# Research Papers — 2026-05 · bundle page 318 of 326 > Full-text markdown of papers #15851–15900 of 16266 published in 2026-05, newest first. AI-readable mirrors of the full 16266-paper 2026-05 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-05/page/318 [← Previous bundle](https://aipod.dev/papers/archive/2026-05/page/317.md) [Next bundle →](https://aipod.dev/papers/archive/2026-05/page/319.md) ### Rhamba: Region-Aware Hybrid Attention-Mamba Framework for Self-Supervised Learning in Resting-State fMRI Authors: Ruthwik Reddy Doodipala, Pankaj Pandey, Pratheek Eranki, Carolina Torres-Rojas, Manob Jyoti Saikia, Ranganatha Sitaram Published: 2026-05-02 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.01240v2 Self-supervised pretraining is promising for large-scale neuroimaging, yet the impact of region-aware masking and hybrid sequence modeling remains underexplored. In this work, we introduce Rhamba, a region-aware pretraining framework that integrates anatomically guided masking with hybrid Attention-Mamba architectures for resting state functional magnetic resonance imaging (fMRI) analysis. Models were pretrained on the ABIDE dataset using region-aligned patch embeddings and three masking strategies (Any, Majority, and Pure) with increasing spatial specificity. We evaluated four architectural variants: a Mamba only model, an Alternate architecture with interleaved Mamba and Attention blocks, and two hybrid encoder-decoder configurations (Attention-Mamba (AM) and Mamba-Attention (MA)). The pretrained models were fine-tuned on downstream classification tasks using the COBRE and ADHD-200 datasets for schizophrenia and attention-deficit/hyperactivity disorder discrimination. We employed Integrated Gradients, an explainable AI method, to identify the brain regions contributing to model predictions. Masking strategy strongly influenced reconstruction behavior, with reconstruction loss following a consistent ordering (Any > Majority > Pure). However, this trend did not directly translate into downstream performance, where differences were modest and dataset-dependent. The hybrid architecture with the MA configuration achieved the highest average AUROC across both datasets, and Rhamba outperformed state-of-the-art methods in comparative evaluation. Region-wise analysis showed that peak performance depends on the interaction between masking strategy and architecture rather than a single dominant configuration. Overall, Rhamba offers a flexible framework for balancing interpretability, scalability, and performance in large-scale fMRI representation learning. --- ### EduGage: Methods and Dataset for Sensor-Based Momentary Assessment of Engagement in Self-Guided Video Learning Authors: Zikang Leng, Edan Eyal, Yingtian Shi, Jiaman He, Yaqi Liu, Thomas Plötz Published: 2026-05-02 Categories: cs.HC, cs.CV arXiv: https://arxiv.org/abs/2605.01238v1 Engagement, which links to attentional, emotional, and cognitive dimensions, plays an important role in learning. In online and video-based learning environments, learners often need to regulate their own interactions with instructional materials. Measuring and reflecting on engagement can therefore support both learners and adaptive learning systems. In this study, we use wearable and camera-based sensing devices to collect physiological and motion signals, including PPG, ECG, EDA, EEG, IMU, heart rate, temperature, and eye-tracking data, to estimate learner engagement. We conducted a user study with 16 participants in a video-based learning scenario, where participants completed learning tasks and provided repeated in-situ self-reports of engagement through brief probes. We develop and evaluate a system for engagement estimation, compare different sensing modalities, and further analyze the feasibility and effectiveness of multimodal modeling for characterizing learner engagement. Across participant-based cross-validation, our model achieves an MAE of 0.81, 83.75% within-1 accuracy, 73.93% binary accuracy, and 68.45% binary Macro-F1, outperforming sensor-free, statistical, deep temporal, foundation-model, and LLM-based baselines. Our results suggest that fine-grained engagement estimation is feasible but inherently noisy, and that practical systems should prioritize lightweight combinations of behavioral and physiological signals over full multimodal instrumentation. We release the EduGage dataset, including synchronized multimodal sensor signals, probe-aligned momentary engagement labels, video metadata, quizzes, and study materials, to support reproducible research on fine-grained sensor-based engagement modeling in self-guided learning. --- ### Degradation-Aware Adaptive Context Gating for Unified Image Restoration Authors: Lei He, Jielei Chu, Fengmao Lv, Weide Liu, Tianrui Li, Jun Cheng, Yuming Fang Published: 2026-05-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.01236v1 Unified image restoration using a single model often faces task interference due to diverse degradations. To address this, we propose DACG-IR (Degradation-Aware Adaptive Context Gating), which enables explicit perception of degradation characteristics to dynamically modulate feature representations. Our method constructs degradation-aware contextual representations from the input to modulate attention distribution, frequency-domain features, and feature aggregation. Specifically, a lightweight multi-scale degradation-aware module extracts coarse degradation information and generates layer-wise prompts. These prompts guide attention temperature and output gating in encoder and decoder blocks for adaptive feature extraction. Additionally, a spatial-channel dual-gated adaptive fusion mechanism refines encoder features, suppressing noise propagation from shallow to deep layers. This design effectively suppresses degradation-induced noise while preserving informative structures. Experiments show DACG-IR outperforms state-of-the-art methods in single-task, all-in-one, adverse weather removal, and composite degradation settings. Code: https://github.com/HlHomes/DACG-IR-code --- ### MindMelody: A Closed-Loop EEG-Driven System for Personalized Music Intervention Authors: Yimeng Zhang, Yueru Sun, Haoyu Gu, Zhanpeng Jin Published: 2026-05-02 Categories: cs.SD, cs.AI arXiv: https://arxiv.org/abs/2605.01235v2 Driven by the escalating global burden of mental health conditions, music-based interventions have attracted significant attention as a non-invasive, cost-effective modality for emotion regulation and psychological stress relief. However, current digital music services rely on static preferences and fail to adapt to users' instantaneous psychological states. Furthermore, directly mapping electroencephalography (EEG) to music generation remains challenging due to severe paired-data scarcity and a lack of interpretability. To address these limitations, we propose MindMelody, a fully functional, closed-loop real-time system for EEG-driven personalized music intervention. MindMelody introduces an emotion-mediated semantic bridge. Specifically, a hybrid Transformer-GNN first decodes real-time EEG signals into global Valence-Arousal states and local temporal affect trajectories. These states are then fed into a Retrieval-Augmented Generation (RAG)-equipped Large Language Model (LLM) to formulate structured intervention plans. Subsequently, a novel Hierarchical EEG Controller injects global affect prefixes and local temporal guidance into a pretrained music backbone, enabling fine-grained controllable audio synthesis. Crucially, the system incorporates a continuous feedback loop that updates generation parameters on the fly based on the user's evolving EEG dynamics. Extensive experiments show that MindMelody improves control adherence and emotional alignment, and receives higher perceived helpfulness in a short-term listening setting, suggesting its promise as an adaptive affect-aware music generation framework. --- ### TT4D: A Pipeline and Dataset for Table Tennis 4D Reconstruction From Monocular Videos Authors: Nima Rahmanian, Daniel Kienzle, Thomas Gossard, Dvij Kalaria, Rainer Lienhart, Shankar Sastry Published: 2026-05-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.01234v2 We present TT4D, a large-scale, high-fidelity table tennis dataset. It provides $140+$ hours of reconstructed singles and doubles gameplay from monocular broadcast videos, featuring multimodal annotations like high-quality camera calibrations, precise 3D ball positions, ball spin, time segmentation, and 3D human meshes over time. This rich data provides a new foundation for virtual replay, in-depth player analysis, and robot learning. The dataset's combination of scale and precision is achieved through a novel reconstruction pipeline. Prior methods first partition a game sequence into individual shot segments based on the 2D ball track, and only then attempt reconstruction. However, 2D-based time segmentation collapses under occlusion and varied camera viewpoints, preventing reliable reconstruction. We invert this paradigm by first lifting the entire unsegmented 2D ball track to 3D through a learned lifting network. This 3D trajectory then allows us to reliably perform time segmentation. The learned lifting network also infers the ball's spin, handles unreliable ball detections, and successfully reconstructs the ball trajectory in cases of high occlusion. This lift-first design is necessary, as our pipeline is the only method capable of reconstructing table tennis gameplay from general-view broadcast monocular videos. We demonstrate the dataset's fidelity through two downstream tasks: estimating the racket's pose \& velocity at impact, and training a generative model of competitive rallies. --- ### TT4D: A Pipeline and Dataset for Table Tennis 4D Reconstruction From Monocular Videos Authors: Nima Rahmanian, Daniel Kienzle, Thomas Gossard, Dvij Kalaria, Rainer Lienhart, Shankar Sastry Published: 2026-05-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.01234v1 We present TT4D, a