arXiv:2606.05262v1 Announce Type: new
Abstract: Uncrewed aerial vehicles (UAVs) are increasingly considered as aerial platforms capable of providing both sensing and communication services, representing a promising paradigm for intelligent transportation systems. This paper investigates the optimal time allocation for a UAV-enabled integrated sensing and communication (ISaC) system operating in the X-band for vehicular networks. We analyze the trade-off between sensing accuracy and communication performance under practical UAV constraints and fading effects, considering both single-shadowing and double-shadowing channel models. An optimization framework is developed to allocate time between sensing and communication while guaranteeing minimum communication rates and sufficient sensing reliability. Simulation results demonstrate adaptive time allocation strategies, highlighting how UAV-to-ground channel conditions and target distances influence the balance between sensing and communication in smart mobility scenarios.
Science Journals
arXiv:2511.20158v2 Announce Type: replace
Abstract: While continual visual instruction tuning (CVIT) has shown promise in adapting multimodal large language models (MLLMs), existing studies predominantly focus on models without safety alignment. This critical oversight ignores the fact that real-world MLLMs inherently require such mechanisms to mitigate potential risks. In this work, we shift our focus to CVIT for safety-aligned MLLMs and observe that during continual adaptation, the model not only suffers from task forgetting but also exhibits degradation in its safety. Achieving a harmonious balance between safety and task performance remains a crucial challenge. To address this, we propose Harmonious Parameter Adaptation (HPA), a post-training framework composed of focusing-based parameter partition, harmoniously balanced parameter selection, and orthogonal parameter adjustment. Specifically, HPA partitions parameters into two types based on their focus on safety or task performance, and selects the focused ones to preserve from a balanced perspective. In addition, HPA imposes orthogonality constraints on parameter updates to further alleviate catastrophic forgetting. Extensive experiments on the CVIT benchmark and safety evaluation datasets demonstrate that HPA better maintains high safety and mitigates forgetting than existing baselines. Code is available at https://github.com/Minato-Zackie/HPA.
arXiv:2606.05104v2 Announce Type: replace
Abstract: Knowledge benchmarks for LLMs face three issues: scaling-driven designs that do not operationalize disciplinary representativeness; flat-payment annotation that permits lazy consensus; and unaudited ranking instability under bounded test budgets. We introduce KINA, an 899-item benchmark across 261 fine-grained disciplines, with two formal results. First, we cast representativeness as a coverage-style objective over expert-elicited anchors and operationalize disciplinary representativeness through a proxy, yielding a (1-1/e) greedy approximation (Proposition 1); the guarantee applies to the proxy, not to population representativeness. Second, we prove a bonus-on-bar tournament weakly FOSD-dominates flat payment in released-review quality, with incentive-compatibility threshold B > Delta C / Delta p_min (Theorem 1). Evaluating 42 models from 13 labs, the top model, Gemini-3.1-Pro-Preview, reaches 53.17%, followed by Claude-Opus-4.6 at 49.92% and GPT-5.4 at 48.55%, leaving substantial headroom below saturation. The full leaderboard shows a tiered structure rather than a smooth total order: a small frontier tier lies above 48%, a dense strong-model tier spans roughly 38-45%, and low-performing models remain only modestly above the 10% chance baseline. Tool augmentation adds up to 5.17 points across the five tool-use evaluations, with gains varying substantially across models. We report bootstrap ranking-stability statistics to make bounded-budget variance explicit and to discourage over-interpretation of adjacent ranks.
arXiv:2606.05238v1 Announce Type: new
Abstract: LLM agents have made rapid progress on software engineering and ML research tasks, but these advances often assume access to a working runnable environment. For research artifacts released alongside published papers, setting up such an environment from a fresh machine remains a major bottleneck. Existing environment setup benchmarks do not cover the full scope of research artifact deployment, which involves multi-language toolchains, system-level dependencies beyond containers (e.g. GPU/CUDA and kernel configurations), and legacy artifact compatibility. We introduce DeployBench, a multi-domain benchmark of 51 research-artifact deployment tasks spanning AI/ML, computer systems, and scientific computing, covering all these dimensions. Each task is verified by a hidden pipeline that executes the paper's designated experiment and checks its outputs. Evaluating four state-of-the-art LLMs with OpenHands yields pass-rates from 7.8% - 51.0% . Failures are dominated by a completion-judgment problem: 97 of 154 are agent-terminated self-stops, where the agent's pre-finish checks validate a different or weaker target than the paper-specific task requires. DeployBench highlights the gap between current agents and autonomous deployment, and offers a realistic testbed for scientific research agents.
