Searcharxiv⌕ Search

arXiv subjects

Yohan Jo

Publications and source records attributed to Yohan Jo.

At least 37 records · Page 2Linked to original sources

Human Psychometric Questionnaires Mischaracterize LLM Behavior

We examine whether human psychometric questionnaires can serve as reliable tools for characterizing and predicting LLM behavior in everyday user interactions. We analyze eight open-source LLMs by comparing their value and personality profiles derived from two different methods: Likert self-reports on established questionnaires (PVQ-40/21 and BFI-44/10) and generation probabilities over value-laden responses to everyday user queries. The two profiles diverge substantially. Within-construct item consistency, often cited as evidence of stable LLM dispositions, disappears in generation probabilities. We find that established questionnaire items contain explicit lexical cues that allow models to recognize the target construct and respond in alignment-consistent, socially desirable ways, whereas realistic user queries contain far less recognizable cues. In addition, demographic persona prompts shift models' responses to human questionnaires in ways consistent with real human patterns, but no such shifts appear in the generation setting, highlighting that human questionnaires overestimate LLMs' ability to faithfully reproduce expected psychological traits when role-playing demographic personas. Overall, our study indicates that questionnaire scores alone should not be treated as evidence of LLMs' response tendencies in realistic user interactions, and supports generation-probability profiling with ecologically valid items as a complementary behavioral measure.

cs.CL↗

Knowledge Editing for Masked Diffusion Language Models

Knowledge editing aims to update or correct factual knowledge in a language model. A widely used approach, locate-then-edit, first localizes a fact within the model and then edits the weights there. To date, such methods have been developed exclusively for autoregressive models (ARMs). Whether they work for masked diffusion models (MDMs), which model text bidirectionally and generate by iterative denoising rather than next-token prediction, remains an open question. We address it by transferring locate-then-edit to MDMs and comparing multiple MDMs with their matched ARMs. Our central finding has two parts. First, where an edit should be applied transfers between them: the same early-to-mid-layer MLP at the last subject token is most effective for both. Second, this shared location does not guarantee a shared outcome. Single-token edits succeed in both, but as targets grow longer, editing degrades far more sharply in the MDMs than in the ARMs. The failure stems from how the edited fact is generated: producing a multi-token target passes through intermediate states in which the target is partially unmasked, for which the edit was never optimized. Guided by this diagnosis, we introduce a simple correction that optimizes the edit including such states, substantially restoring multi-token performance. Our code is available at https://github.com/holi-lab/MDM-KE.

cs.CL↗

ArcANE: Do Role-Playing Language Agents Stay in Character at the Right Time?

Role-playing language agents (RPLAs) simulate specific characters and personas across applications such as entertainment, companionship, interactive storytelling, and education. Faithful role-play requires more than producing plausible, in-character responses: as a character's values and behavior change over a narrative, an RPLA should reflect the character's state at the relevant stage. However, existing benchmarks largely treat characters as fixed personas or test only what they know at a given point in the narrative. We introduce ArcANE (Arc-Aware Narrative Evaluation), a benchmark for evaluating whether an RPLA follows a character's development across a narrative. ArcANE first builds an Arc that maps how a character's values, motivations, or relationships change over the story. The benchmark then scores how well an RPLA's responses fit the corresponding stages of the Arc, covering three distinct scenario types: scenes from the novel, new situations within its world, and situations outside that world. We evaluate six models under six ways of providing narrative context. In every model, using the Arc up to the queried chapter yields the best performance, outperforming the strongest non-Arc context by 2.2-8.4 points. These results suggest that faithful role-play requires evolving character states and tracking their trajectory, rather than merely retrieving relevant episodic evidence.

cs.CL↗

Think Like a Doctor: Conversational Diagnosis through the Exploration of Diagnostic Knowledge Graphs