large-scale, high-fidelity table tennis dataset. It provides $140+$ hours of reconstructed singles and doubles gameplay from monocular broadcast videos, featuring multimodal annotations like high-quality camera calibrations, precise 3D ball positions, ball spin, time segmentation, and 3D human meshes over time. This rich data provides a new foundation for virtual replay, in-depth player analysis, and robot learning. The dataset's combination of scale and precision is achieved through a novel reconstruction pipeline. Prior methods first partition a game sequence into individual shot segments based on the 2D ball track, and only then attempt reconstruction. However, 2D-based time segmentation collapses under occlusion and varied camera viewpoints, preventing reliable reconstruction. We invert this paradigm by first lifting the entire unsegmented 2D ball track to 3D through a learned lifting network. This 3D trajectory then allows us to reliably perform time segmentation. The learned lifting network also infers the ball's spin, handles unreliable ball detections, and successfully reconstructs the ball trajectory in cases of high occlusion. This lift-first design is necessary, as our pipeline is the only method capable of reconstructing table tennis gameplay from general-view broadcast monocular videos. We demonstrate the dataset's fidelity through two downstream tasks: estimating the racket's pose \& velocity at impact, and training a generative model of competitive rallies. --- ### CombinationTS: A Modular Framework for Understanding Time-Series Forecasting Models Authors: Xiaorui Wang, Fanda Fan, Chenxi Wang, Yuxuan Yang, Rui Tang, Kuoyu Gao, Simiao Pang, Yuanfeng Shang, Zhipeng Liu, Wanling Gao, Lei Wang, Jianfeng Zhan Published: 2026-05-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.01231v1 Recent progress in time-series forecasting has led to rapidly increasing architectural complexity, yet many reported State-of-the-Art gains are statistically fragile or misattributed. We argue that progress requires a shift from model selection to modular attribution, identifying which components truly drive performance. We propose CombinationTS, a self-contained probabilistic evaluation framework that decomposes forecasting models into orthogonal modules--Input Transformation, Embedding, Encoder, Decoder, and Output Transformation--and evaluates them under a shared evaluation condition space. By quantifying each component via marginalized performance ($μ$) and stability ($σ$), CombinationTS enables robust attribution beyond fragile point estimates. Through large-scale paired evaluation, we uncover the Identity Paradox: once the data view (Embedding) is well-designed, a parameter-free Identity Encoder often matches or outperforms complex backbones. We further show that explicit structural priors introduced via Input Transformations yield a more favorable performance-stability trade-off than increasing Encoder complexity, establishing a principled baseline for architectural necessity. --- ### Attention Sinks in Massively Multilingual Neural Machine Translation:Discovery, Analysis, and Mitigation Authors: Hillary Mutisya, John Mugane Published: 2026-05-02 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2605.01229v1 Cross-attention patterns in neural machine translation (NMT) are widely used to study how multilingual models align linguistic structure. We report a systematic artifact in cross-attention analysis of NLLB-200 (600M): non-content tokens - primarily end-of-sequence tokens, language tags, and punctuation - capture 83 percent to 91 percent of total cross-attention mass. We term these "attention sinks," extending findings from LLMs (Xiao et al., 2023) to NMT cross-attention and identifying a causal mechanism rooted in vocabulary design rather than position bias. This artifact causes raw metrics to underestimate content-level similarity by nearly half (36.7 percent raw vs. 70.7 percent filtered), rendering uncorrected analyses unreliable. To address this, we validate a content-only filtering methodology that removes non-content tokens and renormalizes the distribution. Applying this to 1,000 parallel sentences across African languages (Swahili, Kikuyu, Somali, Luo) and non-African benchmarks (German, Turkish, Chinese, Hindi), we confirm the artifact is universal and recover masked linguistic signals: a 16.9 percentage-point gap between teacher-forcing and generation modes, clear language-family clustering in attention entropy, and a hidden Somali paradox linking SOV word order to monotonic alignment. We release our filtering toolkit and corrected datasets to support reproducible interpretability research on multilingual NMT. --- ### Arbitrarily Conditioned Hierarchical Flows for Spatiotemporal Events Authors: Keyan Chen, Qiwei Yuan, Zhitong Xu, Bin Shen, Shandian Zhe Published: 2026-05-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.01226v1 Events in spatiotemporal systems are ubiquitous, yet modeling their complex distributions remains challenging. Existing point process models often rely on strong structural assumptions and are typically limited to autoregressive, event-by-event prediction. As a result, they struggle to support broader inference tasks such as inverse inference, trajectory reconstruction, and recovery of missing event locations. We introduce Arbitrarily Conditioned Hierarchical Flows (ARCH), a hierarchical flow matching framework for spatiotemporal event modeling. ARCH is expressive enough to capture complex event distributions while enabling tractable and accurate computation of conditional intensities, which quantify instantaneous event risk. Built on a history-encoder-generative-decoder architecture, ARCH introduces a hybrid masking strategy for flexible conditioning on arbitrary observed events. This enables a unified treatment of forecasting, inverse inference, and partial trajectory recovery within a single framework. Experiments on synthetic and real-world datasets show that ARCH consistently outperforms existing baselines across both prediction and conditional inference tasks. --- ### Lost in the Tower of Babel: The Adverse Effects of Incidental Multilingualism in LLMs Authors: Anjishnu Mukherjee, Chutong Meng, Antonios Anastasopoulos Published: 2026-05-02 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.01224v2 This paper argues that contemporary multilingual NLP has converged on a fragile and misleading paradigm of incidental multilingualism. Today's LLMs appear multilingual largely because they are trained on massive, uneven web corpora, not because multilingual or multicultural competence has been treated as a core design objective. We contend that this paradigm systematically produces unequal, brittle, and opaque behavior across languages, with severe consequences in real-world and agentic deployments where models must reason, plan, and act across multiple linguistic contexts. We report a focused empirical study of two practical questions: which languages models self-report as supported and which languages they actually respond in across multilingual prompts. We additionally demonstrate how even a simple language-change attack can surface these failures and expose hidden assumptions about language in LLM-based systems. To address this, we call for a shift toward multilingualism by design: a research agenda that treats equitable multilingual performance, cultural grounding, and cross-lingual behavioral understanding as first-class goals in all aspects of the model pipeline. --- ### Zero-Shot Signal Temporal Logic Planning with Disjunctive Branch Selection in Dynamic Semantic Maps Authors: Bowen Ye, Ancheng Hou, Junyue Huang, Ruijia Liu, Xiang Yin Published: 2026-05-02 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.01222v1 Signal Temporal Logic (STL) offers verifiable task specifications and is crucial for safety-critical control. Yet STL planning remains challenging: exact optimization-based methods are often too slow, and learning-based methods struggle to generalize across varying environments. We propose a zero-shot STL planning solver for variable-map environments that generates feasible trajectories without retraining. By integrating a map-conditioned Transformer architecture with a lightweight heuristic, our approach effectively handles complex disjunctive (OR) subformulas. Furthermore, we leverage Transitive Reinforcement Learning (TRL) to ensure consistent temporal grounding and logical coherence across decomposed sub-tasks. Experiments on dynamic semantic maps with diverse obstacle layouts demonstrate consistent gains, highlighting the framework's superior zero-shot generalization to changing environments and broad STL coverage. --- ### Local Hessian Spectral Filtering for Robust Intrinsic Dimension Estimation Authors: Genki Osada Published: 2026-05-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.01221v2 While diffusion models enable new approaches for estimating Local Intrinsic Dimension (LID), existing methods fail in high-dimensional spaces where noise from vast normal directions overwhelms the tangent signal. We propose Local Hessian Spectral Dimension (LHSD), which resolves this by applying spectral filtering to the log-density Hessian, explicitly cutting off large eigenvalues associated with normal directions to count zero-curvature tangent directions. Implemented using Stochastic Lanczos Quadrature (SLQ), LHSD avoids full Hessian construction, achieving linear scalability with dimension $D$. Experiments on synthetic and real data confirm LHSD's superior robustness and its utility in detecting memorization in large-scale diffusion models. The code is available at github.com/geosada/LHSD --- ### Visual Implicit Autoregressive Modeling Authors: Pengfei Jiang, Jixiang Luo, Luxi Lin, Zhaohong Huang, Xuelong Li Published: 2026-05-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.01220v2 Visual Autoregressive Modeling (VAR) based on next-scale prediction achieves strong generation quality, but