arXiv:2606.05186v1 Announce Type: new
Abstract: Budget-constrained micro-pretraining often requires triaging many candidate recipes on a shared accelerator before larger search budgets are spent. We study whether a staged fractional-factorial workflow can recover stable early effect structure in this setting. On a fixed autoresearch-derived single-GPU training loop, we run 613 experiments across pilot and follow-up screens at 2, 5, and 10 minutes; full 16-condition seeded reruns at 5 and 10 minutes; targeted seeded anchor checks; same-host greedy and matched-cost random baselines; a 60-minute bridge package; and bounded Windows A100 and Linux L40S anchor continuations through 24 hours. Main penalties from total batch, depth, and width are largest at short budgets and relax as budget increases. Within the predeclared seeded full-screen families, D, A, B, and C retain non-zero estimates at 5 and 10 minutes after within-budget Benjamini-Hochberg correction, while E does not. Random search can reach strong incumbents in this 32-condition space, but repeatedly in the same low-penalty region and without factor attribution. The 60-minute bridge anchor has the lowest mean, although that package does not separate workflow refinement from the larger bridge model's capacity advantage. In bounded 12-hour and 24-hour three-anchor continuations on both hosts, the bridge has the lowest sample mean while the non-bridge ordering stays host-sensitive. We therefore present a bounded methods result: use short designed screens to identify high-penalty directions, confirm promising anchors under repeated runs, and refine locally inside the reduced space. The evidence supports a bridge-centered recommendation through 24 hours on two hosts, not hardware-invariant ranking or general hyperparameter-optimization superiority.
arXiv:2606.05995v1 Announce Type: new
Abstract: Verbal harassment is a growing source of psychological stress for people around the world. It occurs both online and offline and relies on language to demean, threaten, or discredit its targets. Unlike other stressors such as loss or uncertainty, verbal harassment aims at silencing its targets by eroding their sense of being heard and weakening their perceived ability to respond. Many individuals lack access to adequate and timely support, however, when they experience such harassment. People increasingly turn to conversational artificial intelligence (AI) such as ChatGPT or dedicated AI companions for emotional support, raising questions about whether it can facilitate the same psychological benefits as actual human empathy. We focus on online contexts as a prevalent application of verbal harassment. We develop and test a psychological framework identifying three key linguistic signals of empathic listening (perspective-taking, emotional validation, and action orientation), that together restore a sense of feeling heard and enhance coping in the context of verbal harassment. We find that LLMs consistently produce language exhibiting stronger empathic-listening markers than human non-experts and trained mental health professionals, promoting more approach-oriented (vs. avoidance-oriented) coping strategies. A subsequent behavioral study shows that these linguistic signals boost recipients' sense of feeling heard and increase their coping self-efficacy. These findings reveal how specific linguistic features create empathic connections between humans and advanced conversational AI and can enhance people's psychological resilience. Our results highlight the potential for AI to serve as a scalable source of emotional support, especially when human support is unavailable or insufficient.
arXiv:2606.06004v1 Announce Type: new
Abstract: Dialect resources occupy a unique position at the intersection of scientific description, cultural preservation, and computational infrastructure. Large language models offer powerful capabilities for accelerating dialect resource development through retrieval-grounded drafting, corpus navigation, metadata enrichment, and annotation workflow support. However, the same systems pose substantial risks: they can contribute to dialect erasure by privileging prestige varieties, homogenizing orthography, and enabling synthetic feedback loops that reduce linguistic diversity over time. These risks are particularly acute for language varieties characterized by diglossia, limited written standardization, or marginalized speaker communities. This paper makes three contributions. First, we integrate insights from variationist sociolinguistics and corpus linguistics to formalize the generator-eraser paradox as a theoretical framework for understanding the dual nature of LLM-assisted dialect work. Second, we derive 12 community guidelines that operationalize this framework into implementable design requirements for dialect resource creation and documentation. Third, we provide an in-depth case study of Arabic dialects, including a structured comparison of widely used resources, to demonstrate how these guidelines address language-specific challenges including diglossia, orthographic variability, and community governance. The contribution is conceptual and operational rather than experimental, with the goal of enabling dialect communities and resource builders across languages to adopt LLMs without sacrificing authenticity, variation, or sovereignty.