Conversational diagnosis requires multi-turn history-taking, where an agent asks clarifying questions to refine differential diagnoses under incomplete information. Existing approaches often rely on the parametric knowledge of a model or assume that patients provide rich and concrete information, which is unrealistic. To address these limitations, we propose a conversational diagnosis system that explores a diagnostic knowledge graph to reason in two steps: (i) generating diagnostic hypotheses from the dialogue context, and (ii) verifying hypotheses through clarifying questions, which are repeated until a final diagnosis is reached. Since evaluating the system requires a realistic patient simulator that responds to the system's questions, we adopt PatientSim, a persona-driven patient simulator, together with patient profiles from MIMIC-IV. We further adapt it with low-specificity symptom reporting to reflect how real-world patients describe symptoms vaguely during early clinical encounters. Experiments show improved diagnostic accuracy and efficiency over strong baselines, and physician evaluations support the realism of our simulator and the clinical utility of the generated clarifying questions. Our code will be released upon publication.

cs.AI↗

SpokenUS: A Spoken User Simulator for Task-Oriented Dialogue

Robust voice agents require exposure to the full diversity of how people interact through speech. However, obtaining enough spoken interactions is prohibitively expensive. Building spoken user simulators that address this requires large-scale spoken task-oriented dialogue (TOD) data encompassing spoken user behaviors, yet existing datasets are limited in scale and domain coverage, with no systematic pipeline for augmenting them. To address this, we introduce SpokenTOD, a spoken TOD dataset of 52,390 dialogues and 1,034 hours of speech augmented with four spoken user behaviors---cross-turn slots, barge-in, disfluency, and emotional prosody---across diverse speakers and domains. Building on SpokenTOD, we present SpokenUS, a spoken user simulator grounded in TOD that decides when to speak through a dedicated turn-taking head. SpokenUS achieves comparable goal coverage to much larger models while substantially outperforming all baselines in human MOS, disclosing slot values gradually across the dialogue as humans do rather than front-loading them. Further analysis confirms that SpokenUS's spoken behaviors pose meaningful challenges to voice agents, making it a practical tool for evaluating more robust spoken dialogue systems. Our code is available at https://github.com/holi-lab/SpokenUS.

cs.CL↗

Mechanism Shift During Post-training from Autoregressive to Masked Diffusion Language Models

Post-training pretrained autoregressive models (ARMs) into masked diffusion models (MDMs) provides an efficient route to diffusion language modeling, but it remains unclear whether the resulting models reuse inherited autoregressive computation or reorganize it for non-autoregressive generation. We compare two 7B ARM-MDM families across four controlled diagnostic tasks and find a task-dependent mechanism shift. On prefix-dominant tasks, MDMs largely preserve inherited high-attribution pathways or exhibit only modest changes in where computation occurs. On globally constrained tasks, the reorganization is substantially stronger, with task-relevant computation shifting toward earlier layers. This depth-wise pattern persists across prompt resampling, circuit budgets, and tested inference budgets, while targeted ablations support the functional importance of the identified structures under the tested intervention protocols. At the component level, diagnostic probes suggest that ARMs rely more strongly on sharply specialized components, whereas MDMs exhibit weaker single-component specialization and more diffuse output-space alignment. Together, these results suggest that diffusion post-training selectively preserves or reorganizes inherited computation according to task structure, rather than uniformly replacing autoregressive mechanisms.

cs.LG↗

Fine-Grained Multi Image Object Hallucination Benchmark

Multimodal Large Language Models (MLLMs) are increasingly deployed in multi-image scenarios requiring complex reasoning across visual contexts. However, current MLLMs remain fundamentally limited by object hallucination-generating plausible yet factually inconsistent descriptions about objects. Existing benchmarks, designed primarily for single-image settings or providing only high-level multi-image assessments, cannot systematically diagnose how visual complexity and reasoning demands trigger hallucination. To address this gap, we introduce MIOH, a fine-grained multi-image object hallucination benchmark that systematically evaluates object hallucination across four foundational tasks (existence, counting, attribute, position) through three multi-image reasoning patterns (comprehensive, comparative, selective) under three controlled adversarial pressures (visual context scale, perceptual difficulty, contextual bias). Through evaluation of 29 models, we reveal that even state-of-the-art systems like GPT-5 and Gemini-2.5-Pro exhibit distinct failure patterns across different reasoning patterns and tasks. Our evaluation reveals that hallucination stems not merely from perceptual failures but from integration-stage limitations when maintaining object representations across multiple images. MIOH provides a controlled framework for analyzing multi-image object hallucination and serves as a critical evaluation tool for developing more reliable multimodal AI systems.