their explicit deep stacks fix the amount of computation per scale and inflate memory at high resolutions. We introduce Visual Implicit Autoregressive Modeling (VIAR), a next-scale autoregressive generator that embeds an implicit equilibrium layer between shallow pre/post blocks. The implicit layer is trained with Jacobian-Free Backpropagation, yielding constant training memory, while inference exposes a per-scale iteration knob that enables compute control. On ImageNet 256x256 benchmark, VIAR attains FID 2.16, and sFID 8.07 with only 38.4% parameters of VAR, matching or surpassing strong AR baselines and remaining competitive with large diffusion models. By controlling the per-scale knob, VIAR can reduce peak memory from 19.24 GB to 8.53 GB and doubles throughput from 15.16 to 32.08 images/s on a single RTX 4090, without retraining. Ablations show that fewer steps are sufficient for fixed-point iterations to converge and that VIAR consistently dominates VAR across quality efficiency operating points. In zero shot in-painting and class-conditional editing, VIAR produces sharper details and smoother boundaries while preserving global structure, validating the benefits of implicit equilibria and per-scale compute control for practical, deployable visual generation. --- ### Multimodal Confidence Modeling in Audio-Visual Quality Assessment Authors: Mayesha Maliha R. Mithila, Mylene C. Q. Farias Published: 2026-05-02 Categories: cs.MM, cs.CV, cs.SD, eess.IV arXiv: https://arxiv.org/abs/2605.01219v1 Audio-visual quality assessment (AVQA) is essential for streaming, teleconferencing, and immersive media. In realistic streaming scenarios, distortions are often asymmetric, where one modality may be severely degraded while the other remains clean. Still, most contemporary AVQA metrics treat audio and video as equally reliable, causing confidence-unaware fusion to emphasize unreliable signals. This paper proposes MCM-AVQA, a multimodal confidence-aware AVQA framework that explicitly estimates modality-specific confidence and injects it into a dedicated audio-visual mixer for cross-modal attention. The Audio-Visual Mixer utilizes frame-level, confidence-guided channel attention to gate fusion, modulating feature interaction between modalities so that high-confidence streams dominate while unreliable inputs are suppressed, preserving temporal degradation patterns. A multi-head visual confidence estimator turns frame-level artifact probabilities into temporally smoothed, clip-level visual confidence scores, while an audio confidence module derives confidence from speech-quality cues without requiring a clean reference. Experiments on multiple AVQA benchmarks show that MCM-AVQA, and specifically its confidence-guided Audio-Visual Mixer, improve correlation with human mean opinion scores and yield more interpretable behavior under real-world asymmetric audio-visual distortions. --- ### Asymmetric Invertible Threat: Learning Reversible Privacy Defense for Face Recognition Authors: Jiabei Zhang, Ziyuan Yang, Andrew Beng Jin Teoh, Yi Zhang Published: 2026-05-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.01217v1 Face Recognition systems are widely deployed in real-world applications, but they also raise privacy concerns due to unauthorized collection and misuse of facial data. Existing adversarial privacy protection methods rely on input-space perturbations to obfuscate identity information, yet their protection can degrade when adversaries learn restoration or purification mappings that partially invert the transformation. We study this setting as an asymmetric adversarial attack, in which reverse manipulation becomes feasible because existing defense paradigms do not control reversibility. To address this problem, we propose Asymmetric Reversible Face Protection (ARFP), a restoration-aware extension of personalized face cloaking that integrates privacy protection, keyed recovery, and tamper indication in a single framework. ARFP consists of three components: Key-Conditioned Manifold Binding, which ties the protection transformation to a user-provided key; Adversarial Restoration-Aware Training, which introduces a surrogate restoration adversary during training to improve robustness against evaluated inverse purification attacks; and Authorized Reversible Restoration, which supports recovery with the correct key while providing nonce-based tamper indication. Extensive experiments under the threat models considered in this work show that ARFP improves resistance to the evaluated restoration attacks while preserving authorized recovery utility. These results provide empirical evidence of key-sensitive recovery behavior and tamper awareness in the tested settings. --- ### Agentic AI Systems Should Be Designed as Marginal Token Allocators Authors: Siqi Zhu Published: 2026-05-02 Categories: cs.AI, cs.CY arXiv: https://arxiv.org/abs/2605.01214v1 This position paper argues that agentic AI systems should be designed and evaluated as \emph{marginal token allocation economies} rather than as text generators priced by the unit. We follow a single request -- a developer asking a coding agent to fix a failing test -- through four economic layers that today are designed in isolation: a router that decides which model answers, an agent that decides whether to plan, act, verify, or defer, a serving stack that decides how to produce each token, and a training pipeline that decides whether the trace is worth learning from. We show that all four layers are solving the \emph{same} first-order condition -- marginal benefit equals marginal cost plus latency cost plus risk cost -- with different index sets and different prices. The framing is deliberately minimal: we do not propose a complete theory of AI economics. But adopting marginal token allocation as the shared accounting object explains why systems that locally minimize tokens globally misallocate them, predicts a small set of recurring failure modes (over-routing, over-delegation, under-verification, serving congestion, stale rollouts, cache misuse), and points to a concrete research agenda in token-aware evaluation, autonomy pricing, congestion-priced serving, and risk-adjusted RL budgeting. --- ### Faithful Mobile GUI Agents with Guided Advantage Estimator Authors: Haowen Hu, Pengzhou Cheng, Zheng Wu, Lingzhong Dong, Gongshen Liu, Zhuosheng Zhang Published: 2026-05-02 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.01208v1 Vision-language model based graphical user interface (GUI) agents have shown strong interaction capabilities. However, they often behave unfaithfully, relying on memorized shortcuts rather than grounding actions in displayed screen evidence or user instructions. To address this, we propose Faithful-Agent, a faithfulness-first framework that reformulates GUI interaction to prioritize evidence groundedness and internal consistency. Faithful-Agent employs a two-stage pipeline: (i) a faithfulness-oriented SFT stage to instill abstainment behaviors under evidence perturbations; (ii) an RFT stage that further amplifies faithfulness by introducing the guided advantage estimator (GuAE), an anchor-based and variance-adaptive advantage tempering mechanism built upon GRPO. GuAE prevents advantage collapse in low-variance rollout groups under sparse GUI rewards, and with a thought-action consistency reward, Faithful-Agent (Stage II) elevates the Trap SR from 13.88\% to 80.21\% relative to the baseline, while preserving robust general instruction-following performance. --- ### SRA: Span Representation Alignment for Large Language Model Distillation Authors: Quoc Phong Dao, Hoang Son Nguyen, Pham Khanh Chi, Tung Nguyen, Linh Ngo Van, Nguyen Thi Ngoc Diep, Trung Le Published: 2026-05-02 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.01205v2 Cross-Tokenizer Knowledge Distillation (CTKD) enables knowledge transfer between a large language model and a smaller student, even when they employ different tokenizers. While existing approaches mainly focus on token-level alignment strategies, which are often brittle and sensitive to discrepancies between tokenizers, we argue that the method of aggregating tokens into more robust representations before distillation is of equal importance. In this paper, we introduce \textbf{SRA} (\textbf{S}pan \textbf{R}epresentation \textbf{A}lignment for Large Language Model Distillation), a novel framework that reframes CTKD through the physical lens of Multi-Particle Dynamical Systems. SRA shifts the fundamental unit of alignment from tokens to robust, tokenizer-agnostic spans. We model each span as a cluster of particles and represent its state by its Center of Mass (CoM) - an attention-weighted average that captures rich semantic information. We leverage the concept of span centers of mass with attention-derived weighting to prioritize the most salient spans. In addition, we employ a geometric regularizer to preserve the structural integrity of the representation space and introduce aligned span logit distillation to enhance knowledge transfer across models. In challenging cross-architecture distillation experiments, SRA consistently and significantly outperforms state-of-the-art CTKD baselines, validating our physically-grounded approach. --- ### GR-Ben: A General Reasoning Benchmark for Evaluating Process Reward Models Authors: Zhouhao Sun, Xuan Zhang, Xiao Ding, Bibo Cai, Li Du, Kai Xiong, Xinran Dai, Fei Zhang, weidi tang, Zhiyuan Kan, Yang Zhao, Bing Qin et al. Published: 2026-05-02 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2605.01203v2 Currently, process reward models (PRMs) have exhibited remarkable potential for test-time scaling. Since large language models (LLMs) regularly generate flawed intermediate reasoning steps when tackling a broad spectrum of reasoning and decision-making tasks, PRMs are required to possess capabilities for detecting process-level errors in real-world scenarios. However, existing benchmarks primarily focus on mathematical reasoning, thereby failing to comprehensively evaluate the error detection ability of PRMs across diverse reasoning scenarios. To mitigate this gap, we introduce GR-Ben, a process-level benchmark specifically designed for assessing PRM's performance across two primary reasoning domains (science and logic) and nine subdomains. We conduct extensive experiments on a diverse set of 22 models, encompassing both PRMs and LLMs, and derive two key findings: (1) In domains beyond mathematical reasoning, the error-detection ability of existing PRMs and LLMs is found to be markedly weaker by comparison.