

Who Wins the Conflict? Mechanistic Interpretability of Text Bias in Audio LLMs

Authors: Hyebin Cho, Suho Yoo, Jaehyuk Jang, Changick Kim, Joon Son Chung

Abstract: While Audio Large Language Models (Audio LLMs) excel at multimodal understanding, they suffer from text dominance, a bias where models blindly favor text over acoustic evidence, causing hallucinations. However, the internal mechanisms underlying how these models behave when audio and textual inputs contradict each other remain unexplored. In this work, we present the first mechanistic analysis of… ▽ More While Audio Large Language Models (Audio LLMs) excel at multimodal understanding, they suffer from text dominance, a bias where models blindly favor text over acoustic evidence, causing hallucinations. However, the internal mechanisms underlying how these models behave when audio and textual inputs contradict each other remain unexplored. In this work, we present the first mechanistic analysis of this phenomenon by tracing the propagation of internal representations across layers. Our investigation reveals three key findings: (i) text dominance is systematically and empirically across models; (ii) while text and audio rely on functionally distinct pathways, they ultimately converge into a shared semantic space in late layers; and (iii) the text pathway does not erase audio information, but rather actively suppresses intact audio representations. Building on these insights, we leverage back-patching, a training-free intervention that routes late-layer audio activations back into earlier layers. This amplifies the audio representations, enabling them to overcome textual suppression. Our evaluation shows that back-patching consistently reduces text dominance, paving the way for mechanistic multimodal alignment under conflict. △ Less

Submitted 17 June, 2026; originally announced June 2026.

Comments: Preprint

Acoustic Prompting via Stage-wise Modulation for Few-Shot Learning in Audio Language Models

Authors: Hyebin Cho, Jaehyuk Jang, Changick Kim, Joon Son Chung

Abstract: Audio-Language Models (ALMs) have shown remarkable success in zero-shot audio classification by aligning audio waveforms with text. Recent efforts to improve downstream performance focus on learning optimal text prompts. However, previous approaches focus on the text encoder, leaving the potential of learnable prompts within the audio encoder unexplored. In this paper, we propose a novel framework… ▽ More Audio-Language Models (ALMs) have shown remarkable success in zero-shot audio classification by aligning audio waveforms with text. Recent efforts to improve downstream performance focus on learning optimal text prompts. However, previous approaches focus on the text encoder, leaving the potential of learnable prompts within the audio encoder unexplored. In this paper, we propose a novel framework that introduces trainable prompts into the audio encoder to capture task-specific acoustic features. We demonstrate that integrating audio-side prompt learning with existing text-side approaches enhances few-shot adaptation. Through extensive experiments across 11 datasets show that integrating our method as a plug-and-play module alongside existing text prompt tuning generally leads to performance improvements. These findings suggest that explicitly modulating the audio representation space effectively complements text-only prompting approaches. The code is available at https://github.com/hyebin-c/aspl. △ Less

Submitted 14 June, 2026; originally announced June 2026.

Comments: Accepted to INTERSPEECH 2026

Superconducting diode effect in magnetic superconductors realized by nonreciprocal domain-wall dynamics

Authors: Dong Hui Han, Suk Bum Chung, Se Kwon Kim

Abstract: A superconducting diode effect is shown to arise in ferromagnetic superconductors through the nonreciprocal dynamics of magnetic domain walls. Specifically, we show that current-driven dynamics of a magnetic domain wall under a certain external field can exhibit a nonreciprocal Walker breakdown, possessing two distinct direction-dependent critical currents beyond which the domain wall precesses co… ▽ More A superconducting diode effect is shown to arise in ferromagnetic superconductors through the nonreciprocal dynamics of magnetic domain walls. Specifically, we show that current-driven dynamics of a magnetic domain wall under a certain external field can exhibit a nonreciprocal Walker breakdown, possessing two distinct direction-dependent critical currents beyond which the domain wall precesses continuously. In ferromagnetic superconductors, the constant rotation of a domain wall is shown to give rise to phase slips, opening up dissipation channels, whereby the nonreciprocal Walker breakdown is mapped to the superconducting diode effect. For the nonreciprocal Walker breakdown of a magnetic domain wall, we analytically examine its dependence on the magnetic field and the Gilbert damping and verify the theoretical results with micromagnetic simulations. We then extend the analysis to ferromagnetic superconductors by considering additional effects from the superconductivity and identify criteria for experimental conditions to realize the predicted superconducting diode effect. Our work demonstrates that topological defects, such as domain walls, in magnetic superconductors can serve as an intrinsic nanoscale platform for nonlinear nonreciprocal superconducting functionalities within a single homogeneous material, circumventing the need for complicated engineered heterostructures and thereby enabling the miniaturization of superconducting devices down to the nanometer scale that is challenging to achieve with conventional Josephson junctions. △ Less

Submitted 7 June, 2026; originally announced June 2026.

PAI: Preserving Amplitude Information in Representation-Based Time-Series Anomaly Detection

Authors: Kang Zhang, Wei Jian Lau, Shoushou Ren, Dong Lin, Joon Son Chung, Chuanhao Sun

Abstract: Representation-based time-series anomaly detection algorithms significantly outperform other methods on diverse anomaly detection tasks. However, we notice that they suffer from a major limitation in our evaluation - their learned embeddings are often amplitude-agnostic. Losing amplitude information can degrade performance on amplitude related anomalies, and this failure is prevalent across all ex… ▽ More Representation-based time-series anomaly detection algorithms significantly outperform other methods on diverse anomaly detection tasks. However, we notice that they suffer from a major limitation in our evaluation - their learned embeddings are often amplitude-agnostic. Losing amplitude information can degrade performance on amplitude related anomalies, and this failure is prevalent across all existing representation-based methods. To address aforementioned issues, we propose a new anomaly scoring scheme named PAI. PAI consists of two complementary modules, a diagnostic module and a final score augmentation function. The diagnostic module compares cosine and Euclidean scoring on the same representation bank to test whether amplitude information is already captured in the learned representation. Then in final score augmentation function, PAI computes a point-wise median and MAD deviation score and a local mean-shift score-which are fused with the representation score to produce the final anomaly score. On the TSB-AD-U-Eva and TAB UV datasets, PAI improves all four evaluated representation-based methods across every reported metric, achieving average VUS-PR gains of 98.4% and 36.8%, respectively. Among all evaluated combinations, PaAno + PAI achieves the best performance, outperforming the state-of-the-art method by 15%. Further evaluation on bootstrap confidence intervals, anomaly-type breakdowns, and a TS2Vec input-normalization ablation further support the proposed scheme. These results suggest that explicitly retaining amplitude information is important for representation-based time-series anomaly detection, which has been underemphasized in existing scoring schemes. Code is available at: https://github.com/pantheon5100/PAI △ Less

Submitted 7 June, 2026; originally announced June 2026.

Comments: 15 pages

An algebraic multiscale preconditioner for large sparse SPD matrices

Authors: Yingjie Zhou, Shubin Fu, Eric Tsz Shun Chung

Abstract: We present a two-grid algebraic multiscale preconditioner for large sparse symmetric positive definite systems arising from elliptic problems with highly heterogeneous coefficients. The coarse space is constructed directly from the system matrix by graph partitioning and local generalized eigenvalue solvers, yielding basis functions that capture the low-energy modes responsible for slow convergenc… ▽ More We present a two-grid algebraic multiscale preconditioner for large sparse symmetric positive definite systems arising from elliptic problems with highly heterogeneous coefficients. The coarse space is constructed directly from the system matrix by graph partitioning and local generalized eigenvalue solvers, yielding basis functions that capture the low-energy modes responsible for slow convergence. The method requires no geometric information, making it suitable for unstructured and matrix-only settings, and its construction is naturally parallelizable. Numerical results for heterogeneous Darcy flow problems show robustness with respect to coefficient contrast and problem size, better performance than standard algebraic multigrid on challenging large-scale cases, and good parallel scalability. △ Less

Submitted 3 June, 2026; originally announced June 2026.