cs.CV↗

Future Policy Approximation for Offline Reinforcement Learning in LLM Reasoning

Reinforcement learning (RL) has emerged as a key driver of post-training for complex reasoning in large language models (LLMs), yet online RL introduces substantial instability and computational overhead. Offline RL offers a compelling alternative by decoupling generation from training; however, offline algorithms for reasoning remain under-optimized relative to their online counterparts. We revisit the potential of policy-gradient-style offline RL and address a central challenge in offline learning: gradient entanglement. In long-horizon reasoning trajectories, correct and incorrect solutions share substantial token overlap, causing gradient updates from incorrect trajectories to suppress tokens that are also critical for correct ones. We propose Future Policy Approximation (FPA), a simple offline policy-gradient method that weights gradients using an estimate of the future policy rather than the current policy, enabling proactive gradient reweighting. We estimate the future policy through logit- space extrapolation. Across three models, seven mathematical reasoning benchmarks, and three code-generation benchmarks, FPA consistently improves over strong offline baselines, including DPO, RPO, KTO, and vanilla offline RL. FPA stabilizes long-horizon training, where vanilla objectives degrade, and achieves accuracy comparable to state-of-the-art RLVR methods such as GRPO and DAPO at a fraction of the GPU hours.

cs.CL↗

Bridging the Knowledge-Prediction Gap in LLMs on Multiple-Choice Questions

While large language models (LLMs) perform strongly on diverse tasks, their trustworthiness is limited by erratic behavior that is unfaithful to their internal knowledge. In particular, LLMs often fail on multiple-choice questions (MCQs) even if they encode correct answers in their hidden representations, revealing a misalignment between internal knowledge and output behavior. We investigate and mitigate this knowledge-prediction gap on MCQs through a three-step analysis of hidden representations. First, we quantify the prevalence and magnitude of the gap across models and datasets. Second, we provide a geometric interpretation by identifying distinct knowledge and prediction subspaces in the residual stream. Third, we introduce KAPPA, a lightweight inference-time intervention that aligns the two subspaces within the residual stream to reduce the knowledge-prediction gap. Our results provide a geometric and interpretable explanation of the knowledge-prediction gap in LLMs. Furthermore, KAPPA effectively reduces the gap across diverse MCQ benchmarks and models, and generalizes to free-form settings.

cs.CL↗

Dual Mechanisms of Value Expression: Intrinsic vs. Prompted Values in Large Language Models

Large language models can express values in two main ways: (1) intrinsic expression, reflecting the model's inherent values learned during training, and (2) prompted expression, elicited by explicit prompts. Given their widespread use in value alignment, it is paramount to clearly understand their underlying mechanisms, particularly whether they mostly overlap (as one might expect) or rely on distinct mechanisms. We analyze this largely understudied problem at the mechanistic level using two approaches: (1) value vectors, feature directions representing value mechanisms extracted from the residual stream, and (2) value neurons, MLP neurons that contribute to value vectors. We demonstrate that intrinsic and prompted value mechanisms partly share common components crucial for inducing value expression, generalizing across languages and reconstructing theoretical inter-value correlations in the model's internal representations. Yet, each mechanism also possesses unique components that fulfill distinct roles. In particular, the intrinsic mechanism activates in more diverse value-related scenarios and promotes response diversity, whereas the prompted mechanism strengthens instruction compliance, taking effect even in distant tasks like jailbreaking.

cs.CL↗

Psychometric Item Validation Using Virtual Respondents with Trait-Response Mediators

As psychometric surveys are increasingly used to assess the traits of large language models (LLMs), the need for scalable survey item generation suited for LLMs has also grown. A critical challenge here is ensuring the construct validity of generated items, i.e., whether they truly measure the intended trait. Traditionally, this requires costly, large-scale human data collection. To make it efficient, we present a framework for virtual respondent simulation using LLMs. Our central idea is to account for mediators: factors through which the same trait can give rise to varying responses to a survey item. By simulating respondents with diverse mediators, we identify survey items that yield responses robustly correlated with intended traits across these mediators. Experiments on three psychological trait theories (Big5, Schwartz, VIA) show that our mediator generation methods and simulation framework effectively identify high-validity items. LLMs demonstrate the ability to generate plausible mediators from trait definitions and to simulate respondent behavior for item validation. Our problem formulation, metrics, methodology, and dataset open a new direction for cost-efficient survey development and a deeper understanding of how LLMs simulate human survey responses. We release our dataset and code to support future work.