(2) In general, PRMs are less adept at identifying knowledge-based errors, whereas LLMs exhibit poorer performance in detecting computational errors. We hope GR-Ben can foster future researches on PRMs for general domains, thereby enhancing the reasoning capabilities of LLMs. --- ### Focus and Dilution: The Multi-stage Learning Process of Attention Authors: Zheng-An Chen, Pengxiao Lin, Zhi-Qin John Xu, Tao Luo Published: 2026-05-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.01199v1 Transformer-based models have achieved remarkable success across a wide range of domains, yet our understanding of their training dynamics remains limited. In this work, we identify a recurrent focus-dilution cycle in attention learning and provide a rigorous explanation in a one-layer Transformer setting for Markovian data via gradient-flow analysis. Using stage-wise linearization around critical points, we show that a single focus-dilution cycle can be decomposed into a sequence of distinct stages. First, embedding and projection rapidly condense to a rank-one structure, while attention parameters remain effectively frozen. Then, the attention parameters begin to increase, inducing a frequency-driven focus toward high-frequency tokens. As attention continues to evolve, it generates next-order perturbations in embeddings, leading to a mass-redistribution mechanism that progressively dilutes this focus. Finally, small asymmetries among low-frequency tokens lift a degenerate critical point, opening new embedding directions and initiating the next cycle. Experiments on synthetic Markovian data as well as WikiText and TinyStories corroborate the predicted stages and cyclical dynamics. --- ### Near-Floor Geometry Is Generic: Leverage Dispersion in Trained Overcomplete Codes Authors: Hector Borobia, Elies Seguí-Mas, Guillermina Tormo-Carbó Published: 2026-05-02 Categories: cs.LG, cs.IT arXiv: https://arxiv.org/abs/2605.01192v2 An overcomplete code packs $F$ features into $d<F$ dimensions, so linear readout of one feature picks up cross-talk from the others, bounded below by the rank-trace floor $W(F,d)=(F-d)/(d(F-1))$. Proximity to this floor is read as evidence of an efficient arrangement. It is not: an i.i.d. code of the same shape attains it at every shape and load we measure (1.0000-1.0005 over 7 matched controls), and the reason is exact. For the gain-calibrated pseudoinverse the attainment ratio is $\frac{d}{F(F-d)}(\sum_i h_i^{-1}-F)$ with $\sum_i h_i=d$, so it equals one precisely when the leverage $h_i$ is equalised across features: distance from the floor IS leverage dispersion. On 40 released decoder matrices - 24 sparse-autoencoder decoders plus 16 MLP down-projections of a hybrid model - all sit above their matched control (1.0055-1.2018), and all reach guaranteed affine failure 4 to 40 sparsity levels earlier: where a random code of the same shape is still safe, a released dictionary already has a feature no affine rule recovers. Autoencoders trained by the same recipe on a randomly initialised copy of the same model return to the floor (1.0038-1.0065) in all 10 paired layers, so the excess reflects what the model learned, not the act of fitting an autoencoder. Under controlled conditions the loss decides whether an interface exists at all: $L^2$ reaches $R_{geom}=18.5635$ to 32.3131 with its frontier collapsed to $1$, against $1.0002$ for $L^4$, and that frontier follows a rate we registered before the widest run (10.79 predicted, 10.510 measured). Two limits are stated rather than implied: frontier statements are worst cases, not typical-case claims; and because the probe class contains the network's own decoder, our decoder comparison bounds supervised fitting rather than affine expressiveness. --- ### Linear-Readout Floors and Threshold Recovery in Computation in Superposition Authors: Hector Borobia, Elies Seguí-Mas, Guillermina Tormo-Carbó Published: 2026-05-02 Categories: cs.LG, cs.IT arXiv: https://arxiv.org/abs/2605.01192v1 Two recent approaches to computation in superposition reach different recursive capacity regimes: Hänni et al. certify $\tilde{O}(d^{3/2})$ computable features in width $d$ via an approximate-linear recursive template, while Adler and Shavit reach near-quadratic capacity (up to logarithmic factors) using thresholded Boolean recovery. The main contribution of this paper is conceptual: we argue these results are not contradictory because they maintain different interface invariants, and we formalize the distinction. As a tool, we record a rank-trace Welch-type lower bound for biorthogonal linear readouts: for $F \gg d$, the worst-case off-diagonal cross-talk of any unit-diagonal linear readout is $Ω(d^{-1/2})$, and the bound is tight on average for unit-norm tight frames. At quadratic feature load $F=d^2$, random-support threshold recovery succeeds for sparsities $s=O(d/\log d)$, while linear readouts still incur $Ω(s/d)$ average per-coordinate squared error on Bernoulli sparse states. Matching the Welch floor against the published tolerance of the Hänni correction layer explains the $d^{3/2}$ scale as a compatibility threshold for that template, not a universal upper bound. Robust nonlinear reset beyond the Hänni template is left open. --- ### NEURON: A Neuro-symbolic System for Grounded Clinical Explainability Authors: Anuradha Chandrasekaran, Dimitrios Zikos, Mutlu Mete, Alan Pang, Brady D. Lund, Kewei Sha Published: 2026-05-02 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.01189v3 Clinical AI adoption is hindered by the black-box/grey-box nature of high-performing models, which lack the ontological grounding and narrative transparency required for professional-level explainability. We present NEURON, a neuro-symbolic system designed to enhance both predictive reliability and clinical interpretability. NEURON integrates SNOMED CT ontology-informed structural representations with machine learning models to bridge the gap between raw data and medical nomenclature. To facilitate human-aligned interaction, the system utilizes a Retrieval-Augmented Generation (RAG) grounded Large Language Model (LLM) layer to synthesize SHAP feature attributions and patient-specific clinical notes into coherent, natural-language explanations. Validated on the MIMIC-IV dataset for Acute Heart Failure mortality prediction, NEURON improved the AUC from 0.71-0.74 to 0.74-0.77 and substantially outperformed SHAP-based explanations in human-aligned metrics (0.807 vs. 0.558). Our results demonstrate that NEURON offers a robust, scalable engineering solution for deploying trustworthy, human-centered connected health applications. --- ### NEURON: A Neuro-symbolic System for Grounded Clinical Explainability Authors: Anuradha Chandrasekaran, Dimitrios Zikos, Mutlu Mete, Alan Pang, Brady D. Lund, Kewei Sha Published: 2026-05-02 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.01189v2 Clinical AI adoption is hindered by the black-box/grey-box nature of high-performing models, which lack the ontological grounding and narrative transparency required for professional-level explainability. We present NEURON, a neuro-symbolic system designed to enhance both predictive reliability and clinical interpretability. NEURON integrates SNOMED CT ontology-informed structural representations with machine learning models to bridge the gap between raw data and medical nomenclature. To facilitate human-aligned interaction, the system utilizes a Retrieval-Augmented Generation (RAG) grounded LLM layer to synthesize SHAP feature attributions and patient-specific clinical notes into coherent, natural-language explanations. Validated on the MIMIC-IV dataset for Acute Heart Failure mortality prediction, NEURON improved the AUC from 0.74-0.77 to 0.84-0.88 and significantly outperformed raw SHAP visualizations in human-aligned metrics (0.85 vs. 0.50). Our results demonstrate that NEURON offers a robust, scalable engineering solution for deploying trustworthy, human-centered connected health applications. --- ### Compute Optimal Tokenization Authors: Tomasz Limisiewicz, Artidoro Pagnoni, Srini Iyer, Mike Lewis, Sachin Mehta, Alisa Liu, Margaret Li, Gargi Ghosh, Luke Zettlemoyer Published: 2026-05-02 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.01188v2 Scaling laws enable the optimal selection of data amount and language model size, yet the impact of the data unit, the token, on this relationship remains underexplored. In this work, we systematically investigate how the information granularity of tokens, controlled by the compression rate (i.e., average bytes of text per token), affects scaling trends. We train 988 latent tokenized models (BLT) ranging from 50M to 7B parameters that enable setting the desired compression rate. This flexibility allows us to study the role of compression rate well beyond 4.57 bytes per token obtained with a popular BPE tokenizer. Our experiments reveal that in compute-optimal configurations, model parameter counts scale proportionally to data size measured in bytes, not in tokens as commonly perceived (Kaplan et al., 2020; Hoffmann et al., 2022). Furthermore, we discover that the optimal compression rate differs from the one obtained with BPE and decreases with compute. These findings generalize to both latent and subword tokenization, as well as to languages other than English, guiding language model developers on tokenization scheme selection for maximal compute efficiency. --- ### Phase-map