Inference-Time Scaling for Joint Audio-Video Generation

Authors: Jaemin Jung, Kyeongha Rho, Inkyu Shin, Joon Son Chung

Abstract: Joint audio-video generation aims to synthesize realistic audio-video pairs that are both semantically aligned with text prompts and precisely synchronized. While existing joint audio-video generation models often require substantial training resources to improve fidelity, Inference-Time Scaling (ITS) has recently emerged as a promising training-free alternative in single-modality domains. However… ▽ More Joint audio-video generation aims to synthesize realistic audio-video pairs that are both semantically aligned with text prompts and precisely synchronized. While existing joint audio-video generation models often require substantial training resources to improve fidelity, Inference-Time Scaling (ITS) has recently emerged as a promising training-free alternative in single-modality domains. However, extending ITS from a single modality to multimodal domains is non-trivial, as it requires balancing multiple heterogeneous objectives. In this paper, we present the first comprehensive study of ITS for joint audio-video generation. We first demonstrate that a multi-verifier framework is essential to address the limitations of single-objective guidance, including asymmetric performance trade-offs and verifier hacking. Through systematic analysis, we then identify an optimal multi-verifier combination that yields balanced improvements across all quality dimensions. Finally, to effectively aggregate diverse reward signals, we propose Adaptive Reward Weighting (ARW), a novel test-time optimization algorithm. ARW treats reward aggregation as an online optimization problem, utilizing learnable parameters to calibrate reward variances without requiring prior knowledge of reward distributions, thereby ensuring robust multi-objective selection. Experimental results on VGGSound and JavisBench-mini benchmarks demonstrate that our framework significantly enhances semantic alignment, perceptual quality, and audio-visual synchronization of generated outputs. Synthesized samples and code are available on the project page: https://jung-jaemin.github.io/ITS-AVGen-Proj. △ Less

Submitted 2 June, 2026; originally announced June 2026.

Comments: Accepted by Transactions on Machine Learning Research (TMLR). Project page: https://jung-jaemin.github.io/ITS-AVGen-Proj/

Task-Induced Representational Invariances Depend on Learning Objective in Deep RL

Authors: Manu Srinath Halvagal, Sebastian Lee, SueYeon Chung

Abstract: Reinforcement Learning (RL) has long served as a model for goal-directed animal behavior in neuroscience. Modern deep RL has shown remarkable success across many domains, further strengthening this connection. The ability to learn abstract representations of high-dimensional state spaces underlies much of this success. However, theoretical understanding of these learned representations remains lim… ▽ More Reinforcement Learning (RL) has long served as a model for goal-directed animal behavior in neuroscience. Modern deep RL has shown remarkable success across many domains, further strengthening this connection. The ability to learn abstract representations of high-dimensional state spaces underlies much of this success. However, theoretical understanding of these learned representations remains limited, hindering direct comparisons between models and animal learning. We address this gap by analyzing deep RL representations through the lens of MDP reduction theory. Investigating canonical RL algorithms in a navigation task, we find that even when performance is comparable, the value-based method (DQN) learns representations that are invariant to MDP homomorphism symmetries, while the policy-gradient method (PPO) learns representations invariant to action symmetries. These differences emerge consistently across domains, have downstream consequences for transfer learning, and appear in LLMs in a prompt-dependent manner. Our findings provide a principled approach to comparing learned representations across RL algorithms, with demonstrated practical implications and possible insights for neural coding in the brain. △ Less

Submitted 1 June, 2026; originally announced June 2026.

Universal Quantum Transformer

Authors: Sungyong Chung, Alireza Talebpour

Abstract: Classical continuous-space neural networks fundamentally struggle to lock into exact formal rules, whether mathematical, such as modular arithmetic and non-Abelian group algebra, or linguistic, such as systematic compositional generalization. To approximate these discrete logical rules, they often rely on massive parameter scaling, resulting in stochastic instability even after delayed generalizat… ▽ More Classical continuous-space neural networks fundamentally struggle to lock into exact formal rules, whether mathematical, such as modular arithmetic and non-Abelian group algebra, or linguistic, such as systematic compositional generalization. To approximate these discrete logical rules, they often rely on massive parameter scaling, resulting in stochastic instability even after delayed generalization phenomena known as grokking. Here, we introduce the Universal Quantum Transformer (UQT), a novel, quantum-native computing architecture that uses the physical properties of multi-qubit systems as a universal inductive bias for exact algebraic and compositional reasoning. Rather than translating classical neural mechanisms, our framework relies entirely on parameterized geometric phase embedding and $SU(2)$ wave-interference. We demonstrate that an identical quantum attention circuit, operating on a highly compact 5 or 6 qubit substrate with only 551 to 1,650 trainable parameters, exactly learns three highly distinct formal classes: cyclic modular arithmetic ($\mathbb{Z}_{11}$), non-Abelian algebra (the $S_4$ permutation group), and systematic linguistic compositionality (the SCAN language). While standard classical models, including multi-layer perceptrons (MLPs) and Transformers, exhibit stochastic instability at convergence, the UQT achieves mathematically exact, deterministic generalization. We define this stricter regime as crystallization: a step beyond the well-known phenomenon of grokking. Finally, we deploy the UQT on noisy intermediate-scale quantum (NISQ) hardware, achieving 97.5% accuracy on IBM Quantum computers. These results demonstrate that the UQT provides a structurally suited inductive bias for exact formal reasoning that standard classical continuous-space architectures do not natively provide. △ Less

Submitted 27 July, 2026; v1 submitted 29 April, 2026; originally announced June 2026.

KVoiceBench, KOpenAudioBench, and KMMAU: Agent-Driven Korean Speech Benchmarks for Evaluating SpeechLMs

Authors: Haechan Kim, Seungjun Chung, Inkyu Park, Jihoo Lee, Jonghyun Lee

Abstract: Speech language models (SpeechLMs) have achieved substantial progress by extending large language models (LLMs) to the speech modality. However, SpeechLM evaluation remains heavily centered on English, limiting reliable assessment of multilingual speech capabilities. Straightforward benchmark transfer through ASR, translation, normalization, and TTS can corrupt language-specific instructions, answ… ▽ More Speech language models (SpeechLMs) have achieved substantial progress by extending large language models (LLMs) to the speech modality. However, SpeechLM evaluation remains heavily centered on English, limiting reliable assessment of multilingual speech capabilities. Straightforward benchmark transfer through ASR, translation, normalization, and TTS can corrupt language-specific instructions, answer constraints, and spoken forms; for audio understanding, transferring source-language audio also fails to preserve target-language speaker attributes, accents, and paralinguistic properties. To address these limitations, we propose two human-agent benchmark-construction frameworks: one transfers source-language SpokenQA benchmarks into target-language SpokenQA benchmarks, and the other converts target-language ASR corpora into audio understanding benchmarks using transcriptions and speaker metadata. Using these frameworks, we construct and publicly release three Korean speech benchmarks: KVoiceBench and KOpenAudioBench for Korean SpokenQA, and KMMAU for Korean audio understanding, comprising 12,345 samples in total. We evaluate eight recent SpeechLMs and find that English-Korean performance gaps vary substantially across models and task families, and that SpokenQA and audio understanding rankings diverge, revealing complementary weaknesses invisible to English-only evaluation. △ Less

Submitted 27 May, 2026; originally announced May 2026.