cs.CL↗

MIST: Multimodal Interactive Speech-based Tool-calling Conversational Assistants for Smart Homes

The rise of Internet of Things (IoT) devices in the physical world necessitates voice-based interfaces capable of handling complex user experiences. While modern Large Language Models (LLMs) already demonstrate strong tool-usage capabilities, modeling real-world IoT devices presents a difficult, understudied challenge which combines modeling spatiotemporal constraints with speech inputs, dynamic state tracking, and mixed-initiative interaction patterns. We introduce MIST (the Multimodal Interactive Speech-based Tool-calling Dataset), a synthetic multi-turn, voice-driven code generation task that operates over IoT devices. We find that there is a significant gap between open- and closed-weight multimodal LLMs on MIST, and that even frontier closed-weight LLMs have substantial headroom. We release MIST and an extensible data generation framework to build related datasets in order to facilitate research on mixed-initiative voice assistants which reason about physical world constraints.

cs.CL↗

TriBench-Ko: Evaluating LLM Risks in Judicial Workflows

Large language models (LLMs) are increasingly integrated into legal workflows. However, existing benchmarks primarily address proxy tasks, such as bar examination performance or classification, which fail to capture the performance and risks inherent in day-to-day judicial processes. To address this, we publicly release TriBench-Ko, a Korean benchmark designed to evaluate potential deployment risks of LLMs within the context of verified judicial task requirements. It covers four core tasks: jurisprudence summarization, precedent retrieval, legal issue extraction, and evidence analysis. It jointly assesses model behavior across multiple deployment risk categories, including inaccuracy (hallucination, omission, statutory misapplication), biases (demographic, overcompliance), inconsistencies (prompt sensitivity, non-determinism), and adjudicative overreach. Each item is structured to systematically assess both task performance and a specific risk type based on real judicial decisions. Our evaluation of a range of contemporary LLMs reveals that many models frequently manifest significant risks, most notably struggling with precedent retrieval and failing to capture critical legal information. We provide a comprehensive diagnosis of these LLMs and pinpoint critical areas where LLM-generated outputs in judicial contexts necessitate rigorous inspection and caution. Our dataset and code are available at https://github.com/holi-lab/TriBench-Ko

cs.CL↗

ThinkBrake: Efficient Reasoning via Log-Probability Margin Guided Decoding

Large Reasoning Models (LRMs) allocate substantial inference-time compute to Chain-of-Thought (CoT) reasoning, improving performance on mathematics, scientific QA, and tool usage. However, this introduces overthinking: LRMs often reach a correct intermediate solution, continue reasoning, and overwrite it with an incorrect answer. We first demonstrate that oracle stopping--where we inject at every sentence boundary and select the best stopping point in hindsight--improves average accuracy by 8% while reducing thinking tokens by 72%, exposing substantial overthinking. Motivated by this finding, we propose ThinkBrake, which monitors the log-probability margin between the top continuation token and at sentence boundaries, stopping reasoning when this margin narrows. ThinkBrake requires no training and achieves favorable accuracy-efficiency trade-offs across math, scientific QA, and tool usage benchmarks, reducing thinking token usage by up to 30%. Furthermore, we provide theoretical analysis showing that ThinkBrake is equivalent to test-time realignment with a reward bonus for the token.

cs.CL↗

SpeakerSleuth: Can Large Audio-Language Models Judge Speaker Consistency across Multi-turn Dialogues?