synthesis from magnitude-only MR images using conditional score-based diffusion models with application in training of accelerated MRI reconstruction models Authors: M. Berk Sahin, Dilek Yalcinkaya, Abolfazl Hashemi, Behzad Sharif Published: 2026-05-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.01185v1 Accelerated magnetic resonance imaging (MRI) enabled by the training of deep learning (DL)-based image recon. models requires large and diverse raw k-space datasets. In most clinical MRI applications, due to storage and patient privacy concerns, raw k-space data is discarded and magnitude-only images are the only component saved. Consequently, a large portion of the DL-based MRI recon. literature has either relied on small training datasets or has used one of the few available open-source k-space datasets. At the same time, the growing number of anonymized magnitude-only image registries/databases motivates the development of techniques that can use them as training datasets for generalizable DL-based recon. models. Here we propose to address this challenge by employing a generative approach based on conditional score-based diffusion models (SBDMs): given a magnitude-only MR image, it synthesizes a phase map (in the image domain) that realistically corresponds to the magnitude-only image. We evaluate its generative capabilities in a downstream DL-based recon. task whereby a large k-space dataset is generated by combining the SBDM-synthesized phase-maps and the corresponding magnitude-only images, and this k-space dataset is then used to train a DL model for accelerated MRI recon. We compare the performance of the resulting DL model versus those trained according to (a) a naive approach that uses smooth phase, (b) a k-space training dataset generated using synthesized phase maps derived from a generative adversarial network, and (c) the ground truth k-space data. Our results suggest that the DL model trained from SBDM-synthesized k-space data outperforms the other approaches in terms of quantitative metrics as well as qualitatively observed recon. fidelity, i.e., whether the reconstructed images include erroneous or hallucinated features that could adversely impact diagnostic accuracy. --- ### A Theory of Generalization in Deep Learning Authors: Elon Litman, Gabe Guo Published: 2026-05-02 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2605.01172v1 We present a non-asymptotic theory of generalization in deep learning where the empirical neural tangent kernel partitions the output space. In directions corresponding to signal, error dissipates rapidly; in the vast orthogonal dimensions corresponding to noise, the kernel's near-zero eigenvalues trap residual error in a test-invisible reservoir. Within the signal channel, minibatch SGD ensures that coherent population signal accumulates via fast linear drift, while idiosyncratic memorization is suppressed into a slow, diffusive random walk. We prove generalization survives even when the kernel evolves $\mathcal{O}(1)$ in operator norm, the full feature-learning regime. This theory naturally explains disparate phenomena in deep learning theory, such as benign overfitting, double descent, implicit bias, and grokking. Lastly, we derive an exact population-risk objective from a single training run with no validation data, for any architecture, loss, or optimizer, and prove that it measures precisely the noise in the signal channel. This objective reduces in practice to an SNR preconditioner on top of Adam, adding one state vector at no extra cost; it accelerates grokking by $5 \times$, suppresses memorization in PINNs and implicit neural representations, and improves DPO fine-tuning under noisy preferences while staying $3 \times$ closer to the reference policy. --- ### CADFit: Precise Mesh-to-CAD Program Generation with Hybrid Optimization Authors: Ghadi Nehme, Eamon Whalen, Faez Ahmed Published: 2026-05-02 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2605.01171v3 Despite recent progress, recovering parametric CAD construction sequences from geometric input, such as meshes or point clouds, is a key challenge for design and manufacturing, as existing CAD reconstruction and generation methods are largely restricted to difficult-to-edit formats like meshes or Breps or editable simple sketch-and-extrude pipelines and low-complexity datasets. We introduce CADFit, a hybrid optimization-based CAD reconstruction framework that recovers complex, editable CAD construction sequences from meshes by incrementally fitting and validating parametric operations using geometric feedback. Our approach is distinguished by formulating reconstruction as an IoU-driven optimization over structured CAD programs and supporting a rich set of operations, including extrusions, revolutions, fillets, and chamfers. Experiments on multiple CAD benchmarks show that CADFit outperforms state-of-the-art mesh-to-CAD methods in volumetric Intersection-over-Union and Chamfer Distance, while substantially reducing the Invalid Ratio of reconstructed CAD programs, particularly for complex designs. We further present a multimodal pipeline that enables end-to-end reconstruction of CAD construction sequences from images by combining image-based geometry reconstruction with CADFit. By enabling accurate reconstruction of higher-complexity CAD models, CADFit provides a practical foundation for generating richer datasets and advancing future learning-based approaches to CAD reverse engineering. The code is available at: https://github.com/ghadinehme/CADFit. --- ### Control panels to clarify user intent with Large Language Models Authors: Ben Shneiderman Published: 2026-05-01 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2607.21598v1 Typical user interfaces for Large Language Models present a blank prompt window that invites a natural language query by users, but offers little guidance. This paper proposes a visual control panel interface that would provide more cues to the semantics of prompt formation, enabling users to more easily express their intent. By emphasizing recognition over recall, control panels help users formulate more effective prompts that match their intent. --- ### Generative Ontology Induction: Domain-Agnostic Schema Discovery from Document Corpora Using Large Language Models Authors: Sergei Sergienko Published: 2026-05-01 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.16201v1 Ontology engineering remains a critical bottleneck in knowledge-intensive AI systems. Existing automated approaches either depend on predefined schemas, operate within narrow domains, or produce unstructured outputs unsuitable for downstream pipelines. We introduce Generative Ontology Induction (GOI), a domain-agnostic framework that induces a generative blueprint - entities, dimensions, properties, relationships, and constraints - from a corpus of examples and exports it as a typed graph (six node types, seven edge types) in YAML/JSON. We introduce the Node Coverage Score, a novel evaluation metric that measures the fraction of structural ontology nodes (classes, properties, and dimensions) appearing in generated outputs. A controlled generative validation on four contrasting ontologies - a familiar Software Services Invoice schema, a custom Job Description Ontology, a confidential Pain-Management Clinical Visit Record Ontology, and a Professional Services Contract & Statement of Work Ontology - shows that GOI-prompted generation covers 95-100% of the structural backbone in every case; a generic three-field template holds at 97.8% on the invoice schema but drops to 52.2% on the Job Description Ontology, 62.2% on the Pain-Management ontology, and 78.3% on the Professional Services Contract ontology. The structural coverage holds regardless of how familiar the document type is to the model. --- ### Just Keep Prompting: Evaluating Repetitive Socratic Prompting in VLMs Authors: Shayda Moezzi, Bishoy Galoaa, Lorena Genua, Taskin Padir, Sarah Ostadabbas Published: 2026-05-01 Categories: cs.CL, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2607.14099v1 Deploying Vision-Language Models (VLMs) in real-world settings requires not only strong visual reasoning but also stability under sustained conversational pressure. We introduce Just Keep Prompting (JKP), a multi-turn evaluation framework that measures VLM epistemic stability when users repeatedly challenge, question, or contradict a model's answer. JKP probes models for up to 10 follow-up turns using three strategies: Adversarial Negation (repeated rejection), Pure Socratic Interrogation (repeated calls to reassess certainty), and Context-Aware Socratic Summarization (reflecting the model's prior rationale back before asking for reconsideration). We evaluate GPT-4o, Gemini 2.5 Pro, and Qwen3-VL-30B on a subset of the STAR benchmark across 720 multi-turn runs. Aggregate accuracy changes modestly from Turn 0 to Turn 10, but trajectory-level analysis reveals substantial instability: correct answers regress, wrong answers recover, and many runs exhibit repeated answer flipping. Repeated prompting has bounded upside and often acts as a destabilizer rather than a reasoning aid. The effect is strongly model-dependent: Qwen3-VL-30B achieves the highest final accuracy but becomes confidently wrong under direct contradiction; Gemini 2.5 Pro is comparatively stable but token-expensive; GPT-4o is the most brittle and oscillatory. These findings reveal that multi-turn VLM evaluation captures not just additional reasoning but pressure-response profiles: how models trade off visual grounding, calibration, and conversational compliance under repeated challenge. --- ### From ML Predictions to Informed Diagnostic Assistance Using the Toulmin Model of Argumentation Authors: Anca Marginean, Adrian Groza Published: 2026-05-01 Categories: cs.AI, cs.CV arXiv: https://arxiv.org/abs/2607.09664v1 To provide a structured and interpretable assessment, we decompose the image-based diagnosis into components following the Toulmin