Junhyuk Choi

Junhyuk Choi, Sohhyung Park, Chanhee Cho et al.

While LLM-as-a-Judge is widely used in automated evaluation, existing validation practices primarily operate at the level of observed outputs, offering limited insight into whether LLM judges themselves function as stable and reliable measurement instruments. To address this limitation, we introduce a two-phase diagnostic framework for assessing reliability of LLM-as-a-Judge, grounded in Item Response Theory (IRT). The framework adopts Graded Response Model (GRM) of IRT and formalizes reliability along two complementary dimensions: (1) intrinsic consistency, defined as the stability of measurement behavior under prompt variations, and (2) human alignment, capturing correspondence with human quality assessments. We empirically examine diverse LLM judges with this framework, and show that leveraging IRT-GRM yields interpretable signals for diagnosing judgments systematically. These signals provide practical guidance for verifying reliablity of LLM-as-a-Judge and identifying potential causes of unreliability.

Seojin Hwang, Minju Kim, Junhyuk Choi et al.

Large Language Models (LLMs) are increasingly deployed in high-stakes decision-making settings such as legal reasoning, where consistency under factually equivalent inputs is critical. However, we find that fact-preserved but differently framed inputs can significantly destabilize LLM decisions. To systematically investigate this problem, we introduce Fragile, a large-scale benchmark that isolates fact-preserving semantic framing across three controlled dimensions: value-tinted narration, temporal slice, and narrative vividness. Our experiments reveal a high susceptibility of LLMs to framing, with an average decision flip rate of 28.6%. We find that simple prior prompt-level and activation-level interventions not only fail to suppress framing sensitivity but actively amplify it. We therefore propose Valign, a representation-level method that explicitly targets these framing dimensions by anchoring decisions to a stable value prior, steering hidden states toward the model's value-consistent direction, and projecting out temporal-vividness-sensitive directions from the model's hidden states. Valign consistently reduces framing-induced decision flips, demonstrating that robust mitigation requires directly targeting the internal pathways in which framing operates.

Yonghyun Jun, Junhyuk Choi, Jihyeong Park et al.

Despite the rapid proliferation of Role-Playing Agents (RPAs) based on Large Language Models (LLMs), the structural dimensions defining a character's identity remain weakly formalized, often treating characters as arbitrary text inputs. In this paper, we propose the concept of \textbf{Character Identity}, a multidimensional construct that disentangles a character into two distinct layers: \textbf{(1) Parametric Identity}, referring to character-specific knowledge encoded from the LLM's pre-training, and \textbf{(2) Attributive Identity}, capturing fine-grained behavioral properties such as personality traits and moral values. To systematically investigate these layers, we construct a unified character profile schema and generate both Famous and Synthetic characters under identical structural constraints. Our evaluation across single-turn and multi-turn interactions reveals two critical phenomena. First, we identify \textit{"Fame Fades"}: while famous characters hold a significant advantage in initial turns due to parametric knowledge, this edge rapidly vanishes as models prioritize accumulating conversational context over pre-trained priors. Second, we find that \textit{"Nature Remains"}: while models robustly portray general personality traits regardless of polarity, RPA performance is highly sensitive to the valence of morality and interpersonal relationships. Our findings pinpoint negative social natures as the primary bottleneck in RPA fidelity, guiding future character construction and evaluation.

Junhyuk Choi, Jeongyoun Kwon, Heeju Kim et al.

Multi-agent systems utilizing large language models often assign authoritative roles to improve performance, yet the impact of authority bias on agent interactions remains underexplored. We present the first systematic analysis of role-based authority bias in free-form multi-agent evaluation using ChatEval. Applying French and Raven's power-based theory, we classify authoritative roles into legitimate, referent, and expert types and analyze their influence across 12-turn conversations. Experiments with GPT-4o and DeepSeek R1 reveal that Expert and Referent power roles exert stronger influence than Legitimate power roles. Crucially, authority bias emerges not through active conformity by general agents, but through authoritative roles consistently maintaining their positions while general agents demonstrate flexibility. Furthermore, authority influence requires clear position statements, as neutral responses fail to generate bias. These findings provide key insights for designing multi-agent frameworks with asymmetric interaction patterns.

Junhyuk Choi, Yeseon Hong, Minju Kim et al.

Large Language Models (LLMs) show impressive conversational abilities but sometimes show identity drift problems, where their interaction patterns or styles change over time. As the problem has not been thoroughly examined yet, this study examines identity consistency across nine LLMs. Specifically, we (1) investigate whether LLMs could maintain consistent patterns (or identity) and (2) analyze the effect of the model family, parameter sizes, and provided persona types. Our experiments involve multi-turn conversations on personal themes, analyzed in qualitative and quantitative ways. Experimental results indicate three findings. (1) Larger models experience greater identity drift. (2) Model differences exist, but their effect is not stronger than parameter sizes. (3) Assigning a persona may not help to maintain identity. We hope these three findings can help to improve persona stability in AI-driven dialogue systems, particularly in long-term conversations.

Junhyuk Choi, Jihwan Seol, Nayeon Kim et al.