Comments: 16 pages, 4 figures

Two Speeds of Learning: A Representation-Readout Decomposition of Grokking and Double Descent

Authors: Chi-Ning Chou, Oscar Uzdelewicz, Neng-Chun Chiu, Yao-Yuan Yang, SueYeon Chung

Abstract: Training loss and accuracy are the standard signals used to monitor generalization during deep neural network training. Two well-documented phenomena complicate this picture: in grokking, train loss falls rapidly while test performance improves abruptly only after a long delay; in epoch-wise double descent, train loss decreases monotonically while test loss or error rises and falls. Existing accou… ▽ More Training loss and accuracy are the standard signals used to monitor generalization during deep neural network training. Two well-documented phenomena complicate this picture: in grokking, train loss falls rapidly while test performance improves abruptly only after a long delay; in epoch-wise double descent, train loss decreases monotonically while test loss or error rises and falls. Existing accounts are often task-specific, and a task-agnostic analysis framework for diagnosing and explaining these phenomena across realistic tasks and architectures is missing. We address this challenge by analyzing two competing processes that underlie learning dynamics: representation learning in the encoder and readout calibration in the final classifier. Using tools from representational geometry, neural tangent kernels, and linear probing, we show that both processes are active throughout training, with the fluctuations of their relative speed giving rise to seemingly anomalous generalization dynamics. Applying the representation-readout decomposition to grokking across a wide range of tasks and architectures, we find that the readout is train-biased before grokking onset, and representation learning is gradual but not absent, contrary to the lazy-to-rich account. The framework further provides diagnostic signatures distinguishing spurious from genuine generalization: in a previously reported MNIST grokking example and an epoch-wise double descent example, apparent delayed or non-monotone generalization is shown to arise from representation degradation and readout misalignment induced by non-standard training recipes. Together, these results establish the representation-readout decomposition as a top-down framework for understanding learning dynamics and revealing underlying algorithms for interpretability research. △ Less

Submitted 28 May, 2026; v1 submitted 26 May, 2026; originally announced May 2026.

Raon-Speech Technical Report

Authors: Beomsoo Kim, Changho Choi, Dohyun Kim, Dongki Lee, Ethan Ewer, Eunchong Kim, Gyeongman Kim, Haechan Kim, Hyeonghwan Kim, Inkyu Park, Jihun Yun, Jihwan Moon, Jiyun Kim, Joonghyun Bae, Junhyuck Kim, Minkyu Kim, Sehun Lee, Seungjun Chung, Sungwoo Cho, Dongmin Park, Dongwon Kim, Hara Kang, Jonghyun Lee, Keon Lee, Kangwook Lee , et al. (1 additional authors not shown)

Abstract: We present Raon-Speech, a top-performing 9B-parameter speech language model (SpeechLM) for English and Korean speech understanding, answering, and generation, and Raon-SpeechChat, a high-performing full-duplex extension for natural real-time conversation. Raon-Speech successfully transforms a pre-trained LLM into a SpeechLM that both understands and generates speech while preserving strong text ca… ▽ More We present Raon-Speech, a top-performing 9B-parameter speech language model (SpeechLM) for English and Korean speech understanding, answering, and generation, and Raon-SpeechChat, a high-performing full-duplex extension for natural real-time conversation. Raon-Speech successfully transforms a pre-trained LLM into a SpeechLM that both understands and generates speech while preserving strong text capabilities. It trains on 1.38M hours of highly curated English and Korean speech and text datasets with the following training stages: (1) speech modules alignment, (2) end-to-end SpeechLM pre-training with knowledge distillation, and (3) multi-task preference optimization-based post-training. Across 42 English and Korean speech and text benchmarks, Raon-Speech establishes the strongest overall profile on speech-centric tasks in our comparison against eight similarly sized recent audio foundation models, including Qwen2.5-Omni and Fun-Audio-Chat, while preserving strong text question answering performance. Building upon it, Raon-SpeechChat enables natural full-duplex conversation by continual training on 119K hours of time-aligned real and synthetic dialogue data. It proceeds through three complementary training stages: (1) causal encoder adaptation, (2) full-duplex pre-training, (3) full-duplex fine-tuning for voice and role-control. On multiple full-duplex benchmarks, Raon-SpeechChat shows its clearest strengths on the turn-taking and interruption-sensitive behaviors covered by FDB v1.0, and remains competitive across the broader full-duplex evaluation suite. We open-source all model checkpoints, the training and inference pipeline, and an interactive demo. △ Less

Submitted 8 April, 2026; originally announced May 2026.

AtelierEval: Agentic Evaluation of Humans & LLMs as Text-to-Image Prompters

Authors: Hanjun Luo, Zhimu Huang, Sylvia Chung, Yiran Wang, Yingbin Jin, Jialin Li, Jiang Li, Xinfeng Li, Hanan Salam

Abstract: Text-to-image (T2I) systems increasingly rely on upstream prompters, either humans or multimodal large language models (MLLMs), to translate user intent into detailed prompts. Yet current benchmarks fix the prompt and only evaluate T2I models, leaving the prompting proficiency of this upstream component entirely unmeasured. We introduce AtelierEval, the first unified benchmark that quantifies prom… ▽ More Text-to-image (T2I) systems increasingly rely on upstream prompters, either humans or multimodal large language models (MLLMs), to translate user intent into detailed prompts. Yet current benchmarks fix the prompt and only evaluate T2I models, leaving the prompting proficiency of this upstream component entirely unmeasured. We introduce AtelierEval, the first unified benchmark that quantifies prompting proficiency across 360 expert-crafted tasks. Grounded in a cognitive view, it spans three task categories and instantiates tasks using a taxonomy of real-world challenges, with a dual interface for both humans and MLLMs. To enable scalable and reliable evaluation, we propose AtelierJudge, a skill-based, memory-augmented agentic evaluator. It produces subjective and objective scores for prompt-image pairs, achieving a Spearman correlation of 0.79 with human experts, approaching human performance. Extensive experiments benchmark 8 MLLMs against 48 human users across 4 T2I backends, validate AtelierEval as a robust diagnostic tool, and reveal the superiority of mimicry over planning, advocating for an image-augmented direction for future prompters. Our work is released to support future research. △ Less

Submitted 21 May, 2026; originally announced May 2026.

Comments: Accepted by ICML 2026

CAdam: Context-Adaptive Moment Estimation for 3D Gaussian Densification in Generative Distillation

Authors: SeungJeh Chung, Geonho Park, Misong Kim, HyeongYeop Kang

Abstract: Adaptive densification is the engine of 3D Gaussian Splatting (3DGS). However, when transposed to the optimization-based Generative Distillation paradigm, this reconstruction-native mechanism reveals fundamental limitations, resulting in inefficient representations cluttered with redundant primitives. We diagnose this failure as a Densification Dilemma stemming from the stochastic nature of genera… ▽ More Adaptive densification is the engine of 3D Gaussian Splatting (3DGS). However, when transposed to the optimization-based Generative Distillation paradigm, this reconstruction-native mechanism reveals fundamental limitations, resulting in inefficient representations cluttered with redundant primitives. We diagnose this failure as a Densification Dilemma stemming from the stochastic nature of generative guidance: the standard magnitude-based accumulation indiscriminately aggregates transient noise alongside geometric signals, making it difficult to strike a balance between over-densification and under-fitting. To resolve this, we introduce Context-Adaptive Moment Estimation (CAdam), a novel framework that reinterprets densification as a statistically grounded signal verification problem. CAdam leverages the first moment of gradients to exploit the interference principle, where stochastic fluctuations cancel out via destructive interference while consistent geometric drifts accumulate via constructive interference, effectively disentangling the underlying signal from the generative noise floor. This is further augmented by a quantile-based context awareness and an intrinsic Signal-to-Noise Ratio (SNR) gating mechanism, which ensure robust adaptation across optimization stages and enable the soft termination of densification. Extensive experiments across diverse objectives (SDS, ISM, VFDS) and strong generative 3DGS backbones show that CAdam reduces Gaussian count by 85%-97% relative to standard densification while preserving overall comparable perceptual quality. These results highlight signal-aware density control as a practical way to improve memory efficiency in optimization-based generative distillation. △ Less

Submitted 20 May, 2026; originally announced May 2026.

