arXiv subjects

Jungho Cho

Publications and source records attributed to Jungho Cho.

7 recordsLinked to original sources

arXiv2026

Solar Open 2 Technical Report

We present Solar Open 2, a 250B-A15B Mixture-of-Experts language model built for long-horizon agentic tasks, scaled up from Solar Open 1 (Solar Open 100B). To hold entire agent trajectories in a single context, Solar Open 2 reaches a 1M-token window through a hybrid attention stack that interleaves one softmax layer among every three linear-attention layers, using no positional encoding and a gated delta rule extended to negative eigenvalues. To train at this scale under a fixed compute budget, we make training efficient in two ways: a stronger starting point, and higher-value data. For the starting point, we initialize Solar Open 2 from Solar Open 1, transferring the 5.69B-parameter shared skeleton that survives the architectural change and learning everything else through full pre-training. For the data, we curate for value per token: quality- and rarity-aware data curation and mixture-ratio optimization refine a 20T pool into a 10T mixture that, at equal token budget, outperforms the Solar Open 1 recipe. To build its agent skills, we train twelve domain specialists across purpose-built scenarios, then consolidate them into a single model by Multi-teacher On-Policy Distillation (MOPD). Against comparably sized open-weight models on English benchmarks, Solar Open 2 leads on MMLU-Pro, LiveCodeBench, and the APEX-Agents agentic suite, and stays competitive with the strongest (DeepSeek-V4-Flash and MiMo-V2.5) elsewhere. On Korean benchmarks, Solar Open 2 records the highest average of any model compared, including fast-tier closed APIs, and on Ko-GDPval, an in-house Korean officework-agent benchmark, it is competitive with DeepSeek-V4-Pro (1.6T) at less than a sixth of its size.

cs.CL↗

arXiv2026

A High-frequency Geodetic VLBI Experiment for Optical Clock Comparison

An intercontinental metrological clock comparison between Italy and the Republic of Korea was performed by means of geodetic K-band VLBI observations. The comparison involved the hydrogen masers (H-masers) used at Medicina and Sejong radio telescopes. The same clocks were simultaneously compared by a satellite link and by high-precision optical clocks maintained at the National Metrology Institutes, KRISS in Korea and INRIM in Italy, and delivered to VLBI antennas via optical fiber. The H-masers frequency difference was estimated by extrapolating the clock rate from VLBI data using two geodetic VLBI software. This was subsequently compared with clock differences derived by satellite link and by local optical clocks. Results obtained with different approaches were in agreement at the level of $10^{-15}$ s/s. This pilot study demonstrates that standard high-frequency (K-band) geodetic VLBI campaigns could be a viable approach to conduct intercontinental clock comparisons, now only possible via satellite links. This uncertainty can be reduced thanks to the planned installation of new-generation, broadband, high-frequency receivers on the involved telescopes. K/Q/W-band geodetic observations will allow an improvement of the accuracy of the resulting group delays through broad bandwidth synthesis from 20 to 100 GHz. Furthermore, the Frequency Phase Transfer (FPT) method will also be explored together with the use of PCAL systems installed at the radio telescopes to improve phase stability and thus allow a better estimation of the station clock parameters.

astro-ph.IM↗

arXiv2026

User-Oriented Multi-Turn Dialogue Generation with Tool Use at scale

The recent paradigm shift toward large reasoning models (LRMs) as autonomous agents has intensified the demand for sophisticated, multi-turn tool-use capabilities. Yet, existing datasets and data-generation approaches are limited by static, predefined toolsets that cannot scale to the complexity of open-ended human-agent collaboration. To address this, we initially developed a framework for automated task-oriented multi-turn dialogue generation at scale, utilizing an LRM-based simulator to dynamically generate high-value, domain-specific tools to solve specified tasks. However, we observe that a purely task-oriented design often results in "solely task-solving" trajectories, where the agent completes the objective with minimal interaction, failing to generate the high turn-count conversations seen in realistic scenarios. To bridge this gap, we shift toward a user-oriented simulation paradigm. By decoupling task generation from a dedicated user simulator that mimics human behavioral rules - such as incremental request-making and turn-by-turn feedback - we facilitate more authentic, extended multi-turn dialogues that reflect the iterative nature of real-world problem solving. Our generation pipeline operates as a versatile, plug-and-play module capable of initiating generation from any state, ensuring high scalability in producing extended tool-use data. Furthermore, by facilitating multiple task completions within a single trajectory, it yields a high-density dataset that reflects the multifaceted demands of real-world human-agent interaction.

