arXiv subjects

Hyunwoo Kim

Publications and source records attributed to Hyunwoo Kim.

At least 19 recordsLinked to original sources

arXiv2026

Complete Donor-Set Guidance for Frozen Diffusion Docking Models

Metalloprotein docking requires a ligand pose to satisfy both global pocket fit and the identity and geometry of metal-coordinating atoms. General-purpose diffusion docking models do not explicitly condition generation on externally specified complete ligand donor sets, and their predicted poses can exhibit incorrect metal-donor relationships. Metal-aware models address coordination during training and generation; here we ask whether external donor information can correct already-trained general-purpose models without retraining. We propose Complete-Identity Contract Guidance (CICG), which applies a supplied complete donor set through translation, rotation, and torsion updates, and Historical Contract Transfer (HCT), which supplies donor identities from earlier complexes at compatible receptor sites. We compare HCT with trajectory-local controls retaining the historical donor count with or without elemental composition. On 86 assessable HCT-covered queries, HCT-guided CICG increases frozen-confidence-selected joint success from 21.5% to 36.6% on SurfDock and from 25.0% to 34.9% on DiffDock-L. The joint criterion requires correct global ligand placement, complete native donor recovery, coordination geometry, preserved stereochemistry, and absence of fatal receptor clashes. Composition-matched guidance reaches 33.14% and 34.30%, leaving the additional benefit of fixed historical membership unresolved. Our results therefore support donor-set guidance as a conditional correction for frozen docking models, while identifying donor-source reliability and physical-quality preservation as the main barriers to broader use.

physics.chem-ph↗

arXiv2026

TutorCaseBench: A Case-Grounded Benchmark for Evaluating LLM Tutors in Supporting Students with Challenging Behaviors

LLMs are increasingly used as AI tutors, yet existing benchmarks overlook the challenging behaviors common among students with learning and behavioral difficulties. We introduce TutorCaseBench, a benchmark that grounds AI-tutor evaluation in qualitative case studies from pedagogy, educational psychology, and special education. Our pipeline converts documented student categories into case-specific scenarios, tutor-student dialogues, and rubrics grounded in source evidence. The benchmark contains 923 dialogues across five challenge types, evaluated through open-ended response generation (1,560 items) and multiple-choice selection (2,537 items). Across 11 state-of-the-art LLMs, even the strongest model satisfies a rubric criterion in only 55.8% of open-ended cases, while models are better at selecting appropriate strategies than generating situated responses. These results reveal a gap between intervention recognition and context-specific adaptation, especially for open-weight models. All data are publicly available at https://github.com/kixlab/TutorCaseBench.

cs.CL↗

arXiv2026

From Simulated Citizens to Simulated Deliberation: Challenges in Representation and Interaction

Can LLM agents simulate how a real public would deliberate? For such simulations to be meaningful, agents must reflect the population's opinions, and their views must be shaped by what others say during discussion. We evaluate these requirements across six LLMs using national surveys, controlled deliberation experiments, and a real deliberative poll. Comparing census-grounded Korean personas against national surveys, we find that agents often produce overly concentrated opinions and fail to reliably reproduce demographic differences. During simulated deliberation, agents exchange arguments, respond to one another, and frequently change positions. Yet similar changes occur even when agents cannot see one another's contributions, while simply restating their starting positions sharply reduces movement. We further test whether these limitations persist in a real-world setting, using agents constructed from actual respondents in the 2017 Shin-Kori deliberative poll. Even with real respondent profiles, agents fail to reproduce the panel's initial opinion distribution or how its opinions change across survey waves. Our findings show that generating deliberative discussions, representing public opinion, and reproducing interaction-driven opinion change are distinct capabilities. These results suggest a different potential role for LLM deliberation: surfacing arguments and perspectives, rather than simulating a real public.