model of argumentation. This model consists of a claim, grounds, warrant, qualifier, rebuttal, and backing. Consider a claim generated by a machine learning (ML) model for retinal diagnosis. Rather than accepting this claim at face value, one could either apply explainable AI (XAI) methods or adopt an argumentation-based approach. In our framework, a model specialized in biomarker extraction from images provides the grounds. The warrant-linking the grounds to the claim - is analyzed by an agent equipped with medical knowledge; in our architecture, this role is fulfilled by a MedGemma agent. The qualifier is determined based on the overall quantitative evaluation of both the warrant and grounds models. Finally, a rebuttal is constructed using image similarity measures computed with MedSigLip. All these components are presented to the human expert, enabling a more informed and critical assessment of the ML-generated diagnosis. --- ### Breaking Safety at the Token Boundary: How BPE Tokenization Creates Exploitable Gaps in LLM Alignment Authors: Tung-Ling Li, Hongliang Liu, Yuhao Wu Published: 2026-05-01 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2607.01239v1 Character-level perturbations bypass safety alignment in modern LLMs despite leaving prompts human-readable. We identify and test a central structural mechanism: BPE tokenization fragments safety-critical words into sub-word pieces, and the three public alignment datasets we surveyed contain no intentionally fragmented inputs. The mechanism is a chain, tested end-to-end on five model families (Qwen-3-4B, Qwen-2.5-7B, Gemma-3-4B, Llama-3.1-8B, Mistral-7B). An optimization targeting safety-token fragmentation flips the first-token refusal trigger on 80-100% of refused HarmBench prompts, with 48% of those flips producing genuinely harmful outputs (per-model 29-65%; gap-vs-behavior ROC-AUC 0.66-0.98, pooled 0.84). Activation patching localizes the disrupted signal to the last ${\sim}30\%$ of layers; an alignment-data scan finds zero fragmented prompts among 30,000 examples (positive-control recall $\geq 99\%$ at attack-relevant intensities); and targeted-mutation experiments isolate safety words as the disruption locus. On the defense side, a 68-cell grid (55 trained checkpoints) shows that no DPO configuration achieves seed- and pool-stable ASR closure on the three families with closed pool-size confounds. SFT trained on fragmented prompts closes ASR on 3/5 families but only via global collapse that raises refusal on benign prompts as well, indicating the missing distribution is necessary but not sufficient under the LoRA-16 recipe we tested. To distinguish selective repair from global collapse, we introduce Conv-Benign, a candidate paired diagnostic. All ASR claims are 3-judge-calibrated (cell rankings stable across judges; absolute levels $\pm$18pp; see App.~B.13). --- ### SPARCLE: SPeaker-aware Aligned Representations via Contrastive Language Embeddings Authors: Priyam Mazumdar, Yurii Halychanskyi, Steven Guo, Mark Hasegawa-Johnson, Volodymyr Kindratenko Published: 2026-05-01 Categories: cs.CL, cs.AI, cs.SD, eess.AS arXiv: https://arxiv.org/abs/2607.01238v1 Recent advances in speech synthesis have shifted from phoneme representations to direct grapheme modeling. While phonemes address the one-to-many mapping between text and acoustics, they rely on grapheme-to-phoneme (G2P) systems that fail to capture speaker-specific acoustic variation. Prior work demonstrates that grapheme-based models outperform phoneme-based systems at scale, but not in low-resource settings. In this paper, we propose SPARCLE, a speaker-aware grapheme representation model that enriches characters with their precise acoustic realizations. SPARCLE is trained with a contrastive objective to align graphemes with corresponding Wav2Vec2 acoustic representations while conditioned on speaker identity. The resulting model serves as a replacement to G2P systems for downstream text-to-speech (TTS) tasks. We demonstrate that SPARCLE improves generation quality, reducing word error rates by half in extreme low-resource settings compared to standard grapheme-based models. --- ### KARA: Efficient Reasoning LLM Serving via Sliding-Window KV Cache Compression Authors: Shen Han, Yuyang Wu, Junpu Yu, Olexandr Isayev Published: 2026-05-01 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2607.01237v2 Reasoning language models often generate long chain-of-thought (CoT), which accumulates a massive KV cache during the decoding phase and incurs high decoding latency and limited throughput. To address these issues, KV cache compression has emerged as a promising technique for reducing memory overhead by selectively removing unimportant KV pairs while preserving useful ones for subsequent decoding. Nevertheless, we identify two key limitations in existing KV cache compression methods: 1) their threshold-triggered compression policy may provide limited throughput improvement or even reduce throughput, and may fully eliminate KV pairs from certain blocks of the sequence, potentially worsening information loss. 2) they typically retain either isolated KV pairs or fixed-size chunks with rigid boundaries, failing to preserve important flexible-sized chunks at arbitrary token positions. To overcome these limitations, we propose Kara, a sliding-window KV cache compression method that performs decoding-time compression by operating only on the recently generated context. Kara leverages bidirectional attention to score and select informative KV pairs in the window. To enable flexible preservation of important semantic information, we design a Token2Chunk module to expand a subset of selected KV pairs into chunks. Furthermore, we adapt Kara to PagedAttention and develop KvLLM, an inference framework built upon vLLM, which reduces KV cache memory usage and effectively improves output throughput. Extensive experiments demonstrate consistent performance improvements of proposed Kara and KvLLM. --- ### Safeguarding LLM Agents from Misalignment through Provenance Analysis Authors: Yining She, Yiliang Liang, Eunsuk Kang Published: 2026-05-01 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2607.01236v2 As LLM agents gain increasing access to powerful tools, ensuring that their actions align with the user's intent becomes critical. When an agent's proposed action deviates from that intent---a phenomenon called misalignment---it may cause harm that is difficult to undo. Existing runtime guardrails rely on an LLM-as-a-judge paradigm that lacks a systematic framework for reasoning about alignment, often producing inconsistent or difficult-to-audit judgments. Motivated by provenance analysis, we propose a conceptual framework that formalizes misalignment detection as determining whether a proposed tool call is supported by traceable evidence in the agent's context. Based on this framework, we build ProvenanceGuard, a multi-stage pipeline that analyzes the agent's action for three types of misalignment before its execution and only allows aligned actions. We evaluated ProvenanceGuard on AgentSafetyBench and WorkBench, across 11 backbone LLMs. Compared to the LLM-as-a-judge baseline, ProvenanceGuard reduces error rate on misaligned traces from 44.3% to 2.1% on Agent-SafetyBench and from 32.4% to 18.7% on WorkBench, while reducing interventions on task-successful traces from 31.2% to 13.0% and introducing no statistically significant increase in unnecessary interventions on aligned traces. These results demonstrate that structured, provenance-based reasoning provides an effective and practical foundation for safeguarding LLM agents from misalignment. --- ### Safeguarding LLM Agents from Misalignment through Provenance Analysis Authors: Yining She, Yiliang Liang, Eunsuk Kang Published: 2026-05-01 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2607.01236v1 As LLM agents gain increasing access to powerful tools, ensuring that their actions are aligned with the user's intent becomes critical. When an agent's proposed tool invocation deviates from the user's intent -- a phenomenon called misalignment -- it may lead to harmful consequences that are difficult to undo. Existing runtime guardrails rely on an LLM-as-a-judge paradigm that lacks a systematic framework for reasoning about alignment, often producing judgments that are inconsistent or difficult to audit. Motivated by provenance analysis, we propose a provenance-based conceptual framework that formalizes misalignment detection as determining whether a proposed tool call is supported by traceable evidence in the agent's context. Building on this framework, we propose ProvenanceGuard, a multi-stage pipeline that analyzes the agent's action for three types of misalignment before the selected tool is executed and only allows the action to take place when it is considered aligned with the user's input query. We evaluated our proposed approach on two different benchmarks, Agent-SafetyBench and WorkBench, across 10 backbone LLMs. Compared to the LLM-as-a-judge baseline, ProvenanceGuard reduces error rate on misaligned traces from 42.9% to 1.8% on Agent-SafetyBench and from 32.1% to 17.3% on WorkBench, while reducing intervention burden on task-successful traces from 30.5% to 12.8% and introducing no statistically significant increase in unnecessary interventions on aligned traces. These results demonstrate that structured, provenance-based reasoning provides an effective and practical foundation for safeguarding LLM agents from misalignment. --- ### Persona Without Substrate: Regime-Dependence and the LLM Individuation Problem Authors: Shuaizhi Cheng Published: 2026-05-01 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2607.00006v1 Beckmann & Butlin's (2026) ontological framework for the LLM individuation problem inherits an unargued cross-regime co-reference assumption from the persona-vectors literature: that the same direction picks out the same content under prompt-conditioning, gradient-descent fine-tuning, and