Comments: Accepted to SIGGRAPH 2026 Conference Papers. 12 pages, 8 figures

Raon-OpenTTS: Open Models and Data for Robust Text-to-Speech

Authors: Semin Kim, Seungjun Chung, Taehong Moon, Sangheon Lee, Minyoung Ahn, Keon Lee, Nam Soo Kim, Jaewoong Cho, Ludwig Schmidt, Kangwook Lee, Dongmin Park

Abstract: Recent advances in text-to-speech (TTS) models show impressive speech naturalness and quality, yet the role of large-scale open data in driving this progress remains underexplored. In this work, we introduce Raon-OpenTTS, an open TTS model that performs competitively with state-of-the-art closed-data TTS models, and Raon-OpenTTS-Pool, a large-scale open dataset for reproducible TTS training. Raon-… ▽ More Recent advances in text-to-speech (TTS) models show impressive speech naturalness and quality, yet the role of large-scale open data in driving this progress remains underexplored. In this work, we introduce Raon-OpenTTS, an open TTS model that performs competitively with state-of-the-art closed-data TTS models, and Raon-OpenTTS-Pool, a large-scale open dataset for reproducible TTS training. Raon-OpenTTS-Pool consists of 615K hours of 240M speech segments aggregated from publicly available English speech corpora and web-sourced recordings. With a model-based filtering pipeline applied to Raon-OpenTTS-Pool, we derive Raon-OpenTTS-Core, a curated, high-quality subset of 510K hours and 194M speech segments. Using Raon-OpenTTS-Core, we train Raon-OpenTTS, a series of diffusion transformer (DiT)-based TTS models from 0.3B to 1B parameters. On multiple benchmarks, Raon-OpenTTS-1B shows comparable performance to state-of-the-art models such as Qwen3-TTS and CosyVoice 3, which are trained on several million hours of proprietary speech data. Notably, on Seed-TTS-Eval, Raon-OpenTTS-1B achieves a word error rate (WER) of 1.78% and a speaker similarity (SIM) of 0.749, ranking second on WER and first on SIM among recent open-weight TTS baselines. On CV3-Hard-EN, Raon-OpenTTS-1B achieves a WER of 6.15% and a SIM of 0.775, ranking first on both metrics. Furthermore, to support robust evaluation, we introduce Raon-OpenTTS-Eval, a structured benchmark for assessing TTS robustness across diverse acoustic conditions including clean, noisy, in-the-wild, and expressive speech. On Raon-OpenTTS-Eval, Raon-OpenTTS-1B achieves the best average WER and SIM among all evaluated models, and the second-best human preference, as measured by comparative mean opinion score (CMOS). Our data pool, filtering pipeline, training code, and checkpoints are publicly available at https://github.com/krafton-ai/RAON-OpenTTS. △ Less

Submitted 15 June, 2026; v1 submitted 20 May, 2026; originally announced May 2026.

Four-Body Gravitational Microlensing Events Involving Both a Binary Lens and a Binary Source

Authors: Cheongho Han, Chung-Uk Lee, Andrzej Udalski, Michael D. Albrow, Sun-Ju Chung, Andrew Gould, Youn Kil Jung, Kyu-Ha~Hwang, Yoon-Hyun Ryu, Yossi Shvartzvald, In-Gu Shin, Jennifer C. Yee, Weicheng Zang, Hongjing Yang, Doeon Kim, Dong-Jin Kim, Seung-Lee Kim, Dong-Joo Lee, Sang-Mok Cha, Yongseok Lee, Byeong-Gon Park, Richard W. Pogge, Przemek Mróz, Michał K. Szymańskim Jan Skowron, Radosław Poleski , et al. (9 additional authors not shown)

Abstract: We present detailed analyses of three anomalous microlensing events--KMT-2021-BLG-0209, KMT-2021-BLG-0901, and OGLE-2025-BLG-0356--identified from a systematic re-examination of KMTNet light curves for which previous modeling attempts failed or left persistent residuals. Although all three events show caustic-related features consistent with binary-lens microlensing, we find that their full light-… ▽ More We present detailed analyses of three anomalous microlensing events--KMT-2021-BLG-0209, KMT-2021-BLG-0901, and OGLE-2025-BLG-0356--identified from a systematic re-examination of KMTNet light curves for which previous modeling attempts failed or left persistent residuals. Although all three events show caustic-related features consistent with binary-lens microlensing, we find that their full light-curve structures can be described by four-body configurations that required four-body configurations involving a binary lens and a binary source. In KMT-2021-BLG-0209, weak caustic-exit residuals arise from a faint companion source undergoing an additional caustic interaction. In KMT-2021-BLG-0901, a late-time re-brightening is produced when the secondary source encounters the resonant caustic long after the primary. For OGLE-2025-BLG-0356, we test the degeneracy between 3L1S and 2L2S interpretations of a short isolated anomaly and find that the 2L2S model provides a significantly better fit. Source colors and magnitudes indicate binary sources composed of (G8V, M3V), (G8V, K2V), and (G6V, G8V) stars for the three events, respectively. Bayesian inference suggests that the lenses are predominantly low-mass binaries, including one system (KMT-2021-BLG-0901) with a companion consistent with a brown dwarf. These events add to the growing sample of well-characterized 2L2S systems and underscore the importance of systematically testing complex models, particularly in anticipation of the high-precision microlensing data expected from the Roman Space Telescope survey. △ Less

Submitted 17 May, 2026; originally announced May 2026.

Comments: 11 pages, 6 figures

Geometric Fault Identification via Mirror Descent Learning

Authors: Mahdi Taheri, Haeyoon Han, Soon-Jo Chung, Fred Y. Hadaegh

Abstract: This paper develops a fault detection and identification (FDI) method for nonlinear control-affine systems under simultaneous actuator and sensor faults. We adopt a geometric approach to study the isolability of faults in the sense of the principal angles between subspaces corresponding to each actuator and sensor fault. As for the fault identification, a hybrid estimator that consists of a Luenbe… ▽ More This paper develops a fault detection and identification (FDI) method for nonlinear control-affine systems under simultaneous actuator and sensor faults. We adopt a geometric approach to study the isolability of faults in the sense of the principal angles between subspaces corresponding to each actuator and sensor fault. As for the fault identification, a hybrid estimator that consists of a Luenberger-like observer with contraction guarantees is developed. Moreover, neural networks are embedded in the mentioned observer to estimate actuator and sensor faults. Considering that the training dataset for neural networks cannot be representative of every fault scenario, the last layer of each network is adapted using mirror descent-based laws. The mirror descent-based adaptive laws impose isolability conditions for fault channels and do not assume a quadratic parameter estimation space to consider the geometry of the fault subspaces. A Lyapunov-based analysis establishes that the state and parameter estimation errors are uniformly ultimately bounded. The effectiveness of our proposed FDI method is illustrated on the 3-axis attitude control system of a spacecraft. △ Less

Submitted 16 May, 2026; originally announced May 2026.