cs.AI↗

arXiv2026

PADO-Pauli: A GPU-Accelerated Implementation of Pauli Propagation

Variational quantum algorithms utilize parameterized circuits many times during training, at new parameter values, over batches of data, and with gradients. Pauli propagation approximates the expectation values of such circuits classically by evolving the observable in the Heisenberg picture as a truncated sum of Pauli strings. We present PADO-Pauli, a GPU engine for Pauli propagation that compiles a circuit and its observables once into sparse matrices in which each rotation angle contributes only a cosine and a sine factor, so every later evaluation reuses the compiled program. It evaluates batches of parameters and data inputs embedded as rotation angles. By default its gradients come from a vector-Jacobian product registered with PyTorch's automatic differentiation, so they flow through the angles to any classical layer that produces them. PADO-Pauli also provides truncation by Pauli weight and coefficient size, and noise channels. Zero-filtering removes terms that do not contribute to an expectation value in the all-zero reference state and makes the batched forward pass about 1.5 times faster. Compared to cuPauliProp on an NVIDIA A100, PADO-Pauli evaluates up to 10 times faster on 4 to 12 qubits and uses up to 3.1 times less memory on 16 qubits. We check its accuracy against exact statevector simulation up to 28 qubits. These features make PADO-Pauli a practical toolkit for training variational quantum algorithms, including quantum machine learning models.

quant-ph↗

arXiv2026

Before They Can Solve: Predicting Post-Training Coding-Agent Performance from Base Models

How can we predict which base checkpoint is worth an expensive round of agentic post-training? End-to-end pass@$K$ tests whether successful behavior already appears in a base model's distribution, but it is a poor fit for agentic coding: many base checkpoints cannot reliably produce the well-formed tool invocation required to complete a task end-to-end. Single-shot or short-horizon tasks avoid these tool-calling failures by collapsing a multi-step interaction into a fixed prompt and a single patch, but they sidestep the core capability we care about: maintaining coherent state over many tool-using steps as the repository evolves. To bridge this gap, we treat successful post-trained agent trajectories as a lookahead signal of base-model potential. Replaying each trajectory and rerunning tests after every code-changing step identifies the decisive step: the first step whose cumulative patch flips the repository from failing to passing, certifying that the recorded action solves the task given the prior context. Motivated by a coverage principle for agentic traces, we build three screens at this step that do not require a base checkpoint to drive the harness from a cold start: (i) Decisive-Action BPB (bits per byte) measures the probability mass on the certified action, (ii) Patch MCQ tests the checkpoint's choice between that action and alternatives rejected by the same verifier, and (iii) prefix-conditioned pass@$K$ evaluates support for functionally-correct generations and credits any continuation that the tests accept. Across ten pairs of public base and post-trained models, all three screens rank the cohort in close agreement with post-trained SWE-bench Verified pass@$1$. As our methods need only a benchmark's successful trajectories and its verifier, they can be applied to turn future agentic coding benchmarks into base-model evaluations.

cs.AI↗

arXiv2026

LIBERO-Para: A Diagnostic Benchmark and Metrics for Paraphrase Robustness in VLA Models

Vision-Language-Action (VLA) models achieve strong performance in robotic manipulation by leveraging pre-trained vision-language backbones. However, in downstream robotic settings, they are typically fine-tuned with limited data, leading to overfitting to specific instruction formulations and leaving robustness to paraphrased instructions underexplored. To study this gap, we introduce LIBERO-Para, a controlled benchmark that independently varies action expressions and object references for fine-grained analysis of linguistic generalization. Across seven VLA configurations (0.6B-7.5B), we observe consistent performance degradation of 22-52 pp under paraphrasing. This degradation is primarily driven by object-level lexical variation: even simple synonym substitutions cause large drops, indicating reliance on surface-level matching rather than semantic grounding. Moreover, 80-96% of failures arise from planning-level trajectory divergence rather than execution errors, showing that paraphrasing disrupts task identification. Binary success rate treats all paraphrases equally, obscuring whether models perform consistently across difficulty levels or rely on easier cases. To address this, we propose PRIDE, a metric that quantifies paraphrase difficulty using semantic and syntactic factors. Our benchmark and corresponding code are available at: https://github.com/cau-hai-lab/LIBERO-Para

cs.LG↗

arXiv2026

MatcherCompass: A Deployment-Aware Benchmark to Guide Image Matcher Selection in the Wild