arXiv:2606.06006v1 Announce Type: new
Abstract: Nitric oxide (NO) is important in biological, atmospheric, plasma, industrial, and astrophysical environments, where reliable electron-collision data support modelling charged-particle interactions with matter. Its well-known experimental properties make it suitable for assessing how the target electronic-structure description affects low-energy electron scattering calculations. In this work, NO properties were evaluated using B3LYP, M06-2X, PBE0, and $\omega$B97X-D3, with basis sets ranging from minimal to quadruple-zeta quality. Bond length, dipole moment, ionisation potential, and polarisability were compared with experiment to assess the sensitivity of the target description to the functional and basis set. The aug-cc-pVQZ basis set was then used to generate target models for ab initio R-matrix calculations over 0.1--20 eV. The total cross sections show low-energy resonance features, with the strongest functional dependence around the broad peak near 0.8--1.0 eV. A sharper, higher-energy structure is also observed below 2 eV, shifting from 1.74 to 1.82 eV depending on the functional. Differential cross sections show modest functional sensitivity, with more noticeable angular differences at 7.5 and 10 eV. These results show that the DFT functional and basis set affect the target properties, with the resulting target description influencing low-energy electron-scattering observables of NO. The comparison supports $\omega$B97X-D3/aug-cc-pVTZ geometry optimisation followed by aug-cc-pVQZ target-property calculations as a practical protocol for R-matrix modelling of NO.
arXiv:2606.06011v1 Announce Type: new
Abstract: In this work, we propose a framework that combines multi-agent reinforcement learning (MARL) with model-based control to achieve safe, dynamically feasible actions in cooperative multi-agent tasks. Multi-agent reinforcement learning provides the advantage of learning cooperative policies for multi-agent teams from discrete non-differentiable rewards in a long planning horizon. Model-predictive control is robust and offers safe, dynamically feasible actions in a fast replanning framework for short horizons. We propose an algorithm that extends actor-critic model predictive control for MARL which we refer to as multi-agent actor-critic model predictive control (MA-AC-MPC). We demonstrate the capabilities of this algorithm by applying it to a multi-agent pursuit-evasion scenario. Specifically, we compare the evader team's strategy using the MA-AC-MPC model and a multi-layer perceptron model (MA-AC-MLP). The pursuer team uses augmented proportional navigation as it is accepted as an advanced adversarial control law. We also provide an example with a heterogeneous environment where a drone and omni-wheeled rover cooperate to achieve repeatable and successful landing with 100% success rate in hardware for MA-AC-MPC compared to 60% for MA-AC-MLP. We demonstrate the robustness of the proposed MA-AC-MPC algorithm in hardware for both environments.
arXiv:2605.11632v2 Announce Type: replace
Abstract: Self-generated counterfactual explanations (SCEs) are minimally modified inputs (minimality) generated by large language models (LLMs) that flip their own predictions (validity), offering a causally grounded approach to unraveling black-box LLM behavior. Yet extending them beyond English remains challenging: existing methods struggle to produce valid SCEs in non-dominant languages, and a persistent trade-off between validity and minimality undermines explanation quality. We introduce Macro, a preference alignment framework that applies Direct Preference Optimization (DPO) to multilingual SCE generation, using a composite scoring function to construct preference pairs that effectively translate the trade-off into measurable preference signals. Experiments across four LLMs and seven typologically diverse languages show that Macro improves validity by 12.55\% on average over the chain-of-thought baseline without degrading minimality, while avoiding the severe minimality violations of the translation-based baseline. Compared to supervised fine-tuning, Macro achieves superior performance on both metrics, confirming that explicit preference optimization is essential for balancing this trade-off. Further analyses reveal that Macro increases cross-lingual perturbation alignment and mitigates common generation errors. Our results highlight preference optimization as a promising direction for enhancing multilingual model explanations.