cs.CL↗

arXiv2026

Solar Open Technical Report

We introduce Solar Open, a 102B-parameter bilingual Mixture-of-Experts language model for underserved languages. Solar Open demonstrates a systematic methodology for building competitive LLMs by addressing three interconnected challenges. First, to train effectively despite data scarcity for underserved languages, we synthesize 4.5T tokens of high-quality, domain-specific, and RL-oriented data. Second, we coordinate this data through a progressive curriculum jointly optimizing composition, quality thresholds, and domain coverage across 20 trillion tokens. Third, to enable reasoning capabilities through scalable RL, we apply our proposed framework SnapPO for efficient optimization. Across benchmarks in English and Korean, Solar Open achieves competitive performance, demonstrating the effectiveness of this methodology for underserved language AI development.

cs.CL↗

arXiv2025

System Message Generation for User Preferences using Open-Source Models

System messages play a crucial role in interactions with large language models (LLMs), often serving as prompts to initiate conversations. Through system messages, users can assign specific roles, perform intended tasks, incorporate background information, and specify various output formats and communication styles. Despite such versatility, publicly available datasets often lack system messages and are subject to strict license constraints in industrial applications. Moreover, manually annotating system messages that align with user instructions is resource-intensive. In light of these challenges, we introduce SysGen, a pipeline for generating system messages that better align assistant responses with user instructions using existing supervised fine-tuning datasets that lack system messages. Training open-source models on SysGen data yields substantial improvements in both single-turn (Multifacet) and multi-turn (SysBench) conversation benchmarks. Notably, our method shows strong gains in shorter conversations, suggesting that it enhances early-stage interaction effectiveness. Our qualitative analysis further emphasizes the value of diverse and structured system messages in improving LLM adaptability across varied user scenarios.

cs.CL↗

arXiv2024

A Geodetic and Astrometric VLBI Experiment at 22/43/88/132 GHz

Extending geodetic and astrometric Very Long Baseline Interferometry (VLBI) observations from traditional centimeter wavebands to millimeter wavebands offers numerous scientific potentials and benefits. However, it was considered quite challenging due to various factors, including the increased effects of atmospheric opacity and turbulence at millimeter wavelengths. Here, we present the results of the first geodetic-mode VLBI experiment, simultaneously observing 82 sources at 22/43/88/132 GHz (K/Q/W/D bands) using the Korean VLBI Network (KVN). We introduced the frequency phase transfer (FPT) method to geodetic VLBI analysis, an approach for calibrating atmospheric phase fluctuations at higher frequencies by transferring phase solutions from lower frequencies. With a 2-minute scan, FPT improved the signal-to-noise ratio (SNR) of most fringes, some by over 100%, thereby enhancing the detection rate of weak sources at millimeter wavebands. Additionally, FPT reduced systematic errors in group delay and delay rate, with the weighted root-mean-squares (WRMS) of the post-fitting residuals decreasing from 25.0 ps to 20.5 ps at the W band and from 39.3 ps to 27.6 ps at the D band. There were no notable differences observed in calibrating atmospheric phase fluctuations at the K band (WRMS = 12.4 ps) and Q band (WRMS = 11.8 ps) with the KVN baselines. This experiment demonstrated that the millimeter waveband can be used for geodetic and astrometric applications with high precision.

astro-ph.IM↗

arXiv2023

Measurement-based Close-in Path Loss Modeling with Diffraction for Rural Long-distance Communications

In this letter, we investigate rural large-scale path loss models based on the measurements in a central area of South Korea (rural area) in spring. In particular, we develop new close-in (CI) path loss models incorporating a diffraction component. The transmitter used in the measurement system is located on a hill and utilizes omnidirectional antennas operating at 1400 and 2250 MHz frequencies. The receiver is also equipped with omnidirectional antennas and measures at positions totaling 3,858 (1,262 positions for LOS and 2,596 positions for NLOS) and 4,957 (1,427 positions for LOS and 3,530 positions for NLOS) for 1400 and 2250 MHz, respectively. This research demonstrates that the newly developed CI path loss models incorporating a diffraction component significantly reduce standard deviations (STD) and are independent of frequency, especially for LOS beyond the first meter of propagation, making them suitable for use with frequencies up to a millimeter-wave.

eess.SP↗