Systematic KMTNet Planetary Anomaly Search. XIII. Complete Sample of 2021 Prime Field Planets

Authors: In-Gu Shin, Jennifer C. Yee, Weicheng Zang, Cheongho Han, Andrew Gould, Shude Mao, Chung-Uk Lee, Yoon-Hyun Ryu, Ian A. Bond, Takahiro Sumi, Michael D. Albrow, Sun-Ju Chung, Kyu-Ha Hwang, Youn Kil Jung, Yossi Shvartzvald, Hongjing Yang, Sang-Mok Cha, Dong-Jin Kim, Seung-Lee Kim, Dong-Joo Lee, Yongseok Lee, Byeong-Gon Park, Richard W. Pogge, Fumio Abe, David P. Bennett , et al. (17 additional authors not shown)

Abstract: The Systematic KMTNet Planetary Anomaly Search series was conducted using the KMTNet data archived from $2016$ to $2019$. From this first phase of the series, we reported a total of $50$ planetary systems hidden in the data archive, which represent about $35\%$ of the total microlensing planets discovered from $2016$ to $2019$, demonstrating that this semi-machine-based search is a crucial channel… ▽ More The Systematic KMTNet Planetary Anomaly Search series was conducted using the KMTNet data archived from $2016$ to $2019$. From this first phase of the series, we reported a total of $50$ planetary systems hidden in the data archive, which represent about $35\%$ of the total microlensing planets discovered from $2016$ to $2019$, demonstrating that this semi-machine-based search is a crucial channel for building a complete microlensing planet sample. We continue this series for $2021$ and beyond to expand the microlensing planet sample. In this work for the $2021$ KMTNet high-cadence fields (Prime fields), we find seven hidden planetary systems and three planet candidates. These new planets represent about $33\%$ of the total microlensing planets discovered within the Prime fields observed during the $2021$ bulge season. While the by-eye search is the primary channel for detecting microlensing planets (i.e., two-thirds of microlensing planet discoveries), this work clearly shows that a systematic search series is still necessary for constructing a complete microlensing planet sample. Such a sample is essential for conducting unbiased statistical studies of planet demographics in our Galaxy. Datasets for all the events used for analyses in this work are publicly available: doi:10.5281/zenodo.21472225. △ Less

Submitted 12 August, 2026; v1 submitted 16 May, 2026; originally announced May 2026.

Comments: 33 pages, 18 Tables, 16 Figures. Accepted for publication in AJ

SpeakerLLM: A Speaker-Specialized Audio-LLM for Speaker Understanding and Verification Reasoning

Authors: KiHyun Nam, Jungwoo Heo, Siu Bae, Ha-Jin Yu, Joon Son Chung

Abstract: As audio-first agents become increasingly common in physical AI, conversational robots, and screenless wearables, audio large language models (audio-LLMs) must integrate speaker-specific understanding to support user authorization, personalization, and context-aware interaction. This requires modeling who is speaking, how the voice sounds, and how recording conditions affect speaker cues. Conventi… ▽ More As audio-first agents become increasingly common in physical AI, conversational robots, and screenless wearables, audio large language models (audio-LLMs) must integrate speaker-specific understanding to support user authorization, personalization, and context-aware interaction. This requires modeling who is speaking, how the voice sounds, and how recording conditions affect speaker cues. Conventional speaker verification systems provide strong scalar scores but little linguistic evidence, while current audio-LLMs and speaker-aware language models have limited ability to organize speaker information beyond binary labels or descriptive profiles. We present SpeakerLLM, a speaker-specialized audio-LLM framework that unifies single-utterance speaker profiling, recording-condition understanding, utterance-pair speaker comparison, and evidence-organized verification reasoning within a natural-language interface. We construct verification-reasoning targets and a decision-composition policy that separate profile-level evidence from the final same-or-different decision and organize recording condition, profile evidence, and the decision into a structured trace. At its core, SpeakerLLM uses a hierarchical speaker tokenizer designed to capture multiple granularities of speaker evidence. Utterance-level speaker embeddings summarize identity and profile-level cues, whereas frame-level speaker features preserve fine-grained acoustic descriptors. Experiments show that SpeakerLLM-Base improves speaker-profile and recording-condition understanding over general audio-LLMs, while SpeakerLLM-VR preserves strong generated-verdict accuracy and produces decision traces grounded in the supervised verification reasoning schema. We will release the metadata-enriched supervision dataset and target-construction code for reproducibility. △ Less

Submitted 14 May, 2026; originally announced May 2026.

FiTS: Interpretable Spiking Neurons via Frequency Selectivity and Temporal Shaping

Authors: Jongmin Choi, Joon Son Chung

Abstract: Spiking Neural Networks (SNNs) are a promising framework for event-driven temporal processing. Prior work has improved temporal modeling through richer neuron dynamics and network-level mechanisms such as recurrence and delays, but it remains unclear how individual spiking neurons should specialize within a network. In this work, we introduce FiTS, a spiking neuron that factorizes temporal computa… ▽ More Spiking Neural Networks (SNNs) are a promising framework for event-driven temporal processing. Prior work has improved temporal modeling through richer neuron dynamics and network-level mechanisms such as recurrence and delays, but it remains unclear how individual spiking neurons should specialize within a network. In this work, we introduce FiTS, a spiking neuron that factorizes temporal computation within each neuron into Frequency Selectivity (FS) and Temporal Shaping (TS). The FS module parameterizes each neuron's target frequency as the maximizer of its subthreshold magnitude response, while the TS module reshapes when frequency components contribute to membrane voltage accumulation through group-delay modulation. On auditory benchmarks where frequency selectivity and timing are central to the input structure, FiTS consistently improves over a plain Leaky Integrate-and-Fire (LIF) baseline in simple feedforward SNNs without recurrence or network-level delays, while remaining competitive with strong temporal SNN baselines. Beyond accuracy, the learned target frequencies and group-delay shifts provide interpretable neuron-level summaries of the frequency and timing organization learned within the network. △ Less

Submitted 13 May, 2026; originally announced May 2026.

Comments: 23 pages, 7 figures

Keep What Audio Cannot Say: Context-Preserving Token Pruning for Omni-LLMs

Authors: Chaeyoung Jung, Kyeongha Rho, Joon Son Chung

Abstract: Omnimodal Large Language Models (Omni-LLMs) incur substantial computational overhead due to the large number of multimodal input tokens they process, making token reduction essential for real-world deployment. Existing Omni-LLM pruning methods typically reduce this cost by selecting tokens that are important for the current query or strongly aligned with cross-modal cues. However, such strategies… ▽ More Omnimodal Large Language Models (Omni-LLMs) incur substantial computational overhead due to the large number of multimodal input tokens they process, making token reduction essential for real-world deployment. Existing Omni-LLM pruning methods typically reduce this cost by selecting tokens that are important for the current query or strongly aligned with cross-modal cues. However, such strategies can discard evidence that falls outside these criteria, even when needed for different questions or for understanding context beyond aligned audio-visual cues. To address this limitation, we reframe Omni-LLM token reduction as preserving broad audio-visual context while removing cross-modal redundancy. We propose ContextGuard, an inference-time token pruning framework built on this principle. ContextGuard predicts coarse visual semantics from audio and prunes video tokens whose coarse semantics are likely recoverable from audio, while retaining additional video tokens to preserve localized visual details that audio alone cannot specify. For further compression, our method merges temporally similar video tokens. The framework requires no downstream LLM fine-tuning and uses only an independently trained lightweight predictor. On Qwen2.5-Omni and Video-SALMONN2+ at 3B and 7B scales across six audio-visual benchmarks, ContextGuard outperforms prior inference-time pruning methods while pruning more tokens. Notably, on Qwen2.5-Omni 7B, ContextGuard achieves full-token-level performance on five of six benchmarks while pruning 55% of input tokens. △ Less

Submitted 12 May, 2026; originally announced May 2026.

