arXiv subjects

Po-Chun Hsu

Publications and source records attributed to Po-Chun Hsu.

17 recordsLinked to original sources

arXiv2026

Remote epitaxy beyond polarity

Remote epitaxy through a monolayer two-dimensional material-covered substrate establishes a crystallographic registry across the van der Waals (vdW) surface that enables the epitaxial growth, lift-off and transfer of single-crystalline films. A central belief in remote epitaxy is that the substrate facilitating the phenomenon must be a material with strong ionicity, as the interatomic electrostatic potential fluctuation in covalent and metallic materials is substantially attenuated by two-dimensional materials. Here, we show remote epitaxy is possible when the substrate is a metallic or covalently bonded material and experimentally demonstrate non-polar remote homo- and heteroepitaxy across a wide range of material systems, including both metals and semiconductors. The achieved non-polar remote interactions are designed and engineered by harnessing substrate conductivity and vicinal surface step-edge density. These findings indicate that remote epitaxy is universal and applicable to ionic, metallic, and covalent materials, expanding its capabilities and stimulating a plethora of new fundamental scientific questions about the mechanism of remote epitaxy.

cond-mat.mtrl-sci↗

arXiv2026

Taiwan Safety Benchmark and Breeze Guard: Toward Trustworthy AI for Taiwanese Mandarin

Global safety models exhibit strong performance across widely used benchmarks, yet their training data rarely captures the cultural and linguistic nuances of Taiwanese Mandarin. This limitation results in systematic blind spots when interpreting region-specific risks such as localized financial scams, culturally embedded hate speech, and misinformation patterns. To address these gaps, we introduce TS-Bench (Taiwan Safety Benchmark), a standardized evaluation suite for assessing safety performance in Taiwanese Mandarin. TS-Bench contains 400 human-curated prompts spanning critical domains including financial fraud, medical misinformation, social discrimination, and political manipulation. In parallel, we present Breeze Guard, an 8B safety model derived from Breeze 2, our previously released general-purpose Taiwanese Mandarin LLM with strong cultural grounding from its original pre-training corpus. Breeze Guard is obtained through supervised fine-tuning on a large-scale, human-verified synthesized dataset targeting Taiwan-specific harms. Our central hypothesis is that effective safety detection requires the cultural grounding already present in the base model; safety fine-tuning alone is insufficient to introduce new socio linguistic knowledge from scratch. Empirically, Breeze Guard significantly outperforms the leading 8B general-purpose safety model, Granite Guardian 3.3, on TS-Bench (+0.17 overall F1), with particularly large gains in high-context categories such as scam (+0.66 F1) and financial malpractice (+0.43 F1). While the model shows slightly lower performance on English-centric benchmarks (ToxicChat, AegisSafetyTest), this tradeoff is expected for a regionally specialized safety model optimized for Taiwanese Mandarin. Together, Breeze Guard and TS-Bench establish a new foundation for trustworthy AI deployment in Taiwan.

cs.CL↗

arXiv2026

Breeze Taigi: Benchmarks and Models for Taiwanese Hokkien Speech Recognition and Synthesis

Taiwanese Hokkien (Taigi) presents unique opportunities for advancing speech technology methodologies that can generalize to diverse linguistic contexts. We introduce Breeze Taigi, a comprehensive framework centered on standardized benchmarks for evaluating Taigi speech recognition and synthesis systems. Our primary contribution is a reproducible evaluation methodology that leverages parallel Taiwanese Mandarin resources. We provide 30 carefully curated Mandarin-Taigi audio pairs from Taiwan's Executive Yuan public service announcements with normalized ground truth transcriptions. We establish Character Error Rate (CER) as the standard metric and implement normalization procedures to enable fair cross-system comparisons. To demonstrate the benchmark's utility and provide reference implementations, we develop speech recognition and synthesis models through a methodology that leverages existing Taiwanese Mandarin resources and large-scale synthetic data generation. In particular, we fine-tune a Whisper model on approximately 10,000 hours of Taigi synthetic speech data. Our ASR model achieves 30.13% average CER on the benchmark, outperforming existing commercial and research systems. By providing standardized evaluation protocols, diverse training datasets, and open baseline models, we offer a replicable framework with methodologies applicable to various linguistic contexts.