arXiv:2606.05371v1 Announce Type: new
Abstract: Reduced-order modeling of high-dimensional dynamical systems is often hindered by the non-Markovian closure term that represents the effect of unresolved variables on the resolved dynamics. Inspired by the Mori--Zwanzig formalism, in which the closure takes the form of a memory functional of the resolved trajectory, we recast closure modeling as a sequence modeling problem and propose the Mamba-Assisted Closure (MAC) framework: a Mamba-based sequence model, trained to predict the closure from the resolved trajectory, is coupled with the reduced-order governing equations through a numerical integrator to advance the resolved variables in time. A key feature of the framework is its exploitation of the dual representation of state-space models -- the model is trained in a sequence-to-sequence fashion via the convolutional form, and deployed for step-by-step autoregressive rollout via the recurrent form, yielding both efficient long-trajectory training and constant per-step inference cost. On the viscous Burgers' equation and the chaotic two-scale Lorenz '96 system, the MAC model substantially outperforms the Markovian reduced-order model, the GRU-based sequence model, and the Wilks method in predictive accuracy and long-time rollout stability.
arXiv:2605.13830v2 Announce Type: replace
Abstract: Decision tree ensembles (DTE) are a popular model for a wide range of AI classification tasks, used in multiple safety critical domains, and hence verifying properties on these models has been an active topic of study over the last decade. One such verification question is the problem of sensitivity, which asks, given a DTE, whether a small change in subset of features can lead to misclassification of the input. In this work, our focus is to build a quantitative notion of sensitivity, tailored to DTEs, by discretizing the input space of the model and enumerating the regions which are susceptible to sensitivity. We propose a novel algorithmic technique that can perform this computation efficiently, within a certified error and confidence bound. Our approach is based on encoding the problem as an algebraic decision diagram (ADD), and further splitting it into subproblems that can be solved efficiently and make the computation compositional and scalable. We evaluate the performance of our technique over benchmarks of varying size in terms of number of trees and depth, comparing it against the performance of model counters over the same problem encoding. Experimental results show that our tool XCount achieves significant speedup over other approaches and can scale well with the increasing sizes of the ensembles.
arXiv:2606.06087v1 Announce Type: new
Abstract: Agent systems increasingly use textual skills to encode reusable task procedures, but injecting these skills into the prompt at every step incurs substantial context overhead and exposes skill content as plaintext. We present LatentSkill, a framework that converts textual skills into plug-and-play LoRA adapters through a pretrained hypernetwork. LatentSkill stores skill knowledge in weight space rather than context space, removing per-step skill tokens while preserving modular loading, scaling, and composition. On ALFWorld and Search-QA, LatentSkill outperforms the corresponding in-context skill baseline while using substantially fewer prefill tokens: it improves ALFWorld success by 21.4 and 13.4 points on the seen and unseen splits with 64.1% fewer prefill tokens, and improves Search-QA exact match by 3.0 points with 72.2% lower skill-token overhead. Further analysis shows that generated skill LoRAs form a structured semantic geometry, can be precisely controlled via the LoRA scaling coefficient, and can be composed through parameter-space arithmetic when skill components are aligned. These findings suggest that weight-space skills provide an efficient, modular, and less exposed substrate for extending LLM agents.
arXiv:2603.03993v2 Announce Type: replace
Abstract: Multi-head attention enables transformer models to represent multiple attention patterns simultaneously. Empirically, head specialization emerges in distinct stages during training, while many heads remain redundant and learn similar representations. We propose a theoretical model capturing this phenomenon, based on the multi-index and single-location regression frameworks. In the first part, we analyze the training dynamics of multi-head softmax attention under SGD, revealing an initial unspecialized phase followed by a multi-stage specialization phase in which different heads sequentially align with latent signal directions. In the second part, we study the impact of attention activation functions on performance. We introduce the Bayes-softmax attention, which achieves optimal prediction performance in this setting.
arXiv:2606.03067v2 Announce Type: replace-cross
Abstract: A recurring data mining task in complex networks is to determine how individual nodes contribute to system behavior. Existing approaches rely on either static-graph centralities or control-theoretic quantities such as controllability Gramians, which assume linear, time-invariant dynamics. Estimated systems, however, are typically nonlinear and time-varying. We define "emergent contribution (EC)," a finite-horizon measure of a node's dynamical leverage: the metric-weighted energy of its impulse response accumulated along the system trajectory. Computed from the Jacobians of any differentiable model, EC is estimator-agnostic and reduces exactly to average controllability in the linear, time-invariant limit. Our contribution is a characterization of when the two measures agree and diverge. Using a controlled synthetic family with known ground-truth contribution, we construct a phase diagram spanning nonlinearity, regime structure, persistence, and perturbation amplitude. EC and average controllability agree under static or smoothly drifting dynamics and both track ground truth. Divergence emerges under persistent regime switching, is strongest under persistent sign reversal, and disappears when the sign reversal is removed. At extreme perturbation amplitudes, both measures degrade, identifying the limits of local linearization. We place five estimated real systems from several domains within this phase space. Their placement serves as a diagnostic of when EC provides information beyond static controllability and therefore justifies its additional computational cost. On one panel examined in depth, a twenty-seed retraining ensemble reveals a robust variance--leverage dissociation: nodes whose perturbations propagate widely despite low within-system variance, which is not recovered by static centralities nor variance-based summaries.