Probing Cross-modal Information Hubs in Audio-Visual LLMs

Authors: Jihoo Jung, Chaeyoung Jung, Ji-Hoon Kim, Joon Son Chung

Abstract: Audio-visual large language models (AVLLMs) have recently emerged as a powerful architecture capable of jointly reasoning over audio, visual, and textual modalities. In AVLLMs, the bidirectional interaction between audio and video modalities introduces intricate processing dynamics, necessitating a deeper understanding of their internal mechanisms. However, unlike extensively studied text-only or… ▽ More Audio-visual large language models (AVLLMs) have recently emerged as a powerful architecture capable of jointly reasoning over audio, visual, and textual modalities. In AVLLMs, the bidirectional interaction between audio and video modalities introduces intricate processing dynamics, necessitating a deeper understanding of their internal mechanisms. However, unlike extensively studied text-only or large vision language models, the internal workings of AVLLMs remain largely unexplored. In this paper, we focus on cross-modal information flow between audio and visual modalities in AVLLMs, investigating where information derived from one modality is encoded within the token representations of the other modality. Through an analysis of multiple recent AVLLMs, we uncover two common findings. First, AVLLMs primarily encode integrated audio-visual information in sink tokens. Second, sink tokens do not uniformly hold cross-modal information. Instead, a distinct subset of sink tokens, which we term cross-modal sink tokens, specializes in storing such information. Based on these findings, we further propose a simple training-free hallucination mitigation method by encouraging reliance on integrated cross-modal information within cross-modal sink tokens. Our code is available at https://github.com/kaistmm/crossmodal-hub. △ Less

Submitted 11 May, 2026; v1 submitted 11 May, 2026; originally announced May 2026.

Comments: Accepted by ICML 2026

Mass Production of 2023 KMTNet Microlensing Planets. III: Three Planets from the Subprime Field

Authors: Hongyu Li, Zhixing Li, Weicheng Zang, Yoon-Hyun Ryu, Andrzej Udalski, Takahiro Sumi, Hongjing Yang, Jiyuan Zhang, Shude Mao, Michael Albrow, Sun-Ju Chung, Andrew Gould, Cheongho Han, Kyu-Ha Hwang, Youn Kil Jung, In-Gu Shin, Yossi Shvartzvald, Jennifer Yee, Sang-Mok Cha, Dong-Jin Kim, Seung-Lee Kim, Chung-Uk Lee, Dong-Joo Lee, Yongseok Lee, Byeong-Gon Park , et al. (34 additional authors not shown)

Abstract: To complete the analysis of the 2023 KMTNet subprime-field microlensing planetary events identified by its AlertFinder system, we present the analysis of six events, KMT-2023-BLG-(1810, 0084, 1118, 0584, 1697, 2218). We find that the first three events are securely confirmed as planetary, with inferred mass ratios of $\log q \sim -1.9$, $-2.0$, and $-2.6$, respectively. The remaining three events… ▽ More To complete the analysis of the 2023 KMTNet subprime-field microlensing planetary events identified by its AlertFinder system, we present the analysis of six events, KMT-2023-BLG-(1810, 0084, 1118, 0584, 1697, 2218). We find that the first three events are securely confirmed as planetary, with inferred mass ratios of $\log q \sim -1.9$, $-2.0$, and $-2.6$, respectively. The remaining three events exhibit the well-known degeneracy between binary-lens/single-source (2L1S) and single-lens/binary-source (1L2S) models, and two of these also admit viable stellar binary solutions. A Bayesian analysis indicates that the companions in the confirmed planetary events are likely either super-Jupiters orbiting beyond the snow line of M- or K-dwarf hosts or, for two degenerate solutions of KMT-2023-BLG-1118, Saturn-mass planets orbiting late-type M dwarfs. To date, the 2023 KMTNet sample contains 25 unambiguous planetary events, and its mass-ratio distribution is consistent with that of the KMTNet planetary sample from 2016--2019. △ Less

Submitted 8 May, 2026; originally announced May 2026.

Comments: Submitted to PASP

LRS-VoxMM: A benchmark for in-the-wild audio-visual speech recognition

Authors: Doyeop Kwak, Jeongsoo Choi, Suyeon Lee, Joon Son Chung

Abstract: We introduce LRS-VoxMM, an in-the-wild benchmark for audio-visual speech recognition (AVSR). The benchmark is derived from VoxMM, a dataset of diverse real-world spoken conversations with human-annotated transcriptions. We select AVSR-suitable samples and preprocess them in an LRS-style format for direct use in existing AVSR pipelines. Compared with commonly used benchmarks, LRS-VoxMM covers a mor… ▽ More We introduce LRS-VoxMM, an in-the-wild benchmark for audio-visual speech recognition (AVSR). The benchmark is derived from VoxMM, a dataset of diverse real-world spoken conversations with human-annotated transcriptions. We select AVSR-suitable samples and preprocess them in an LRS-style format for direct use in existing AVSR pipelines. Compared with commonly used benchmarks, LRS-VoxMM covers a more diverse range of scenarios and acoustic conditions. We also release distorted evaluation sets with additive noise, reverberation, and bandwidth limitation to support evaluation under severe acoustic degradation. Experimental results show that LRS-VoxMM is considerably harder than LRS3 and that the contribution of visual information becomes more evident as the audio signal degrades. LRS-VoxMM supports more realistic AVSR benchmarking and encourages further research on the role of visual information in challenging real-world conditions. △ Less

Submitted 30 April, 2026; originally announced April 2026.

Comments: Technical report for the LRS-VoxMM dataset release. Project page: https://mm.kaist.ac.kr/projects/voxmm

ICPR 2026 Competition on Low-Resolution License Plate Recognition

Authors: Rayson Laroca, Valfride Nascimento, Donggun Kim, Sanghyeok Chung, Subin Bae, Uihwan Seo, Seungsang Oh, Chi M. Phung, Minh G. Vo, Xingsong Ye, Yongkun Du, Yuchen Su, Zhineng Chen, Sunhee Heo, Hyangwoo Lee, Kihyun Na, Khanh V. Vu Nguyen, Sang T. Pham, Duc N. N. Phung, Trong P. Le, Vy N. Vo Tran, David Menotti

Abstract: Low-Resolution License Plate Recognition (LRLPR) remains a challenging problem in real-world surveillance scenarios, where long capture distances, compression artifacts, and adverse imaging conditions can severely degrade license plate legibility. To promote progress in this area, we organized the ICPR 2026 Competition on Low-Resolution License Plate Recognition, the first competition specifically… ▽ More Low-Resolution License Plate Recognition (LRLPR) remains a challenging problem in real-world surveillance scenarios, where long capture distances, compression artifacts, and adverse imaging conditions can severely degrade license plate legibility. To promote progress in this area, we organized the ICPR 2026 Competition on Low-Resolution License Plate Recognition, the first competition specifically dedicated to LRLPR using real low-quality data collected under operationally relevant conditions. The competition was based on the LRLPR-26 dataset, which comprises 20,000 training tracks and 3,000 test tracks; each training track contains five low-resolution and five high-resolution images of the same license plate. Notably, a total of 269 teams from 41 countries registered for the competition, and 99 teams submitted valid entries in the Blind Test Phase. The winning team achieved a Recognition Rate of 82.13%, and four teams surpassed the 80% mark, highlighting both the high level of competition at the top of the leaderboard and the continued difficulty of the task. In addition to presenting the competition design, evaluation protocol, and main results, this paper summarizes the methods adopted by the top-5 teams and discusses current trends and promising directions for future research on LRLPR. The competition webpage is available at https://icpr26lrlpr.github.io/ △ Less

Submitted 24 April, 2026; originally announced April 2026.

