All-Polymer and Adhesive Electrodes as Electroceuticals for Chronic Wound Healing
Jiwan Jeon, Jin‐Hoon Kim, Yohan Kim, Hyun Roh, Tae Jin Kang, Young Chul Suh, Woosung Lee, Jin‐Woo Park
IF 9.1
Nano Letters
Chronic wounds are complicated by diabetes, surgical complications, and vascular dysfunction, which delay healing and impair cellular repair. Conventional therapies are costly and may injure surrounding tissue. Electroceutical stimulation promotes cell migration and proliferation, yet existing systems show weak adhesion and low efficiency and restrict patient movement. We present a wearable electroceutical system using polymer electrodes to address these limitations. Relative to metals, the polymers provide higher charge injection and storage and strong adhesion for stable stimulation. We characterized electrode behavior under pulsed currents that mimic endogenous fields and simulated the relationship between electrode dimensions and wound size. Electrochemical impedance and chronoamperometry quantified charge transfer under clinically relevant waveforms. To enable mobility, we built a portable module that delivers controlled pulses. Cell-based assays and an animal wound model confirmed biocompatibility and improved closure rates, demonstrating accelerated healing without invasive procedures.
The effect of circadian gene, Per2, on JAK-STAT pathway in inflammation model in vitro 3581
Tae Jin Kang, Venkata Prakash Annamneedi, Jun Woo Park
IF 3.4
The Journal of Immunology
Abstract Description Per2 is core circadian gene to generate circadian rhythms and it is associated with enhancing inflammation. JAK-STAT pathway is a classical signal transduction pathway in keratinocytes to initiates skin inflammation through releasing cytokines and chemokines which leads atopic dermatitis. In this study we investigated the role of clock gene Per2 in skin inflammation using HaCaT cells transfected with siPer2 RNA. The transfected and un-transfected HaCaT cells were stimulated with 25 ng of TNF? and IFN?? each for 6 hours. Cell supernatant, total mRNA, and protein lysate were collected to measure the expression of cytokines, chemokines, and transcription factor STAT1. The inflammatory cytokines such as IL-6 and IL-8 has enhance skin inflammation by chemotaxis. There is a big significantly difference with the level of cytokines such as IL-6 and IL-8, and the level of chemokines such as MRC, TARC in between un-transfected and transfected HaCaT cells. Transfected HaCaT cells released low levels cytokines and chemokines where un-transfected HaCaT cells releasing higher levels with cytokines and chemokines. The expression of transcription factor STAT1 which responsible inflammation showed weak in transfected HaCaT cells. According to these results we hypothesized that the increased expression of Per2 gene results enhancing inflammation in HaCaT cells through upregulation of STAT1 expression, which induce skin inflammation in HaCaT cells by releasing cytokines and chemokines. Funding Sources This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government (MSIT) (No. 2022R1F1A1072616). Topic Categories Innate Immune Responses and Host Defense: Molecular Mechanisms (INM)
All-Polymer and Adhesive Electrodes as Electroceuticals for Chronic Wound Healing
Jiwan Jeon, Jin‐Hoon Kim, Yohan Kim, Hyun Roh, Tae Jin Kang, Young Chul Suh, Woosung Lee, Jin‐Woo Park
IF 9.1
Nano Letters
Chronic wounds are complicated by diabetes, surgical complications, and vascular dysfunction, which delay healing and impair cellular repair. Conventional therapies are costly and may injure surrounding tissue. Electroceutical stimulation promotes cell migration and proliferation, yet existing systems show weak adhesion and low efficiency and restrict patient movement. We present a wearable electroceutical system using polymer electrodes to address these limitations. Relative to metals, the polymers provide higher charge injection and storage and strong adhesion for stable stimulation. We characterized electrode behavior under pulsed currents that mimic endogenous fields and simulated the relationship between electrode dimensions and wound size. Electrochemical impedance and chronoamperometry quantified charge transfer under clinically relevant waveforms. To enable mobility, we built a portable module that delivers controlled pulses. Cell-based assays and an animal wound model confirmed biocompatibility and improved closure rates, demonstrating accelerated healing without invasive procedures.