Field robots operating across time of day and sensing modalities require accurate image correspondences within onboard time and resource budgets. However, accuracy and runtime reported for individual methods on a single device provide limited guidance for choosing a matcher and its configuration on a target platform. We present MatcherCompass, a deployment-aware benchmark for choosing local feature matchers in field robotics. Under common input and pose-evaluation procedures, we compare nine classical and learned matching pipelines across four image resolutions and supported numerical precisions. Four visual conditions cover viewpoint variation, day--night matching in visible and thermal imagery, and daytime visible--thermal matching. We evaluate pose accuracy using the area under the error--recall curve (AUC) at $5^\circ$, $10^\circ$, and $20^\circ$, and measure runtime, GPU memory, and energy per image pair on four GPU platforms spanning workstation and onboard computers. The results show that changes in hardware, input resolution, and numerical precision can move a matcher across a runtime budget boundary, altering the feasible choices. We organize the measurements into a selection guide that returns all configurations satisfying user-specified time and resource limits, together with their accuracy under the selected visual condition. MatcherCompass provides measured evidence for choosing matching pipelines that fit a robot's sensing conditions and computing hardware. Project page: https://matchercompass.github.io/.

cs.RO↗

arXiv2026

Penquiry: A Pen-based Interactive In-situ Q&A System Leveraging LLMs

Pen-based digital devices remain a preferred medium for active, cognitively engaging study. Concurrently, Large Language Models (LLMs) have become indispensable for self-directed learning, enabling students to clarify concepts. However, a fundamental interaction gap exists between the fluid, spatial nature of pen-based workflows and the discrete, keyboard-heavy requirements of LLMs. We present Penquiry, an in-situ question-and-answer system that bridges this gap by enabling learners to pose questions directly on digital study materials via a pen. We characterize two primary interaction challenges in this multimodal transition: a Referential Barrier, which hinders grounding fine-grained visual elements into the query context, and an Expressive Barrier, which forces learners to translate diverse, non-textual intents--such as equations and diagrams--into rigid, typed sentences. To resolve these, Penquiry introduces a mediation layer featuring Content Snapping for unambiguous referencing and Question Autocompletion to expand sparse ink keywords into rich semantic queries. Through two iterative user studies (N = 16 per study), we demonstrate that Penquiry significantly reduces the cognitive and physical overhead of inquiry compared to traditional interfaces, providing a new blueprint for pen-based, in-situ AI interaction

cs.HC↗

arXiv2026

VA-DPO: Valence-Arousal Direct Preference Optimization for Controllable Emotion Generation in Language Models

How precisely can we tell a language model how to feel? Most work on emotional generation answers with a discrete label - happy, angry, sad - which cannot express a target like "mildly downcast but calm." We instead specify the desired affect as a continuous point (v*, a*) in the Valence-Arousal plane and train the model to hit it. Our method, VA-DPO, is a small modification to Direct Preference Optimization: a frozen VA regressor scores each sampled generation by its Euclidean distance to the target, we keep only candidate pairs whose distance gap clears a margin tau, and we optimize a LoRA adapter with the ordinary DPO loss against a frozen reference. The DPO objective itself is unchanged; what is new is how the preference data is built. On Llama-3.1-8B-Instruct this cuts mean VA distance to the target by 33% over system-prompting and 25% over few-shot prompting, lifting valence/arousal correlation to r_v=0.93 and r_a=0.75. The gains carry over to Qwen3-8B and Llama-3.2-3B, and they do not come at the usual price: MMLU is unchanged (Delta=+0.0) and HellaSwag and TruthfulQA are preserved. We release the code, configs, and the preference-construction pipeline.

cs.CL↗

arXiv2026

You Can't Prefer Emotions You Don't Sample: Intensity Undershoot in DPO-Tuned LLMs