Comments: Accepted for presentation at the International Conference on Pattern Recognition (ICPR) 2026

Planetary Exploration 3.0: A Roadmap for Software-Defined, Radically Adaptive Space Systems

Authors: Masahiro Ono, Daniel Selva, Morgan L. Cable, Marie Ethvignot, Margaret Hansen, Andreas M. Hein, Elena-Sorina Lupu, Zachary Manchester, David Murrow, Chad Pozarycki, Pascal Spino, Amanda Stockton, Mathieu Choukroun, Soon-Jo Chung, John Day, Alexander Demagall, Anthony Freeman, Chloe Gentgen, Michel D. Ingham, Charity M. Phillips-Lander, Richard Rieber, Alejandro Salado, Maria Sakovsky, Lori R. Shiraishi, Yisong Yue , et al. (1 additional authors not shown)

Abstract: The surface and subsurface of worlds beyond Mars remain largely unexplored. Yet these worlds hold keys to fundamental questions in planetary science - from potentially habitable subsurface oceans on icy moons to ancient records preserved in Kuiper Belt objects. NASA's success in Mars exploration was achieved through incrementalism: 22 progressively sophisticated missions over decades. This paradig… ▽ More The surface and subsurface of worlds beyond Mars remain largely unexplored. Yet these worlds hold keys to fundamental questions in planetary science - from potentially habitable subsurface oceans on icy moons to ancient records preserved in Kuiper Belt objects. NASA's success in Mars exploration was achieved through incrementalism: 22 progressively sophisticated missions over decades. This paradigm, which we call Planetary Exploration 2.0 (PE 2.0), is untenable for the outer Solar System, where cruise times of a decade or more make iterative missions infeasible. We propose Planetary Exploration 3.0 (PE 3.0): a paradigm in which unvisited worlds are explored by a single or a few missions with radically adaptive space systems. A PE 3.0 mission conducts both initial exploratory science and follow-on hypothesis-driven science based on its own in situ data returns, evolving spacecraft capabilities to work resiliently in previously unseen environments. The key enabler of PE 3.0 is software-defined space systems (SDSSs) - systems that can adapt their functions at all levels through software updates. This paper presents findings from a Keck Institute for Space Studies (KISS) workshop on PE 3.0, covering: (1) PE 3.0 systems engineering including science definition, architecture, design methods, and verification & validation; (2) software-defined space system technologies including reconfigurable hardware, multi-functionality, and modularity; (3) onboard intelligence including autonomous science, navigation, controls, and embodied AI; and (4) three PE 3.0 mission concepts: a Neptune/Triton smart flyby, an ocean world explorer, and an Oort cloud reconnaissance mission. △ Less

Submitted 21 April, 2026; originally announced April 2026.

Journal ref: AIAA ASCEND 2026

Evolution of Lane-Changing Behavior in Mixed Traffic: A Quantum Game Theory Approach

Authors: Sungyong Chung, Tina Radvand, Alireza Talebpour

Abstract: As automated vehicles (AVs) enter mixed traffic, proactively anticipating the evolution of human driving behavior during critical interactions, such as lane changes, is essential. However, classical Evolutionary Game Theory (EGT) fails to capture the complexity of human decision-making during lane changes. Specifically, by strictly assuming independence between agents, classical models calibrated… ▽ More As automated vehicles (AVs) enter mixed traffic, proactively anticipating the evolution of human driving behavior during critical interactions, such as lane changes, is essential. However, classical Evolutionary Game Theory (EGT) fails to capture the complexity of human decision-making during lane changes. Specifically, by strictly assuming independence between agents, classical models calibrated on empirical payoffs predict a convergence to unrealistic full cooperation, contradicting the stable 42% cooperation rate observed in real-world data. To resolve this discrepancy, this study introduces a Quantum Game Theory (QGT) framework. We analyze 7,636 lane-changing interactions from the Waymo Open Motion Dataset (WOMD) to derive empirical payoff matrices via a Quantal Response Equilibrium (QRE) model. Utilizing the Marinatto-Weber (MW) quantization scheme, we introduce an entanglement parameter to mathematically embed latent correlations directly into the payoff structure of a single interaction. Our results identify a human entanglement parameter of $|b|^2_{HDV} \approx 0.52$ that accurately reproduces the observed mixed equilibrium. Furthermore, simulations of three AV deployment strategies (classical, entangled, and inverted) reveal that human adaptation depends critically on the underlying AV algorithm: while cooperative classical AVs maximize system-wide cooperation at high market penetration rates, defective inverted AVs paradoxically yield higher overall cooperation at low penetration rates by prompting more cooperative behaviors from human drivers. Consequently, rather than waiting for large scale deployment to observe these effects, stakeholders can utilize this framework to simulate repeated interactions and proactively anticipate how human driver behavior will evolve in response to specific AV software designs. △ Less

Submitted 16 April, 2026; originally announced April 2026.

Low-Stack HAETAE for Memory-Constrained Microcontrollers

Authors: Gustavo Banegas, Kim Youngbeom, Seo Seog Chung, Vredendaal Christine Van

Abstract: We present a low-stack implementation of the module-lattice signature scheme HAETAE, targeting microcontrollers with 8 kB-16 kB of available SRAM. On such devices, peak stack usage is often the binding constraint, and HAETAE's hyperball-based sampler, large transient polynomial vectors, and variable-length signature payloads (hint and high-bits arrays) pose a particular challenge. To address this… ▽ More We present a low-stack implementation of the module-lattice signature scheme HAETAE, targeting microcontrollers with 8 kB-16 kB of available SRAM. On such devices, peak stack usage is often the binding constraint, and HAETAE's hyperball-based sampler, large transient polynomial vectors, and variable-length signature payloads (hint and high-bits arrays) pose a particular challenge. To address this we introduce (i) Rejection-aware pass decomposition, which isolates encoding to the post-acceptance path; (ii) Component-level early rejection, which short-circuits the response computation when a partial norm already exceeds the bound; and (iii) Reverse-order streaming entropy coding using range Asymmetric Numeral Systems (rANS), which eliminates full hint and high-bits staging buffers. Combined with streamed matrix generation, a two-pass hyperball sampler with streaming Gaussian backend, and row-streamed verification, these techniques bring Signing stack from 71 kB-141 kB in the reference implementation down to 5.8 kB-6.0 kB, key generation to 4.7 kB-5.7 kB, and verification to 4.7 kB-4.8 kB across all three security levels. Our pure C implementation covers all three security levels (HAETAE-2/3/5), whose optimization paths differ due to the public-key domain (d>0 vs. d=0) and rejection structure. We implement our optimization on a Nucleo-L4R5ZI and compare to the reference pqm4 (for HAETAE-2 and -3) and a recently published memory-optimized implementation (targeting HAETAE-5 only). We reduce HAETAE-2, -3, and -5 stack by respectively 75, 86 and 8 % for key generation, 92, 95 and 24 % for signature generation, and 85, 91 and 22 % for verification. Depending on the parameter set, this impacts performance by at most a factor 1.8 and 3.4 for key and signature generation respectively, while even offering a performance improvement up to 18 % for verification. Verification at all security levels fits within 8 kB of RAM (signature buffer + stack) and is 2.34-3.34x faster than ML-DSA m4fstack at each comparable security level. We additionally validate portability under RIOT-OS on ARM Cortex-M4 and RISC-V targets. △ Less

Submitted 17 April, 2026; originally announced April 2026.