Large Audio-Language Models (LALMs) as judges have emerged as a prominent approach for evaluating speech generation quality, yet their ability to assess speaker consistency across multi-turn dialogues remains unexplored. We present \textbf{SpeakerSleuth}, a benchmark evaluating whether LALMs can reliably judge speaker consistency across multi-turn dialogues through three tasks reflecting real-world requirements. We construct 1,818 human-verified evaluation instances across four diverse datasets spanning synthetic and real speech, with controlled acoustic difficulty. Evaluating twelve widely-used LALMs, we find that models struggle to reliably detect acoustic inconsistencies. For instance, given audio samples of the same speaker's turns, some models overpredict inconsistency, whereas others are overly lenient. Models further struggle to identify the exact turns that are problematic. When other interlocutors' turns are provided as textual context, performance degrades dramatically as models prioritize textual coherence over acoustic cues, failing to detect even obvious gender switches for a speaker. On the other hand, models perform substantially better in comparing and ranking acoustic variants, demonstrating inherent acoustic discrimination capabilities. These findings expose a significant bias in LALMs: they tend to prioritize text over acoustics, revealing fundamental modality imbalances that need to be addressed to build reliable audio-language judges. Our code and data are available at https://github.com/holi-lab/SpeakerSleuth.

cs.CL↗

Don't Adapt Small Language Models for Tools; Adapt Tool Schemas to the Models

Small language models (SLMs) enable scalable tool-augmented multi-agent systems where multiple SLMs handle subtasks orchestrated by a powerful coordinator. However, they struggle with tool-use tasks, particularly in selecting appropriate tools and identifying correct parameters. A common failure mode is \textit{schema misalignment}: models hallucinate plausible tool names that are absent from the provided tool schema, due to different naming conventions internalized during pretraining. Rather than training models to adapt to unfamiliar schemas, we propose adapting schemas to align with models' pretrained knowledge. We introduce \textbf{PA-Tool} (Pretraining-Aligned Tool Schema Generation), a training-free method that leverages peakedness, a signal used in contamination detection that indicates pretraining familiarity, to rename tool components. By generating multiple candidates and selecting the candidate with the highest peakedness, PA-Tool identifies pretraining-aligned naming patterns. Experiments on MetaTool and RoTBench show improvements of up to 17\%, with schema misalignment errors reduced by 80\%. PA-Tool enables small models to substantially improve tool-use accuracy without retraining, showing that schema-level interventions can unlock the tool-use potential of resource-efficient models. Our code is available at https://github.com/holi-lab/PA-Tool.

cs.CL↗

Learning to Retrieve User History and Generate User Profiles for Personalized Persuasiveness Prediction

Estimating the persuasiveness of messages is critical in various applications, from recommender systems to safety assessment of LLMs. While it is imperative to consider the target persuadee's characteristics, such as their values, experiences, and reasoning styles, there is currently no established systematic framework to optimize leveraging a persuadee's past activities (e.g., conversations) to the benefit of a persuasiveness prediction model. To address this problem, we propose a context-aware user profiling framework with two trainable components: a query generator that generates optimal queries to retrieve persuasion-relevant records from a user's history, and a profiler that summarizes these records into a profile to effectively inform the persuasiveness prediction model. Our evaluation on the ChangeMyView Reddit dataset shows consistent improvements over existing methods across multiple predictor models, raising F1 from 33% to 47% on Llama-3.3-70B-Instruct. Further analysis shows that effective user profiles are context-dependent and predictor-specific, rather than relying on static attributes or surface-level similarity. Together, these results highlight the importance of task-oriented, context-dependent user profiling for personalized persuasiveness prediction.

cs.CL↗

Towards Motion-aware Referring Image Segmentation

Referring Image Segmentation (RIS) requires identifying objects from images based on textual descriptions. We observe that existing methods significantly underperform on motion-related queries compared to appearance-based ones. To address this, we first introduce an efficient data augmentation scheme that extracts motion-centric phrases from original captions, exposing models to more motion expressions without additional annotations. Second, since the same object can be described differently depending on the context, we propose Multimodal Radial Contrastive Learning (MRaCL), performed on fused image-text embeddings rather than unimodal representations. For comprehensive evaluation, we introduce a new test split focusing on motion-centric queries, and introduce a new benchmark called M-Bench, where objects are distinguished primarily by actions. Extensive experiments show our method substantially improves performance on motion-centric queries across multiple RIS models, maintaining competitive results on appearance-based descriptions. Codes are available at https://github.com/snuviplab/MRaCL

cs.CV↗