Ask a language model to respond "very excitedly," and its output is typically only mildly more energetic. We quantify this effect. We condition an instruction-tuned LLM on a continuous Valence-Arousal (VA) target, where valence measures how pleasant a state is and arousal how activated it is, measure the achieved affect with a frozen regressor, and sweep the requested target from -1 to +1. The response moves far less than asked: the gain, the slope of achieved against requested affect, is only 0.26 for valence and 0.13 for arousal on Llama-3.1-8B, where a faithful controller would score 1. The model systematically undershoots requested emotional intensity, which puts a number on the qualitative observation of Fazzi et al. (2025). Our experiments trace this to the preference-learning pipeline. Training targets from natural corpora such as EmoBank are neutral-heavy, and the sampled candidates themselves rarely reach extreme affect, so Direct Preference Optimization (DPO) is left with no extreme exemplar to prefer. If instead we cover the target space uniformly and sample a hotter, larger candidate pool, valence gain rises from 0.26 to 0.40 +/- 0.02 (3 seeds) and extrapolation error drops, at only a modest in-distribution cost (EmoBank-test VA distance 0.092 to 0.107). The same recipe reproduces on Qwen3-8B (gain_v 0.44, with in-distribution accuracy preserved). Arousal is harder and less reliable: its gain barely moves on average and swings across seeds (0.14 +/- 0.07, against valence's tight +/- 0.02), because raising arousal needs candidates the base model is reluctant to generate. The evidence indicates that faithful intensity is bottlenecked by the extremity of the candidate pool rather than by the conditioning format.

cs.CL↗

arXiv2026

XYBench: Can LLMs Respond Pragmatically to Queries with Misconceptions?

When non-expert users ask LLMs for assistance, their queries can often have misconceptions (e.g., "How do I parse XML with regex?"). In such cases, often referred to as the XY-problem, LLMs must identify the misconception ("regex are fragile") and meaningfully direct the user toward a pragmatic solution that will address the root problem implicit in the request ("use an XML parser"). We introduce XYBench, a benchmark of 8,115 such queries, drawn from technical (StackOverflow/StackExchange) and everyday (WikiHow and a manually-curated subset) domains. We design an evaluation paradigm that assesses model responses along three criteria grounded in cooperative response theory: (a) presence and (b) emphasis on pragmatic solutions, and (c) identification of misconceptions. Our experiments show that even the strongest LLMs predominantly answer the literal request (0.75--0.92) and far less often the intended one (0.33--0.71), while substantially lagging behind humans at identifying misconceptions (at most 63% vs. 79--90%). Further, models overwhelmingly prefer pragmatic responses in a multiple choice setting yet consistently fail to generate them. Oracle ablation experiments show that providing explicit user intent at generation time helps; however a large gap remains, suggesting pragmatic redirection is a fundamentally underdeveloped capability in current LLMs.

cs.CL↗

arXiv2026

SSDi8: Accurate and Efficient 8-bit Quantization for State Space Duality

Recent advances in sequence modeling have highlighted Mamba as a state space architecture offering efficient long-range dependency modeling and providing a viable alternative to Transformers. Building upon this, Mamba-2 introduces the Structured State Space Duality (SSD), which integrates recurrent and attention modes to achieve efficiency and scalability. However, this architectural expansion substantially increases memory and latency overhead, underscoring the need for efficient compression strategies tailored to SSD. In this work, we present SSDi8, the first post-training quantization framework specifically designed for SSD to maintain a persistent INT8 path. SSDi8 introduces a reformulation that decouples element-wise multiplications from matrix multiplications, enabling reuse of quantized activations across modules. Moreover, SSDi8 adaptively quantizes channel-varying activations at cost-effective points, further reducing latency. On the accuracy side, SSDi8 explicitly leverages the intrinsic dimensional decomposition of SSD, exploiting distinct outlier distributions across axes, and incorporates an error correction term based on per-channel error statistics. Comprehensive experiments demonstrate that SSDi8 achieves accuracy comparable to FP16 while delivering up to 1.4x speedup in W4A8 and W8A8 settings. We further validate its robustness in resource-constrained environments by deploying it on the Orin NX device.