cs.CL↗

arXiv2025

Spectrally-selective dynamic radiative thermoregulation via phase engineering

Maintaining comfortable temperatures for buildings, humans, and devices consumes a substantial portion of global energy, underscoring the urgent need for energy-efficient thermoregulation technologies. Dynamic radiative thermal emitters that can switch between passive cooling and heating modes offer a promising solution, but most existing devices exhibit broadband optical responses, resulting in unwanted parasitic heat exchange and limited performance. Here, we introduce an elegant strategy that uses a dielectric cap to transform broadband metal-insulator transition (MIT) materials into spectrally selective dynamic emitters. This design creates a highly tunable Fabry-Perot cavity, enabling a tailored thermal emission spectrum by engineering the reflected-wave phase profile. Our Fresnel-formalism-based phasor diagram analysis reveals two key routes for realizing high spectral selectivity: a high-index dielectric cap and a low-loss metallic MIT state, which are further validated by Bayesian optimization. Following this principle, we demonstrated a wide-angle spectrally-selective thermoregulator operating in the atmospheric transparency window (8-13 um), where the thermal emittance can be electrically tuned from about 0.2 to 0.9 through reversible copper electrodeposition on a germanium cavity. Furthermore, this strategy can be extended to multispectral electrochromic windows, enabling switching between solar heating and spectrally-selective radiative cooling. Our work establishes a versatile and generalizable paradigm for spectral engineering of dynamic thermal emitters, opening opportunities in energy-efficient buildings, wearable thermal comfort, spacecraft thermoregulation, and multispectral camouflage.

physics.optics↗

arXiv2025

Epitaxial Electrodeposition of Fe with Controlled In-Plane Variants for Reversible Metal Anode in Aqueous Electrolyte

The development of reversible metal anodes is a key challenge for advancing aqueous battery technologies, particularly for scalable and safe stationary energy storage applications. Here we demonstrate a strategy to realize epitaxial electrodeposition of iron (Fe) on single-crystal copper (Cu) substrates in aqueous electrolytes. We compare the electrodeposition behavior of Fe on polycrystalline and single-crystalline Cu substrates, revealing that the latter enables highly uniform, dense, and crystallographically aligned Fe growth. Comprehensive electron backscatter diffraction (EBSD) and X-ray diffraction (XRD) analysis confirms the formation of Fe with specific out-of-plane and in-plane orientations, including well-defined rotational variants. Our findings highlight that epitaxial electrodeposition of Fe can suppress dendritic growth and significantly enhance Coulombic efficiency during plating/stripping cycles. This approach bridges fundamental crystallography with practical electrochemical performance, providing a pathway toward high-efficiency aqueous batteries utilizing Earth-abundant materials.

cond-mat.mtrl-sci↗

arXiv2025

High-efficiency broadband active metasurfaces via reversible metal electrodeposition

Realizing active metasurfaces with substantial tunability is important for many applications but remains challenging due to difficulties in dynamically tuning light-matter interactions at subwavelength scales. Here, we introduce reversible metal electrodeposition as a versatile approach for enabling active metasurfaces with exceptional tunability across a broad bandwidth. As a proof of concept, we demonstrate a dynamic beam-steering device by performing reversible copper (Cu) electrodeposition on a reflective gradient metasurface composed of metal-insulator-metal resonators. By applying different voltages, the Cu atoms can be uniformly and reversibly electrodeposited and stripped around the resonators, effectively controlling the gap-surface plasmon resonances and steering the reflected light. This process experimentally achieved >90% diffraction efficiencies and >60% reflection efficiencies in both specular and anomalous modes, even after thousands of cycles. Moreover, these high efficiencies can be extended from the visible to the near- and mid-infrared regimes, demonstrating the broad versatility of this approach in enabling various active optical and thermal devices with different working wavelengths and bandwidths.