The effect of circadian gene, Per2, on JAK-STAT pathway in inflammation model in vitro 3581
Tae Jin Kang, Venkata Prakash Annamneedi, Jun Woo Park
IF 3.4
The Journal of Immunology
Abstract Description Per2 is core circadian gene to generate circadian rhythms and it is associated with enhancing inflammation. JAK-STAT pathway is a classical signal transduction pathway in keratinocytes to initiates skin inflammation through releasing cytokines and chemokines which leads atopic dermatitis. In this study we investigated the role of clock gene Per2 in skin inflammation using HaCaT cells transfected with siPer2 RNA. The transfected and un-transfected HaCaT cells were stimulated with 25 ng of TNF? and IFN?? each for 6 hours. Cell supernatant, total mRNA, and protein lysate were collected to measure the expression of cytokines, chemokines, and transcription factor STAT1. The inflammatory cytokines such as IL-6 and IL-8 has enhance skin inflammation by chemotaxis. There is a big significantly difference with the level of cytokines such as IL-6 and IL-8, and the level of chemokines such as MRC, TARC in between un-transfected and transfected HaCaT cells. Transfected HaCaT cells released low levels cytokines and chemokines where un-transfected HaCaT cells releasing higher levels with cytokines and chemokines. The expression of transcription factor STAT1 which responsible inflammation showed weak in transfected HaCaT cells. According to these results we hypothesized that the increased expression of Per2 gene results enhancing inflammation in HaCaT cells through upregulation of STAT1 expression, which induce skin inflammation in HaCaT cells by releasing cytokines and chemokines. Funding Sources This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government (MSIT) (No. 2022R1F1A1072616). Topic Categories Innate Immune Responses and Host Defense: Molecular Mechanisms (INM)
Judging Against the Reference: Uncovering Knowledge-Driven Failures in LLM-Judges on QA Evaluation
Dongryeol Lee, Yerin Hwang, Tae Jin Kang, Minwoo Lee, Younhyung Chae, Kyomin Jung
ArXiv.org
While large language models (LLMs) are increasingly used as automatic judges for question answering (QA) and other reference-conditioned evaluation tasks, little is known about their ability to adhere to a provided reference. We identify a critical failure mode of such reference-based LLM QA evaluation: when the provided reference conflicts with the judge model's parametric knowledge, the resulting scores become unreliable, substantially degrading evaluation fidelity. To study this phenomenon systematically, we introduce a controlled swapped-reference QA framework that induces reference-belief conflicts. Specifically, we replace the reference answer with an incorrect entity and construct diverse pairings of original and swapped references with correspondingly aligned candidate answers. Surprisingly, grading reliability drops sharply under swapped references across a broad set of judge models. We empirically show that this vulnerability is driven by judges' over-reliance on parametric knowledge, leading judges to disregard the given reference under conflict. Finally, we find that this failure persists under common prompt-based mitigation strategies, highlighting a fundamental limitation of LLM-as-a-judge evaluation and motivating reference-based protocols that enforce stronger adherence to the provided reference.
Don’t Judge Code by Its Cover: Exploring Biases in LLM Judges for Code Evaluation
Jiwon Moon, Yerin Hwang, Dongryeol Lee, Tae Jin Kang, Yongil Kim, Kyomin Jung
With the growing use of large language models (LLMs) as evaluators, their application has expanded to code evaluation tasks, where they assess the correctness of generated code without relying on reference implementations.While this offers scalability and flexibility, it also raises a critical, unresolved question: Can LLM judges fairly and robustly evaluate semantically equivalent code with superficial variations?Functionally correct code often exhibits variations-such as differences in variable names, comments, or formatting-that should not influence its correctness.Yet, whether LLM judges can reliably handle these variations remains unclear.We present the first comprehensive study of this issue, defining six types of potential bias in code evaluation and revealing their systematic impact on LLM judges.Across five programming languages and multiple LLMs, we empirically demonstrate that all tested LLM judges are susceptible to both positive and negative biases, resulting in inflated or unfairly low scores.Moreover, we observe that LLM judges remain vulnerable to these biases even when prompted to generate test cases before scoring, highlighting the need for more robust code evaluation methods.
When Wording Steers the Evaluation: Framing Bias in LLM judges
Yerin Hwang, Dongryeol Lee, Tae Jin Kang, Minwoo Lee, Kyomin Jung
arXiv (Cornell University)
Large language models (LLMs) are known to produce varying responses depending on prompt phrasing, indicating that subtle guidance in phrasing can steer their answers. However, the impact of this framing bias on LLM-based evaluation, where models are expected to make stable and impartial judgments, remains largely underexplored. Drawing inspiration from the framing effect in psychology, we systematically investigate how deliberate prompt framing skews model judgments across four high-stakes evaluation tasks. We design symmetric prompts using predicate-positive and predicate-negative constructions and demonstrate that such framing induces significant discrepancies in model outputs. Across 14 LLM judges, we observe clear susceptibility to framing, with model families showing distinct tendencies toward agreement or rejection. These findings suggest that framing bias is a structural property of current LLM-based evaluation systems, underscoring the need for framing-aware protocols.