arXiv:2606.06219v1 Announce Type: new
Abstract: End-to-end autonomous driving models often struggle to balance multi-modal maneuver generation with real-time inference constraints. While diffusion models successfully capture diverse driving behaviors, their iterative denoising process incurs unacceptable latency for safety-critical deployment. To address this, we propose CLEAR (Cognition and Latent Evaluation for Adaptive Routing), a framework that combines ultra-fast generative planning with deep semantic reasoning. CLEAR employs Drive-JEPA as the visual encoder and replaces the multi-step denoising chain with a single-step conditional drift in a VAE latent space, introducing a conditioning coefficient to balance diversity and expert precision. Meanwhile, we fully fine-tune Qwen~3.5~0.8B on driving QA pairs to extract scene-aware hidden states. These states guide both an Adaptive Scheduler, which selects the conditioning coefficient $\alpha$ and sample count $N$ from a discrete set of predefined schemes, and a cross-attention scorer that selects the optimal trajectory from candidates. On the NAVSIM v1 benchmark, CLEAR achieves a state-of-the-art PDMS of 93.7. Our results demonstrate that high-fidelity, multi-modal planning can be executed efficiently without dense geometric annotations or iterative sampling.
arXiv:2606.05400v1 Announce Type: new
Abstract: Long-horizon autoformalization of research mathematics fails not only at hard lemmas, but at scale: statements drift, dependencies tangle, context decays, and local repairs corrupt distant work. We present LeanMarathon, a multi-agent harness for reliable research-level Lean autoformalization. Its core abstraction is an evolving blueprint: a Lean file that serves simultaneously as formal proof skeleton, natural-language proof graph, and shared system of record. Four contract-scoped agents construct, audit, prove, and repair this blueprint. These agents are coordinated by a two-stage orchestrator that first stabilizes target fidelity through adversarial review and then discharges the proof directed acyclic graph (DAG) from its dynamic leaves upward in parallel CI-gated rounds. LeanMarathon turns one brittle multi-hour run into many local, recoverable, parallel transactions. We evaluate LeanMarathon on two recent research papers spanning four Erd\H{o}s problems (#1051, #1196, #164, #1217). Across three autonomous runs, it formalizes all seven target theorems with no sorry, proving 258 lemmas and theorems. These results show that reliable AI co-mathematics requires not only stronger provers, but durable harnesses that preserve target fidelity across long mathematical developments. The code can be found at https://github.com/YuanheZ/LeanMarathon.
arXiv:2606.06492v1 Announce Type: new
Abstract: Code language models need repository-level context to resolve imports, APIs, and project conventions. Existing methods inject this knowledge as long inputs (retrieved through RAG or dependency analysis) or through per-repository fine-tuning and LoRA -- costly at repository scale and brittle to evolving codebases. We introduce Code2LoRA, a hypernetwork framework that generates repository-specific LoRA adapters, effectively injecting repository knowledge with zero inference-time token overhead. Code2LoRA supports two usage scenarios: Code2LoRA-Static converts a single repository snapshot into an adapter, suitable for comprehension of stable codebases; while Code2LoRA-Evo maintains an adapter backed by a GRU hidden state updated per code diff, suitable for active development of evolving codebases. To evaluate Code2LoRA against parameter-efficient fine-tuning baselines, we build RepoPeftBench, a benchmark of 604 Python repositories with two tracks: a static track with 40K training and 12K test assertion-completion tasks, and an evolution track with 215K commit-derived training and 87K commit-derived test tasks. On the static track, Code2LoRA-Static achieves 63.8% cross-repo and 66.2% in-repo exact match, matching the per-repository LoRA upper bound; on the evolution track, Code2LoRA-Evo achieves 60.3% cross-repo exact match (+5.2 pp over a single shared LoRA). Code2LoRA's code can be found at https://anonymous.4open.science/r/code2lora-6857; the model checkpoints and RepoPeftBench datasets can be found at https://huggingface.co/code2lora.