physics.optics↗

arXiv2025

The Breeze 2 Herd of Models: Traditional Chinese LLMs Based on Llama with Vision-Aware and Function-Calling Capabilities

Llama-Breeze2 (hereinafter referred to as Breeze2) is a suite of advanced multi-modal language models, available in 3B and 8B parameter configurations, specifically designed to enhance Traditional Chinese language representation. Building upon the Llama 3.2 model family, we continue the pre-training of Breeze2 on an extensive corpus to enhance the linguistic and cultural heritage of Traditional Chinese. In addition to language modeling capabilities, we significantly augment the models with function calling and vision understanding capabilities. At the time of this publication, as far as we are aware, absent reasoning-inducing prompts, Breeze2 are the strongest performing models in Traditional Chinese function calling and image understanding in its size class. The effectiveness of Breeze2 is benchmarked across various tasks, including Taiwan general knowledge, instruction-following, long context, function calling, and vision understanding. We are publicly releasing all Breeze2 models under the Llama 3.2 Community License. We also showcase the capabilities of the model running on mobile platform with a mobile application which we also open source.

cs.CL↗

arXiv2024

Enhancing Function-Calling Capabilities in LLMs: Strategies for Prompt Formats, Data Integration, and Multilingual Translation

Large language models (LLMs) have significantly advanced autonomous agents, particularly in zero-shot tool usage, also known as function calling. This research delves into enhancing the function-calling capabilities of LLMs by exploring different approaches, including prompt formats for integrating function descriptions, blending function-calling and instruction-following data, introducing a novel Decision Token for conditional prompts, leveraging chain-of-thought reasoning, and overcoming multilingual challenges with a translation pipeline. Our key findings and contributions are as follows: (1) Instruction-following data improves both function-calling accuracy and relevance detection. (2) The use of the newly proposed Decision Token, combined with synthetic non-function-call data, enhances relevance detection. (3) A tailored translation pipeline effectively overcomes multilingual limitations, demonstrating significant improvements in Traditional Chinese. These insights highlight the potential for improved function-calling capabilities and multilingual applications in LLMs.

cs.CL↗

arXiv2024

RAD-Bench: Evaluating Large Language Models Capabilities in Retrieval Augmented Dialogues

In real-world applications with Large Language Models (LLMs), external retrieval mechanisms - such as Search-Augmented Generation (SAG), tool utilization, and Retrieval-Augmented Generation (RAG) - are often employed to enhance the quality of augmented generations in dialogues. These approaches often come with multi-turn dialogue, where each interaction is enriched by relevant information retrieved from external sources. Existing benchmarks either assess LLMs' chat abilities in multi-turn dialogues or their use of retrieval for augmented responses in single-turn settings. However, there is a gap in evaluating LLMs' ability to leverage retrieval for more precise responses across multiple turns. To address this limitation, we introduce RAD-Bench (Retrieval Augmented Dialogue), a benchmark designed to evaluate LLMs' capabilities in multi-turn dialogues following retrievals, essential for their deployment in context-rich applications. RAD-Bench evaluates two key abilities of LLMs: Retrieval Synthesis and Retrieval Reasoning. These are measured using discriminative questions and retrieved contexts, and corresponding reference answers, assessing how effectively LLMs integrate and reason with context to maintain and enhance conversation quality over multiple turns. Our evaluation results on commonly used LLMs reveal that model performance deteriorates as additional layers of conditions or constraints are applied across conversation turns, even when accurate retrieved contexts are provided. The data and code are available at https://github.com/mtkresearch/RAD-Bench

cs.CL↗

arXiv2024

Let's Fuse Step by Step: A Generative Fusion Decoding Algorithm with LLMs for Robust and Instruction-Aware ASR and OCR