Judging Against the Reference: Uncovering Knowledge-Driven Failures in LLM-Judges on QA Evaluation
Dongryeol Lee, Yerin Hwang, Tae Jin Kang, Minwoo Lee, Younhyung Chae, Kyomin Jung
arXiv (Cornell University)
While large language models (LLMs) are increasingly used as automatic judges for question answering (QA) and other reference-conditioned evaluation tasks, little is known about their ability to adhere to a provided reference. We identify a critical failure mode of such reference-based LLM QA evaluation: when the provided reference conflicts with the judge model's parametric knowledge, the resulting scores become unreliable, substantially degrading evaluation fidelity. To study this phenomenon systematically, we introduce a controlled swapped-reference QA framework that induces reference-belief conflicts. Specifically, we replace the reference answer with an incorrect entity and construct diverse pairings of original and swapped references with correspondingly aligned candidate answers. Surprisingly, grading reliability drops sharply under swapped references across a broad set of judge models. We empirically show that this vulnerability is driven by judges' over-reliance on parametric knowledge, leading judges to disregard the given reference under conflict. Finally, we find that this failure persists under common prompt-based mitigation strategies, highlighting a fundamental limitation of LLM-as-a-judge evaluation and motivating reference-based protocols that enforce stronger adherence to the provided reference.
When Wording Steers the Evaluation: Framing Bias in LLM judges
Yerin Hwang, Dongryeol Lee, Tae Jin Kang, Minwoo Lee, Kyomin Jung
ArXiv.org
Large language models (LLMs) are known to produce varying responses depending on prompt phrasing, indicating that subtle guidance in phrasing can steer their answers. However, the impact of this framing bias on LLM-based evaluation, where models are expected to make stable and impartial judgments, remains largely underexplored. Drawing inspiration from the framing effect in psychology, we systematically investigate how deliberate prompt framing skews model judgments across four high-stakes evaluation tasks. We design symmetric prompts using predicate-positive and predicate-negative constructions and demonstrate that such framing induces significant discrepancies in model outputs. Across 14 LLM judges, we observe clear susceptibility to framing, with model families showing distinct tendencies toward agreement or rejection. These findings suggest that framing bias is a structural property of current LLM-based evaluation systems, underscoring the need for framing-aware protocols.
SiO <sub>2</sub> nanoparticle-coated Ti-6Al-4V spherical powder for powder bed fusion additive manufacturing process
J H Lee, Tae Jin Kang, Ukju Gim, Se‐Hun Kim, Sanghee Jung, Jimin Han, Bin Lee
IF 1.8
Powder Metallurgy
This study investigated a nanoparticle coating method to simultaneously improve the flow characteristics of spherical powder used in powder bed fusion (PBF) additive manufacturing (AM) and the mechanical properties of the resulting additively manufactured specimens. By employing a simple surface dry-coating of nano-sized hydrophobic SiO 2 powder, the flowability of Ti-6Al-4V powder was enhanced. Additionally, the mechanical properties including strength and hardness of the Ti-6Al-4V alloy produced via laser powder bed fusion (L-PBF) were improved by incorporating hydrophobic SiO 2 . The spherical Ti-6Al-4V powder was mixed with 0.1 wt.% SiO 2 using a 3D tubular mixer, and subsequent microstructural analyses confirmed the uniform dispersion of oxide particles. Various cubic specimens were then fabricated using the PBF process to optimize AM parameters. The relative density of the additively manufactured samples revealed that the highest density sample achieved approximately 99%. Tensile tests demonstrated that specimens fabricated from the SiO 2 -coated Ti-6Al-4V powder exhibited a maximum tensile strength of 1188 MPa and an elongation of 11.4%, representing an 11.0% improvement in tensile strength compared to samples without SiO 2 addition. The enhanced mechanical properties and subsequent microstructural analysis suggest that the increase in strength is attributable to oxide dispersion strengthening induced by SiO 2 .
Don't Judge Code by Its Cover: Exploring Biases in LLM Judges for Code Evaluation
Ji‐Won Moon, Yerin Hwang, Dongryeol Lee, Tae Jin Kang, Yongil Kim, Kyomin Jung
ArXiv.org
With the growing use of large language models(LLMs) as evaluators, their application has expanded to code evaluation tasks, where they assess the correctness of generated code without relying on reference implementations. While this offers scalability and flexibility, it also raises a critical, unresolved question: Can LLM judges fairly and robustly evaluate semantically equivalent code with superficial variations? Functionally correct code often exhibits variations-such as differences in variable names, comments, or formatting-that should not influence its correctness. Yet, whether LLM judges can reliably handle these variations remains unclear. We present the first comprehensive study of this issue, defining six types of potential bias in code evaluation and revealing their systematic impact on LLM judges. Across five programming languages and multiple LLMs, we empirically demonstrate that all tested LLM judges are susceptible to both positive and negative biases, resulting in inflated or unfairly low scores. Moreover, we observe that LLM judges remain vulnerable to these biases even when prompted to generate test cases before scoring, highlighting the need for more robust code evaluation methods.