cs.AI↗

arXiv2026

Quark spin-orbit correlations in spin-1 targets

The quark spin-orbit correlation probes the alignment of quark helicity with longitudinal kinetic orbital angular momentum inside a hadron. This correlation is defined by a QCD operator: the position moment of the asymmetric parity-odd quark energy-momentum tensor. The matrix element of this rank-two tensor decomposes into symmetric-traceless, antisymmetric, and trace parts. The symmetric-traceless part is matched to moments of axial generalized parton distributions. The QCD equations of motion relate the antisymmetric part to vector and tensor form factors and set the trace to zero. Using these relations, we derive two gauge-invariant sum rules for the spin-orbit correlation in a spin-$1$ hadron. One gives the correlation in an unpolarized target. The other gives its tensor-polarization dependence, which is absent for spin-$0$ and spin-$1/2$ targets. We estimate the unpolarized spin-orbit correlations for the $ρ$ meson and deuteron using existing lattice and phenomenological inputs, respectively.

hep-ph↗

arXiv2026

DeltaPrompts: Escaping the Zero-Delta Trap in Multimodal Distillation

Distillation enables compact Vision-Language Models (VLMs) to obtain strong reasoning capabilities, yet the prompts driving this process are typically chosen via simple heuristics or aggregated from off-the-shelf datasets. We reveal a critical inefficiency in this approach: up to 69% of the prompts in standard chart / document reasoning datasets are effectively zero-delta, meaning the teacher and student already induce the exact same answer distribution. Training on these prompts provides minimal learning signal, causing student improvement to rapidly saturate regardless of data scale. To escape the zero-delta trap, we return to first principles: distillation fundamentally minimizes distributional divergence, and thus a prompt is valuable only if it exposes a functional capability gap between the teacher and student. We quantify this gap through answer divergence ($Δ$), demonstrating that non-zero divergence is critical for effective scaling. Building on this insight, we propose a staged synthesis pipeline that repurposes existing datasets as seeds, actively targeting student failure modes to produce better prompts. The result is DeltaPrompts, a diverse dataset of 200k synthetic, high-divergence reasoning problems. We evaluate DeltaPrompts across three distinct settings: on-policy distillation with the target teacher-student pair, transfer to a novel model family without regenerating the data, and off-policy fine-tuning of a non-reasoning model. Across all scenarios, DeltaPrompts drives substantial gains, yielding up to 15% relative improvement even on top of a highly-optimized reasoning model (e.g., Qwen3-VL-8B-Thinking) -- averaged over 10 benchmarks spanning chart, document and perception-centric reasoning.

cs.LG↗

arXiv2026

Prune Once: Retraining-Free Task-Agnostic Pruning for Vision-Language Models

Vision-language models (VLMs) have achieved remarkable generalization across diverse multimodal tasks through large-scale pre-training, yet their rapidly increasing computational and memory requirements pose significant challenges for deployment in constrained environments. Existing pruning strategies often depend on task-specific criteria or LLM-oriented importance measures, making them unsuitable for task-agnostic pruning, where no task-specific samples are available at pruning time and the pruned model remains broadly applicable. We introduce a retraining-free VLM pruning framework called PORTA that derives a task- and modality-agnostic importance formulation based on activation variation, estimated from generic calibration data, which reliably captures feature-level representation utility across modalities. PORTA further incorporates an adaptive sparsity allocation mechanism that assigns layer-wise pruning ratios based on output feature variability, avoiding the limitations of uniform sparsity and reducing performance degradation at high compression levels. Extensive experiments across VLM architectures, such as CLIP, BLIP, and Qwen2-VL, demonstrate that PORTA achieves competitive downstream performance under high sparsity without requiring any retraining, supporting efficient VLM compression. Code is available at https://github.com/cau-hai-lab/PORTA.git.