We propose "Generative Fusion Decoding" (GFD), a novel shallow fusion framework designed to integrate large language models (LLMs) into cross-modal text recognition systems for automatic speech recognition (ASR) and optical character recognition (OCR). We derive the necessary formulations to enable GFD to operate across mismatched token spaces of different models by calculating likelihood at the byte level, thereby enabling seamless fusion and synchronous progression during the decoding process. GFD is plug-and-play by design, making it readily compatible with various auto-regressive models without the need for any re-training. GFD proves effective for general ASR and OCR tasks through intermediate and frequent interactions with LLMs, surpassing cascaded methods in English and Mandarin benchmarks. In addition, GFD transfers in-context learning abilities of LLMs and allows for adaptive ASR in instruction-aware and long-context settings, yielding significant WER reductions of up to 17.7\%.

cs.CL↗

arXiv2024

Breeze-7B Technical Report

Breeze-7B is an open-source language model based on Mistral-7B, designed to address the need for improved language comprehension and chatbot-oriented capabilities in Traditional Chinese. This technical report provides an overview of the additional pretraining, finetuning, and evaluation stages for the Breeze-7B model. The Breeze-7B family of base and chat models exhibits good performance on language comprehension and chatbot-oriented tasks, reaching the top in several benchmarks among models comparable in its complexity class.

cs.CL↗

arXiv2023

Advancing the Evaluation of Traditional Chinese Language Models: Towards a Comprehensive Benchmark Suite

The evaluation of large language models is an essential task in the field of language understanding and generation. As language models continue to advance, the need for effective benchmarks to assess their performance has become imperative. In the context of Traditional Chinese, there is a scarcity of comprehensive and diverse benchmarks to evaluate the capabilities of language models, despite the existence of certain benchmarks such as DRCD, TTQA, CMDQA, and FGC dataset. To address this gap, we propose a novel set of benchmarks that leverage existing English datasets and are tailored to evaluate language models in Traditional Chinese. These benchmarks encompass a wide range of tasks, including contextual question-answering, summarization, classification, and table understanding. The proposed benchmarks offer a comprehensive evaluation framework, enabling the assessment of language models' capabilities across different tasks. In this paper, we evaluate the performance of GPT-3.5, Taiwan-LLaMa-v1.0, and Model 7-C, our proprietary model, on these benchmarks. The evaluation results highlight that our model, Model 7-C, achieves performance comparable to GPT-3.5 with respect to a part of the evaluated capabilities. In an effort to advance the evaluation of language models in Traditional Chinese and stimulate further research in this field, we have open-sourced our benchmark and opened the model for trial.

cs.CL↗

arXiv2023

Extending the Pre-Training of BLOOM for Improved Support of Traditional Chinese: Models, Methods and Results

In this paper we present the multilingual language model BLOOM-zh that features enhanced support for Traditional Chinese. BLOOM-zh has its origins in the open-source BLOOM models presented by BigScience in 2022. Starting from released models, we extended the pre-training of BLOOM by additional 7.4 billion tokens in Traditional Chinese and English covering a variety of domains such as news articles, books, encyclopedias, educational materials as well as spoken language. In order to show the properties of BLOOM-zh, both existing and newly created benchmark scenarios are used for evaluating the performance. BLOOM-zh outperforms its predecessor on most Traditional Chinese benchmarks while maintaining its English capability. We release all our models to the research community.