arXiv:2606.03730v2 Announce Type: replace
Abstract: Vision-language models (VLMs) such as CLIP show strong zero-shot generalization but remain highly vulnerable to adversarial attacks. Adversarial training improves robustness but is computationally expensive, motivating test-time defenses. Recent approaches exploit how CLIP's visual representations respond to stochastic perturbations: aggregating predictions across noisy views, constructing Gaussian noise-averaged anchors and interpolating features toward them, or applying counter-perturbations. These strategies improve robustness but often degrade clean accuracy, yielding an unfavorable clean-robust trade-off. We revisit stochastic test-time defenses and identify an underexplored noise-regime transition in CLIP's representation space. Prior work explored perturbations mainly in the weak-noise regime, where adversarial examples can appear unusually stable (false stability). Our analysis shows this reverses as perturbation strength grows: beyond the weak-noise regime, adversarial representations become markedly more unstable than clean ones, giving a clearer separation signal. The transition is consistent across uniform and Gaussian noise, photometric and geometric transforms, datasets, and diverse attacks. It largely disappears in adversarially trained models, suggesting it is tied to the fragile local-basin geometry of adversarial representations in non-robust CLIP. We propose a training-free, plug-in drift-gated mechanism that uses high-noise feature drift as a lightweight gating signal to trigger existing test-time defenses only when adversarial-like instability is detected. Across 13 datasets it consistently improves the clean-robust trade-off. On eight fine-grained datasets, mean clean+adversarial accuracy rises from 65.7% to 71.4% for counterattack defenses and 68.4% to 73.2% for noise-anchoring; on ImageNet and four shifted variants, from 56.1% to 66.2% and 62.1% to 67.6%.
arXiv:2606.06147v1 Announce Type: new
Abstract: End-to-end Vision-Language-Action (VLA) models have shown promise in UAV navigation. However, existing approaches typically rely on historical observations to directly predict actions, often struggling in dense urban environments where severe occlusions and sharp turns result in drastic viewpoint transitions. We argue that the ability to "imagine" future states -- inherent in World Models -- is critical for robust decision-making under such partial observability. To address this, we construct a challenging Urban Canyon Traversal Benchmark, specifically designed to evaluate spatial understanding in scenarios characterized by severe occlusions and drastic viewpoint transitions. To this end, we propose WorldFly, a novel world-model-based VLA framework that employs a dual-branch coupled flow matching mechanism to jointly generate future video predictions and navigation actions, thereby explicitly guiding the agent's policy via spatial imagination. Extensive evaluations on our benchmark demonstrate that WorldFly outperforms other baselines, particularly in unseen environments, validating the effectiveness of integrating world models into embodied aerial agents.
arXiv:2602.16149v3 Announce Type: replace
Abstract: Instruction-guided image-to-image (I2I) editors are increasingly used in consumer and professional visual workflows, where trustworthiness depends not only on prompt compliance but also on equitable preservation of identity-relevant attributes. We formalize two failure modes: Soft Erasure, where requested edits are weakly realized or silently suppressed, and Stereotype Replacement, where edits introduce unrequested, stereotype-consistent demographic attributes. Using a controlled benchmark of 5,040 edited portraits, we evaluate these failures across three recent open-weight editors with vision-language model scoring and human evaluation. Our results show that identity-preservation failures are pervasive and demographically uneven. In particular, 62--71% of outputs exhibit skin lightening, with Indian and Black source portraits affected at 72--75%, compared with 44% for White source portraits, indicating output-level drift toward lighter or more White-presenting appearances when identity constraints are underspecified. In a mitigation case study, prompt-level appearance constraints reduce race-change scores for non-White source portraits by up to 1.48 points, while leaving White source portraits largely unchanged, without modifying model weights. These findings show that identity preservation is not a uniform property of I2I portrait editing systems, but an unevenly distributed trustworthiness failure with direct social consequences. At deployment scale, such silent distortions can shape AI-mediated self-representation and reinforce representational disparities. We introduce a controlled audit protocol for fairness-aware evaluation and governance of generative editing systems. Project page: https://seochan99.github.io/i2i-demographic-bias
arXiv:2605.18937v2 Announce Type: replace
Abstract: Patient-managed Personal Health Records (PHRs) promises to empower patients to better understand their health; but information in the record is complex, potentially hindering insights. In this study, we assess the potential of large language models (LLMs, Gemini 3.0 Flash) to provide helpful answers to user health queries, when provided clinical data from PHRs as context. A total of 2,257 user queries were drawn from 3 different distributions to represent patient questions: shorter web search queries, longer questions derived from templates of chatbot conversations, and questions patients asked to their healthcare team (patient calls). Queries were matched with de-identified PHRs (from a pool of 1,945). Gemini responses were generated (1) without PHR context; (2) with a basic summary of demographics, conditions, and medications; (3) with full, extensive clinical notes. For evaluation, we leveraged an existing rating framework (SHARP), and developed a new framework for specific error modes when interpreting PHRs. Evaluation was performed using autoraters for the full set, and with clinician ratings for a subset (n=95), with both sets of raters knowing the full PHR context. We see significant improvements in the helpfulness of answers to all question types with PHR data (p < 0.001, paired t-test). We also observe potential gains in safety, accuracy, relevance and personalization of answers. Our PHR evaluation framework further identifies gaps in LLM understanding of particular aspects of complex PHRs, such as temporal disorientation, and rare but meaningful confabulations. These results suggest potential for PHR data to help people with a wide range of user needs; and provide a framework for monitoring for gaps in LLM answers based on PHR context. This study motivates further work to assess and realize potential benefits to users from understanding their health records.