Perron-Frobenius Contractive Operator Matching for Data-Driven Reachable Fault Identification and Recovery

Authors: Joshua D. Ibrahim, Mahdi Taheri, Soon-Jo Chung, Fred Y. Hadaegh

Abstract: This paper focuses on data-driven fault detection, identification, and recovery (FDIR) for nonlinear control-affine systems under actuator faults. We create a unified framework in the space of probability densities, rather than on individual trajectories, using fault-indexed Perron--Frobenius (PF) operators to predict the evolution of state distributions under different fault profiles. By leveragi… ▽ More This paper focuses on data-driven fault detection, identification, and recovery (FDIR) for nonlinear control-affine systems under actuator faults. We create a unified framework in the space of probability densities, rather than on individual trajectories, using fault-indexed Perron--Frobenius (PF) operators to predict the evolution of state distributions under different fault profiles. By leveraging the probability-flow representation of the Fokker--Planck equation, we construct deterministic PF operators that reproduce exact stochastic marginals, define forward reachable density families, and establish certifiable 2-Wasserstein bounds on the divergence between fault-driven and nominal density evolutions. These provide quantitative conditions for the detectability and identifiability of various faults. The fault-indexed operators are learned from trajectory data via flow map matching (FMM), and we demonstrate that the observable FMM residual directly bounds the approximation error of the operator in the 2-Wasserstein metric. Additionally, we co-train a contraction certificate that bounds the gap between the learned operator family, the actual fault-driven density flow, and the nominal dynamics. The operator library is then used online for continuous fault parameter fitting over a continuous parameter space to generalize the learned operators to out-of-distribution (OOD) scenarios. To carry out the recovery control, we employ reachable density propagation and Gaussian mixture covariance steering. The proposed framework is validated on a 10-state spacecraft attitude-control system with four reaction wheels. △ Less

Submitted 16 April, 2026; originally announced April 2026.

Seeing Through Touch: Tactile-Driven Visual Localization of Material Regions

Authors: Seongyu Kim, Seungwoo Lee, Hyeonggon Ryu, Joon Son Chung, Arda Senocak

Abstract: We address the problem of tactile localization, where the goal is to identify image regions that share the same material properties as a tactile input. Existing visuo-tactile methods rely on global alignment and thus fail to capture the fine-grained local correspondences required for this task. The challenge is amplified by existing datasets, which predominantly contain close-up, low-diversity ima… ▽ More We address the problem of tactile localization, where the goal is to identify image regions that share the same material properties as a tactile input. Existing visuo-tactile methods rely on global alignment and thus fail to capture the fine-grained local correspondences required for this task. The challenge is amplified by existing datasets, which predominantly contain close-up, low-diversity images. We propose a model that learns local visuo-tactile alignment via dense cross-modal feature interactions, producing tactile saliency maps for touch-conditioned material segmentation. To overcome dataset constraints, we introduce: (i) in-the-wild multi-material scene images that expand visual diversity, and (ii) a material-diversity pairing strategy that aligns each tactile sample with visually varied yet tactilely consistent images, improving contextual localization and robustness to weak signals. We also construct two new tactile-grounded material segmentation datasets for quantitative evaluation. Experiments on both new and existing benchmarks show that our approach substantially outperforms prior visuo-tactile methods in tactile localization. △ Less

Submitted 13 April, 2026; originally announced April 2026.

Comments: CVPR 2026. Project page: https://mm.kaist.ac.kr/projects/SeeingThroughTouch/

Hard-constrained Physics-informed Neural Networks for Interface Problems

Authors: Seung Whan Chung, Stephen T. Castonguay, Sumanta Roy, Michael S. Penwarden, Yucheng Fu, Pratanu Roy

Abstract: Physics-informed neural networks (PINNs) have emerged as a flexible framework for solving partial differential equations, but their performance on interface problems remains challenging because continuity and flux conditions are typically imposed through soft penalty terms. The standard soft-constraint formulation leads to imperfect interface enforcement and degraded accuracy near interfaces. We i… ▽ More Physics-informed neural networks (PINNs) have emerged as a flexible framework for solving partial differential equations, but their performance on interface problems remains challenging because continuity and flux conditions are typically imposed through soft penalty terms. The standard soft-constraint formulation leads to imperfect interface enforcement and degraded accuracy near interfaces. We introduce two ansatz-based hard-constrained PINN formulations for interface problems that embed the interface physics into the solution representation and thereby decouple interface enforcement from PDE residual minimization. The first, termed the windowing approach, constructs the trial space from compactly supported windowed subnetworks so that interface continuity and flux balance are satisfied by design. The second, called the buffer approach, augments unrestricted subnetworks with auxiliary buffer functions that enforce boundary and interface constraints at discrete points through a lightweight correction. We study these formulations on one- and two-dimensional elliptic interface benchmarks and compare them with soft-constrained baselines. In one-dimensional problems, hard constraints consistently improve interface fidelity and remove the need for loss-weight tuning; the windowing approach attains very high accuracy (as low as $O(10^{-9})$) on simple structured cases, whereas the buffer approach remains accurate ($\sim O(10^{-5})$) across a wider range of source terms and interface configurations. In two dimensions, the buffer formulation is shown to be more robust because it enforces constraints through a discrete buffer correction, as the windowing construction becomes more sensitive to overlap and corner effects and over-constrains the problem. This positions the buffer method as a straightforward and geometrically flexible approach to complex interface problems. △ Less

Submitted 15 May, 2026; v1 submitted 9 April, 2026; originally announced April 2026.

Comments: 53 pages, 14 figures

Report number: 25-ERD-052, LLNL-JRNL-2010925 MSC Class: 68T07; 35J25

Candidate Microlensing Brown Dwarfs in Binary Lens Systems from the 2023--2025 Observing Seasons

Authors: Cheongho Han, Andrzej Udalski, Ian A. Bond, Chung-Uk Lee, Michael D. Albrow, Sun-Ju Chung, Andrew Gould, Youn Kil Jung, Kyu-Ha Hwang, Yoon-Hyun Ryu, Yossi Shvartzvald, In-Gu Shin, Jennifer C. Yee, Weicheng Zang, Hongjing Yang, Doeon Kim, Dong-Jin Kim, Seung-Lee Kim, Dong-Joo Lee, Sang-Mok Cha, Yongseok Lee, Byeong-Gon Park, Richard W. Pogge, Przemek Mróz, Michał K. Szymański , et al. (40 additional authors not shown)

Abstract: We present detailed light-curve analyses of ten binary-lens microlensing events observed during the 2023--2025 seasons and selected as candidates for hosting brown-dwarf companions. The sample includes OGLE-2023-BLG-0249, KMT-2023-BLG-1246, OGLE-2023-BLG-0079, KMT-2024-BLG-0072, KMT-2024-BLG-0897, KMT-2024-BLG-1876, KMT-2024-BLG-2379, KMT-2025-BLG-0922, KMT-2025-BLG-1056, and KMT-2025-BLG-2427. Fo… ▽ More We present detailed light-curve analyses of ten binary-lens microlensing events observed during the 2023--2025 seasons and selected as candidates for hosting brown-dwarf companions. The sample includes OGLE-2023-BLG-0249, KMT-2023-BLG-1246, OGLE-2023-BLG-0079, KMT-2024-BLG-0072, KMT-2024-BLG-0897, KMT-2024-BLG-1876, KMT-2024-BLG-2379, KMT-2025-BLG-0922, KMT-2025-BLG-1056, and KMT-2025-BLG-2427. For each event, we carry out modeling of the light curve, explore relevant degeneracies, and, when finite-source effects are present, determine the angular Einstein radius. For OGLE-2023-BLG-0249, we additionally measure the microlens parallax, which allows a direct determination of the lens masses and distance. For the remaining events, we estimate the physical lens properties via Bayesian analyses incorporating Galactic priors. The resulting posteriors show that the lens companions in all systems have median masses in the brown-dwarf regime, and the lenses of two events (KMT-2025-BLG-0922 and KMT-2025-BLG-1056) are consistent with binaries in which both lens components fall within the brown-dwarf mass range. Spanning a wide range of projected separations and distances, these detections illustrate the power of high-cadence microlensing surveys to build a census of brown-dwarf companions, including faint and distant systems beyond the reach of flux-limited methods. △ Less

Submitted 9 April, 2026; originally announced April 2026.

Comments: 17 pages, 12 figures