cs.CV↗

arXiv2026

Social World Models

Humans intuitively navigate social interactions by simulating unspoken dynamics and reasoning about others' perspectives, even with limited information. In contrast, AI systems struggle to structure and reason about implicit social contexts, as they lack explicit representations for unobserved dynamics such as intentions, beliefs, and evolving social states. In this paper, we introduce the concept of social world models (SWMs) to characterize the complex social dynamics. To operationalize SWMs, we introduce a novel structured social world representation formalism (S3AP), which captures the evolving states, actions, and mental states of agents, addressing the lack of explicit structure in traditional free-text-based inputs. Through comprehensive experiments across five social reasoning benchmarks, we show that S3AP significantly enhances LLM performance-achieving a +51% improvement on FANToM over OpenAI's o1. Our ablations further reveal that these gains are driven by the explicit modeling of hidden mental states, which proves more effective than a wide range of baseline methods. Finally, we introduce an algorithm for social world models using S3AP, which enables AI agents to build models of their interlocutors and predict their next actions and mental states. Empirically, S3AP-enabled social world models yield up to +18% improvement on the SOTOPIA multi-turn social interaction benchmark. Our findings highlight the promise of S3AP as a powerful, general-purpose representation for social world states, enabling the development of more socially-aware systems that better navigate social interactions.

cs.AI↗

arXiv2026

Trainability and Mode Separation of Mixed IQP-QCBMs

Quantum circuit Born machines (QCBMs) based on instantaneous quantum polynomial-time (IQP) circuits are promising quantum generative models for their classical trainability. It is known that their ancilla-free form avoids barren plateaus under certain initializations, but remains non-universal. Although adding ancilla qubits raises the expressivity, whether the ancilla-extended model retains local trainability remains unknown. We propose the mixed IQP-QCBM, which generalizes the ancilla-extended circuit as a weighted mixture of ancilla-free IQP circuits, called branches. For a polynomial number of branches, we prove local barren-plateau avoidance from data-agnostic and, under certain assumptions, data-dependent initializations. We further show that the mixed IQP-QCBM can surpass the best ancilla-free IQP circuit only if its branches generate a number of distinct distributions. In particular, we focus on a behavior we call \emph{mode separation}, in which each branch captures a particular feature of the target. Mode separation is hard to attain from an initialization whose branches generate the same distribution: the gradients that would separate them are suppressed while the distributions they generate remain close. This motivates \emph{cluster initialization}, which assigns a different unsupervised data cluster to each branch and provides an initial degree of mode separation. Exact calculations on two 16-bit datasets support the barren-plateau and gradient-suppression claims. On four benchmarks, binary clusters, a two-dimensional Ising model, binarized MNIST, and a 484-spin glass, cluster initialization converges fastest and reaches the lowest mean test $\mathrm{MMD}^2$. We observe that, when achieving the lowest test $\mathrm{MMD}^2$, the mixed IQP-QCBM contains branches specialized to distinguishable data features such as blob patterns, magnetization sectors, or digit shapes.

quant-ph↗

arXiv2026

Frictional Q-Learning

Off-policy reinforcement learning suffers from extrapolation errors when a learned policy selects actions that are weakly supported in the replay buffer. In this study, we address this issue by drawing an analogy to static friction. From this perspective, the replay buffer is represented as a smooth, low-dimensional action manifold, where the support directions correspond to the tangential component, while the normal component captures the dominant first-order extrapolation error. This decomposition reveals an intrinsic anisotropy in value sensitivity that naturally induces a stability condition analogous to a friction threshold. To mitigate deviations toward unsupported actions, we propose Frictional Q-Learning, an off-policy algorithm that encodes supported actions as tangent directions using a contrastive variational autoencoder. We further show that an orthonormal basis of the orthogonal complement corresponds to normal components under mild local isometry assumptions. Extensive empirical results on standard continuous-control benchmarks consistently demonstrate robust and stable performance compared with competitive baselines.

cs.LG↗