Speech-aware Language Models (SpeechLMs) have fundamentally transformed human-AI interaction by enabling voice-based communication, yet they may exhibit acoustic-based gender differentiation where identical questions lead to different responses based on the speaker's gender. This paper propose a new dataset that enables systematic analysis of this phenomenon, containing 9,208 speech samples across three categories: Gender-Independent, Gender-Stereotypical, and Gender-Dependent. We further evaluated LLaMA-Omni series and discovered a paradoxical pattern; while overall responses seems identical regardless of gender, the pattern is far from unbiased responses. Specifically, in Gender-Stereotypical questions, all models consistently exhibited male-oriented responses; meanwhile, in Gender-Dependent questions where gender differentiation would be contextually appropriate, models exhibited responses independent to gender instead. We also confirm that this pattern does not result from neutral options nor perceived gender of a voice. When we allow neutral response, models tends to respond neutrally also in Gender-Dependent questions. The paradoxical pattern yet retains when we applied gender neutralization methods on speech. Through comparison between SpeechLMs with corresponding backbone LLMs, we confirmed that these paradoxical patterns primarily stem from Whisper speech encoders, which generates male-oriented acoustic tokens. These findings reveal that current SpeechLMs may not successfully remove gender biases though they prioritized general fairness principles over contextual appropriateness, highlighting the need for more sophisticated techniques to utilize gender information properly in speech technology.

Junhyuk Choi, Ro-hoon Oh, Jihwan Seol et al.

We introduce VoiceBBQ, a spoken extension of the BBQ (Bias Benchmark for Question Answering) - a dataset that measures social bias by presenting ambiguous or disambiguated contexts followed by questions that may elicit stereotypical responses. Due to the nature of speech, social bias in Spoken Language Models (SLMs) can emerge from two distinct sources: 1) content aspect and 2) acoustic aspect. The dataset converts every BBQ context into controlled voice conditions, enabling per-axis accuracy, bias, and consistency scores that remain comparable to the original text benchmark. Using VoiceBBQ, we evaluate two SLMs - LLaMA-Omni and Qwen2-Audio - and observe architectural contrasts: LLaMA-Omni resists acoustic bias while amplifying gender and accent bias, whereas Qwen2-Audio substantially dampens these cues while preserving content fidelity. VoiceBBQ thus provides a compact, drop-in testbed for jointly diagnosing content and acoustic bias across spoken language models.

Junhyuk Choi, Hyeonchu Park, Haemin Lee et al.

Recent advances in Large Language Models (LLMs) have generated significant interest in their capacity to simulate human-like behaviors, yet most studies rely on fictional personas rather than actual human data. We address this limitation by evaluating LLMs' ability to predict individual economic decision-making using Pay-What-You-Want (PWYW) pricing experiments with real 522 human personas. Our study systematically compares three state-of-the-art multimodal LLMs using detailed persona information from 522 Korean participants in cultural consumption scenarios. We investigate whether LLMs can accurately replicate individual human choices and how persona injection methods affect prediction performance. Results reveal that while LLMs struggle with precise individual-level predictions, they demonstrate reasonable group-level behavioral tendencies. Also, we found that commonly adopted prompting techniques are not much better than naive prompting methods; reconstruction of personal narrative nor retrieval augmented generation have no significant gain against simple prompting method. We believe that these findings can provide the first comprehensive evaluation of LLMs' capabilities on simulating economic behavior using real human data, offering empirical guidance for persona-based simulation in computational social science.

Junhyuk Choi, Minju Kim, Yeseon Hong et al.

As large vision language models(LVLMs) rapidly advance, concerns about their potential to learn and generate social biases and stereotypes are increasing. Previous studies on LVLM's stereotypes face two primary limitations: metrics that overlooked the importance of content words, and datasets that overlooked the effect of color. To address these limitations, this study introduces new evaluation metrics based on the Stereotype Content Model (SCM). We also propose BASIC, a benchmark for assessing gender, race, and color stereotypes. Using SCM metrics and BASIC, we conduct a study with eight LVLMs to discover stereotypes. As a result, we found three findings. (1) The SCM-based evaluation is effective in capturing stereotypes. (2) LVLMs exhibit color stereotypes in the output along with gender and race ones. (3) Interaction between model architecture and parameter sizes seems to affect stereotypes. We release BASIC publicly on [anonymized for review].

Junhyuk Choi, Yeseon Hong, Bugeun Kim

Large Language Models (LLMs) have been recently adopted in interactive systems requiring communication. As the false belief in a model can harm the usability of such systems, LLMs should not have cognitive biases that humans have. Psychologists especially focus on the False Consensus Effect (FCE), a cognitive bias where individuals overestimate the extent to which others share their beliefs or behaviors, because FCE can distract smooth communication by posing false beliefs. However, previous studies have less examined FCE in LLMs thoroughly, which needs more consideration of confounding biases, general situations, and prompt changes. Therefore, in this paper, we conduct two studies to examine the FCE phenomenon in LLMs. In Study 1, we investigate whether LLMs have FCE. In Study 2, we explore how various prompting styles affect the demonstration of FCE. As a result of these studies, we identified that popular LLMs have FCE. Also, the result specifies the conditions when FCE becomes more or less prevalent compared to normal usage.