cs.CL↗

arXiv2023

Federated Deep Reinforcement Learning for THz-Beam Search with Limited CSI

Terahertz (THz) communication with ultra-wide available spectrum is a promising technique that can achieve the stringent requirement of high data rate in the next-generation wireless networks, yet its severe propagation attenuation significantly hinders its implementation in practice. Finding beam directions for a large-scale antenna array to effectively overcome severe propagation attenuation of THz signals is a pressing need. This paper proposes a novel approach of federated deep reinforcement learning (FDRL) to swiftly perform THz-beam search for multiple base stations (BSs) coordinated by an edge server in a cellular network. All the BSs conduct deep deterministic policy gradient (DDPG)-based DRL to obtain THz beamforming policy with limited channel state information (CSI). They update their DDPG models with hidden information in order to mitigate inter-cell interference. We demonstrate that the cell network can achieve higher throughput as more THz CSI and hidden neurons of DDPG are adopted. We also show that FDRL with partial model update is able to nearly achieve the same performance of FDRL with full model update, which indicates an effective means to reduce communication load between the edge server and the BSs by partial model uploading. Moreover, the proposed FDRL outperforms conventional non-learning-based and existing non-FDRL benchmark optimization methods.

cs.AI↗

arXiv2022

STOP: A dataset for Spoken Task Oriented Semantic Parsing

End-to-end spoken language understanding (SLU) predicts intent directly from audio using a single model. It promises to improve the performance of assistant systems by leveraging acoustic information lost in the intermediate textual representation and preventing cascading errors from Automatic Speech Recognition (ASR). Further, having one unified model has efficiency advantages when deploying assistant systems on-device. However, the limited number of public audio datasets with semantic parse labels hinders the research progress in this area. In this paper, we release the Spoken Task-Oriented semantic Parsing (STOP) dataset, the largest and most complex SLU dataset to be publicly available. Additionally, we define low-resource splits to establish a benchmark for improving SLU when limited labeled data is available. Furthermore, in addition to the human-recorded audio, we are releasing a TTS-generated version to benchmark the performance for low-resource domain adaptation of end-to-end SLU systems. Initial experimentation show end-to-end SLU models performing slightly worse than their cascaded counterparts, which we hope encourages future work in this direction.

cs.CL↗

arXiv2021

Bio-inspired vascularized electrodes for high-performance fast-charging batteries designed by deep learning

Slow ionic transport and high voltage drop (IR drop) of homogeneous porous electrodes are the critical causes of severe performance degradation of lithium-ion (Li-ion) batteries under high charging rates. Herein, we demonstrate that a bio-inspired vascularized porous electrode can simultaneously solve these two problems by introducing low tortuous channels and graded porosity. To optimize the vasculature structural parameters, we employ artificial neural networks (ANNs) to accelerate the computation of possible structures with high accuracy. Furthermore, an inverse-design searching library is compiled to find the optimal vascular structures under different industrial fabrication and design criteria. The prototype delivers a customizable package containing optimal geometric parameters and their uncertainty and sensitivity analysis. Finally, the full-vascularized cell shows a 66% improvement of charging capacity than the traditional homogeneous cell under 3.2C current density. This research provides an innovative methodology to solve the fast-charging problem in batteries and broaden the applicability of deep learning algorithm to different scientific or engineering areas.

cond-mat.mtrl-sci↗

arXiv2021

A triple-mode mid-infrared modulator for all-surface radiative thermal management

Thermal management is ubiquitous in the modern world and indispensable for a sustainable future. Radiative heat management provides unique advantages because the heat transfer can be controlled by the surface. However, different surface emissivities require different tuning strategies. Here, we demonstrate a triple-mode mid-infrared modulator that can switch between passive heating and cooling suitable for all types of object surface emissivities. The device is composed of a surface-textured infrared-semi-transparent elastomer coated with a metallic back reflector, which is biaxially strained to sequentially achieve three fundamental modes: emission, reflection, and transmission. By analyzing and optimizing the coupling between optical and mechanical properties, we achieve a performance of emittance contrast = 0.58, transmittance contrast = 0.49, and reflectance contrast = 0.39. The device can provide a new design paradigm of radiation heat regulation to develop the next generation wearable devices, robotics, and camouflage technology.

physics.app-ph↗