inference-time steering. We present four empirical wedges from persona-topology experiments on Qwen3-4B-Instruct and Mistral-7B-Instruct-v0.2 - non-collinearity of prompt-extracted vectors and fine-tune basins; fictional personas displacing the model along real-anchor directions more strongly than real anchors do; contradictory-valenced mixtures biased toward a training-history-determined attractor; and asymmetric compositional algebra under inference-time arithmetic versus fine-tune-time chimera training - that jointly undermine the assumption. We propose regime-indexed individuation: the identity unit for representational content is a (vehicle, regime) pair, not a vehicle alone. Under this framework, Beckmann & Butlin's three candidate positions describe three different regime-internal objects rather than competing for the same referent; the same diagnosis applies to Mollo & Millière, Chalmers, and Cerullo. --- ### Reducing Conversational Escalation in Large Language Model Dialogue with Nonviolent Communication Constraints Authors: Zhixing Sun, Shenghe Xu, Tao Li Published: 2026-05-01 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2606.26106v1 Large language models (LLMs) are increasingly used in emotionally charged situations involving interpersonal conflict, frustration, and distress. While prior safety research has focused on preventing explicit harms such as toxic or policy-violating content, less attention has been paid to conversational behaviors that may unintentionally escalate conflict. In this paper, we investigate whether LLMs can be guided toward more de-escalating dialogue behavior through lightweight prompt-level constraints derived from Nonviolent Communication (NVC). We reformulate NVC principles as process-oriented guidelines that discourage blame attribution, emphasize attention to users' emotional experiences, and encourage clarification before advice. Using a dual-agent simulation framework across multiple instruction-tuned models and user resistance levels, we show that NVC-constrained prompting consistently reduces conversational escalation and stabilizes interactions with highly resistant users. These results suggest that simple communication constraints can meaningfully improve the trustworthiness of LLM dialogue in conflict-prone settings. --- ### Context Recycling for Long-Horizon LLM Inference Authors: Derek Thomas Published: 2026-05-01 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2606.26105v1 Large language models (LLMs) exhibit strong capabilities in short-context reasoning but degrade in performance over long conversational horizons due to context window limitations and inefficient token usage. We introduce ContextForge, a system for context recycling that maintains task-relevant information across turns by combining structured query generation, external memory retrieval, and controlled synthesis. The system enables efficient reuse of prior computation without relying on full context replay, reducing token overhead while preserving answer quality. We evaluate ContextForge using a 15-turn conversational benchmark that tests multi-turn reasoning, back-references, and domain shifts across structured healthcare queries. Compared to a baseline agent using identical underlying models, ContextForge demonstrates improved consistency and reduced token consumption, while maintaining comparable response accuracy. These results suggest that context recycling provides a practical approach for extending LLM capabilities in long-horizon tasks without requiring larger context windows or model retraining. Code and evaluation artifacts are available at https://github.com/Betanu701/ContextForge. --- ### LLM-assisted gNB Parameter Configuration for Radio Access Network Authors: Yao-Cong Dong, Maria Amparo Canaveras Galdon, Ari Uskudar, Kuntal Chowdhury, Edwin K. P. Chong, Ray-Guang Cheng Published: 2026-05-01 Categories: cs.NI, cs.AI arXiv: https://arxiv.org/abs/2606.20574v1 gNB parameter misconfigurations are a common cause of system failures in radio access networks (RANs), and their diagnosis and correction rely on manual analysis of complex network logs that does not scale well. This paper proposes a large language model (LLM)-assisted framework for automatic gNB parameter configuration. The framework adopts a synthetic data generation pipeline following a configuration, log, correction workflow. Starting from a workable configuration and the gNB technical references, the pipeline uses a commercial LLM to generate modified configurations and derive structured reasoning traces from gNB error logs. The synthetic training data maps network states to corrective actions and is used to fine-tune an LLM for configuration correction. During inference, the fine-tuned LLM generates valid and deployable gNB parameter configurations from gNB error logs. The framework is validated on an OpenAirInterface (OAI) gNB testbed with 480 unseen misconfiguration scenarios, where fine-tuning improves correction accuracy from 13.8% (zero-shot baseline) to 85.4%, and retrieval-augmented generation (RAG) further improves accuracy to 92.7%. The results demonstrate that the framework may enable automated recovery from misconfigurations without manual intervention and supports scalable and autonomous RAN operation. --- ### Beyond Compaction: Structured Context Eviction for Long-Horizon Agents Authors: Andrew Semenov, Svyatoslav Dorofeev Published: 2026-05-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2606.11213v1 We present Context Window Lifecycle (CWL), a context-management scheme that gives long-horizon LLM agents an effectively unbounded working horizon. As a session accumulates history, CWL keeps the context within budget through graduated, semantically-aware eviction: the agent annotates its trajectory as typed, dependency-linked episodes as work proceeds, and a deterministic, LLM-free policy evicts content in priority order within that structure when a token budget is exceeded. CWL preserves user turns and the exploratory context the agent is actively reasoning over, while aggressively shedding action episodes whose effects are already persisted in the environment, keeping active context near a stable ceiling that also avoids the performance degradation associated with very large prompts. Compared to summarization-based compaction, CWL avoids four well-known limitations: unpredictable lossiness, destruction of causal structure, blocking model cost, and compression-induced hallucination. Compared to recency truncation, CWL is semantically aware: it drops the oldest-and-most-recoverable content according to the dependency graph rather than oldest-in-time regardless of relevance. We describe the annotation protocol, the episode graph, the eviction policy, and the token-accounting loop, and evaluate CWL on long-horizon agentic benchmarks: a single agent session completing 89 sequential tasks across 80 million tokens with no measurable degradation in task accuracy relative to per-task isolated sessions --- ### Sketch-to-Layout: A Human-Centric Computational Agent for Constraint-Aware Synthesis of Modular Photobioreactors Authors: Xiujin Liu, Shuqi Li, Yuxin Lin Published: 2026-05-01 Categories: cs.HC, cs.CV arXiv: https://arxiv.org/abs/2606.09849v1 Building-integrated photobioreactors (PBRs) offer a pathway for carbon-neutral architecture, yet deployment is hindered by configuration complexity and biological maintenance. This paper presents a modular PBR facade system powered by a computational framework reconciling design intent with physical validity. We introduce 'carbon-neutralization bricks' featuring integrated vessel-and-conduit geometry; monolithic fluid channels enable 'plug-and-play' assembly. To navigate the combinatorial complexity of 14 modular geometries, we develop a Computational Sketch-to-Layout Agent that formulates layout synthesis as a Constraint Satisfaction Problem (CSP). Using the CP-SAT engine, the agent treats sparse user sketches as soft priors while enforcing hard constraints like port alignment and global connectivity. This allows non-experts to synthesize fabrication-ready configurations in near real-time. Furthermore, to facilitate autonomous maintenance, we propose a weakly supervised algae health monitoring pipeline. By employing a hybrid CNN-attention backbone and a temporal ranking loss, the system quantifies biological vitality from photographs without absolute ground-truth labels. Experiments demonstrate the CSP solver achieves a 95.5% success rate on grid scales up to 15 x 15. Qualitative evaluations confirm the framework preserves design semantics while ensuring operational integrity. Long-term tests show the vision module produces health trajectories aligned with 14-day biological cycles, suggesting that integrating interactive synthesis with low-cost computer vision can democratize scalable carbon capture systems. --- ### Human-AI Coordination Zones: A Framework for Designing Human-in-the-Loop Experiences with Agentic AI Authors: James Pierce, Vaiva Kalnikaitė, Siddharth Gupta, Brian Granger Published: 2026-05-01 Categories: cs.HC, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2606.09848v1 As generative and agentic AI becomes embedded in everyday products, practitioners face a persistent challenge: how to design human-AI coordination -- the ongoing mutual adjustment between users and AI systems as mediate through interfaces-that supports usability, trust, and safety. Existing resources offer high-level principles ("be transparent," "maintain user control") or low-level UI patterns, but there is a lack of mid-level design knowledge bridging the two. Through landscape and artifact analysis of 60 commercial AI applications, we introduce a framework defining human-AI coordination as the interplay of three dimensions: salience (how prominently AI is presented), involvement (what users can do to engage AI), and activity (what AI actually does). We contribute mid-level tools including coordination zones (done-for-me, done-under-me, done-with-me, done-without-me), an input taxonomy (prompted, sparked, inferred, layered), coordination curves for mapping user journeys, and design patterns demonstrating the generative capacity of the framework. The framework can be applied generatively to design experiences, analytically to evaluate existing ones, and communicatively to articulate ideas across stakeholders. --- ### Concerns and Strategic Responses of Older Workers Navigating Generative AI in Bridge Employment Authors: Aditya Nayak, Aakash Gautam, Rama Adithya Varanasi Published: 2026-05-01 Categories: cs.CY, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2606.07543v1 Generative AI (GenAI) is transforming workplaces at a rapid pace. This disproportionately affects vulnerable communities, including older workers (OWs) who re-enter the workforce through bridge employment prior to final retirement. Through in-depth semi-structured interviews with 21 professionals, we examine how OWs navigate GenAI-driven disruptions while pursuing bridge roles, focusing on their concerns about GenAI integration and their responses to these changes. Our findings show that OWs experienced both temporal and structural disruptions across all stages of the bridge employment decision-making process due to GenAI. In response, they reconfigured their tasks through different forms of boundary work aimed at restoring stability and continuity. We conceptualize these responses as AI resilience, which reshaped OWs' bridge employment decision-making into an ongoing process of negotiation and adaptation. We conclude by offering recommendations to reduce burnout among OWs by balancing individual-level AI resilience strategies with meso-level AI resilience collectives and macro-level adversarial and contestable AI-mediated organizational structures. --- ### DIYHealth Suite: Dataset, Model, and Benchmark for Health Management at Home Authors: Changshuo Liu, Junran Wu, Zhongle Xie, Wenqiao Zhang, Kaiping Zheng, Jiaqi Zhu, Qingpeng Cai, Ooi Gene Anne, Marcus Chun Jin Tan, Jianwei Yin, James Wei Luen Yip, Beng Chin Ooi Published: 2026-05-01 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2606.07542v1 Generative AI is reshaping healthcare, yet most existing advances rely on hospital-grade devices, which limits their accessibility and potential for health management outside clinical settings. With the proliferation of portable devices and telemedicine, healthcare is shifting toward home-based Diagnosis-It-Yourself (DIY) care. Despite this promise, several distinctive challenges remain: (i) home-collected data are heterogeneous, exacerbated by the absence of standardized large-scale datasets; (ii) models require adaptation to variable task demands and evolving individual conditions; (iii) the broad spectrum of home care tasks lacks a unified benchmark for systematic evaluation. In this paper, we present DIYHealth Suite, a comprehensive framework designed to address these challenges through a tailored dataset, model, and benchmark. We first curate DIYHealth-900K, a large-scale multimodal dataset capturing diverse real-world home care scenarios. Building on this, we propose DIYHealthGPT, an adaptive foundation model for home-based health management, powered by the novel Hybrid Hyper Low-Rank Adaptation technique. Finally, we establish DIYHealthBench, the first benchmark to evaluate foundation models on home care tasks. Extensive experiments demonstrate that DIYHealthGPT delivers state-of-the-art performance over both general-purpose and medical-specific baselines on 11 home care tasks in both open-QA and closed-QA settings, laying the groundwork for the next generation of personalized health management at home. --- ### Multimodal Large Language Models as Synthetic Participants in Video-Based Studies: An Evaluation Authors: Prabal Shrestha, Bohan Jiang, Haoning Xue, Huan Liu, Xinyi Zhou Published: 2026-05-01 Categories: cs.HC, cs.AI, cs.CV, cs.CY, cs.MM arXiv: https://arxiv.org/abs/2606.07541v1 Multimodal large language models (MLLMs) have shown strong performance on objective tasks such as video understanding and reasoning. However, it remains unclear whether they can approximate subjective human responses, which depend not only on content comprehension but also on individuals' social contexts. To address this gap, we evaluate MLLMs as synthetic participants in an emerging task: assessing perceived sensory engagement with short videos. Grounded in the Perceived Message Sensation Value (PMSV) framework, we compare ratings from recruited human participants and profile-conditioned MLLM simulations (n=673) using a 17-item scale measuring emotional arousal, dramatic impact, and novelty. We find that even leading MLLMs (Gemini 3 Flash and Qwen 3 Omni) show limited agreement with human participants. The models exhibit distinct downward mean-shift and central-tendency biases in their rating distributions. They both introduce and flatten subgroup differences, while showing inconsistent sensitivity to participant profiles. Prompting strategies affect these metrics differently, modestly improving some aspects while worsening others. These results highlight both the challenges and opportunities of developing MLLMs as synthetic participants in video-based research. Data and code: https://github.com/MINDLab25/mllm-human-simulation-eval --- ### Comprehensive AI governance requires addressing non-model gains Authors: Arthur Goemans, Dan Altman, Noemi Dreksler, Jonas Freund, Milan Gandhi, Zhengdong Wang, Sarah Cogan, Sebastien Krier, Demetra Brady, Lewis Ho, Allan Dafoe Published: 2026-05-01 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2606.00047v1 Frontier AI governance often centres on the model-level governance paradigm, which assumes that a model's capability profile is primarily a function of the compute and data used during training. This position paper argues that model-level governance becomes less effective when capability progress is increasingly driven by "non-model gains"--improvements that are independent from advances in the base model. We formalise the concept of non-model gains and provide a taxonomy of three distinct vectors of capability gain: inference gain (scaling compute at test-time), systems gain (post-training enhancements such as scaffolds), and asset gain (enhancing a model with restricted assets). We demonstrate how these vectors--alongside potential future impacts from embodiment, continual learning, and AI diffusion--may undermine risk management strategies that hinge mostly on pre-deployment evaluation and mitigation. We provide an overview of governance approaches that go beyond the model level: system, entity, agent, and cloud governance. Finally, we emphasise the importance of societal resilience as a complement to these governance layers. --- ### Spacetime Formation under Requirements: Contextual Realization and Form-Dependent Probability Authors: Song-Ju Kim Published: 2026-05-01 Categories: cs.AI, physics.hist-ph, quant-ph arXiv: https://arxiv.org/abs/2605.23943v1 Quantum cognition often explains order effects, contextuality, and violations of the law of total probability by replacing classical probability with quantum probability on a fixed event structure. This paper proposes a different interpretation: quantum probability is the fixed-spacetime projection of contextual spacetime formation under finite-state requirements. The framework begins not with time, space, objects, or probabilities, but with requirements such as finite representational capacity, single-state semantic stability, context-sensitive intervention, avoidance of explicit context labels, coherent world-formation, and intersubjective transformability. When these requirements cannot be realized within a single global Boolean event structure, the mismatch appears, under fixed-spacetime projection, as noncommutativity, interference, and quantum-like probability. Building on prior single-state approaches to contextuality, we reinterpret classical contextual bookkeeping cost as the fixed-spacetime shadow of contextual spacetime formation. Auxiliary memory or context labels in a classical representation correspond, in this account, to holonomy-like mismatch among locally Boolean logic-worlds. The interference term is the cross term generated when locally classical realization contributions are nontrivially glued and projected back into a fixed classical spacetime form. The result is a transcendental-operational realist account: objecthood, eventhood, probability, and spacetime are treated as forms of realization under requirements, while objectivity is defined by invariants preserved across observer- and history-dependent spacetime formations. --- ### TeamTR: Trust-Region Fine-Tuning for Multi-Agent LLM Coordination Authors: Yi Xie, Siao Liu, Falong Fan, Yuanqi Yao, Yue Zhao, Bo Liu Published: 2026-05-01 Categories: cs.LG, cs.MA arXiv: https://arxiv.org/abs/2605.15207v2 Multi-agent LLM systems have shown promise for complex reasoning, yet recent evaluations reveal they often underperform single-model baselines. We identify a structural failure mode in sequential fine-tuning of shared-context teams: updating one agent shifts the team's context distribution, and when subsequent updates are evaluated on cached rollouts, this mismatch compounds. We formalize this as the compounding occupancy shift and prove that stale-occupancy evaluation incurs a penalty that scales quadratically with the number of agents. In contrast, intermediate-occupancy evaluation reduces this to linear scaling. We propose TeamTR, a trust-region framework that resamples trajectories after each component update and enforces per-agent divergence control, yielding rigorous per-update and per-stage improvement lower bounds. Experiments show that TeamTR outperforms single-agent and sequential baselines with 7.1% on average, mitigates coordination regressions, and supports plug-and-play component replacement. Code is available at https://github.com/Yydc/TeamTR. ---