arXiv:2606.06174v1 Announce Type: new
Abstract: Childhood asthma is a common illness exacerbated by air pollution as well as meteorological and neighborhood-level socioeconomic factors. Modeling asthma exacerbation (AE) in large spatiotemporal datasets requires disentangling impacts from multiple contributors. In this case study, we compared three techniques that balance predictive power with interpretability to predict AE in Hampton Roads, a coastal Virginia region comprising 7 cities and over 1.5 million people. After collating ambient air pollution measurements, weather data, and measures of neighborhood opportunity, we modeled zip code-level acute AE visits to a regional children's hospital and affiliated providers from 2018-2023. Generalized linear models (GLM) provided a baseline while neural networks (NN) served as a maximally predictive target. To bridge between statistical models and deep learning, we developed a framework based on sparse dictionary learning to identify and interpret parsimonious nonlinear interacting equations. After comparing each model's predictive performance, we estimated relative risks for AE due to input exposure variables and found consensus across frameworks. Our work links statistical and interpretable machine learning models to highlight possible synergistic interactions influencing AE, and may enable future studies to guide public health interventions in coastal Virginia.
arXiv:2606.06177v1 Announce Type: new
Abstract: Speech translation (ST) is increasingly adopted in user applications, yet its evaluation largely focuses on decontextualized testbeds and holistic quality, rather than end users' communication needs. We introduce Ouvia, an evaluation framework for measuring user-perceived usability of speech translation outputs in real-world settings. Ouvia focuses on one-to-one communication: an English speaker needs to convey a request to a Portuguese speaker, and the message is automatically translated. Through a custom web app and multi-phase study design, we collect more than 1,750 such interactions in healthcare and everyday situations, mediated by four ST systems, involving speakers from three English dialects and two genders. We find that modern ST serves people only to a limited extent -- only around half of interactions are rated as usable -- with significant gaps in reported usability across demographic groups. Moreover, among quality metrics, we find that QA-based evaluation is a substantially stronger predictor of real-world usability than standard approaches. Together, these findings stress the importance of situated, user-centered evaluation frameworks that go beyond holistic quality scores and attend to who the technology serves -- and how well.
arXiv:2606.06266v1 Announce Type: new
Abstract: Hate speech detection is inherently subjective: people from different demographic groups perceive the same content very differently. Collecting enough annotations from multiple demographic groups is costly and difficult to scale. Persona-conditioned Large Language Models (models prompted to adopt a specific demographic identity) have been proposed as a way to simulate diverse perspectives at scale. But do they actually reflect how different groups disagree? We evaluate three aspects of human social judgement: (i) whether personas from different groups disagree in human-like ways (inter-group disagreement), (ii) whether they become more sensitive when content targets their own identity (in-group sensitivity), and (iii) whether they can accurately predict how another group would react (vicarious prediction). Our results show that no model consistently captures all three dimensions, and performance is highly model-dependent and does not emerge reliably from minimal identity prompts alone. However, vicarious prompting with Llama 3.1 yields the highest cross-group agreement in most demographic axes and provides the closest overall approximation to human disagreement patterns, indicating that this configuration may provide a more reliable setting for automatic annotation aligned with human judgements.