# Research Papers — 2026-04 · bundle page 57 of 233 > Full-text markdown of papers #2801–2850 of 11612 published in 2026-04, newest first. AI-readable mirrors of the full 11612-paper 2026-04 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-04/page/57 [← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/56.md) [Next bundle →](https://aipod.dev/papers/archive/2026-04/page/58.md) ### Agentic AI for Personalized Physiotherapy: A Multi-Agent Framework for Generative Video Training and Real-Time Pose Correction Authors: Abhishek Dharmaratnakar, Srivaths Ranganathan, Anushree Sinha, Debanshu Das Published: 2026-04-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.21154v1 At-home physiotherapy compliance remains critically low due to a lack of personalized supervision and dynamic feedback. Existing digital health solutions rely on static, pre-recorded video libraries or generic 3D avatars that fail to account for a patient's specific injury limitations or home environment. In this paper, we propose a novel Multi-Agent System (MAS) architecture that leverages Generative AI and computer vision to close the tele-rehabilitation loop. Our framework consists of four specialized micro-agents: a Clinical Extraction Agent that parses unstructured medical notes into kinematic constraints; a Video Synthesis Agent that utilizes foundational video generation models to create personalized, patient-specific exercise videos; a Vision Processing Agent for real-time pose estimation; and a Diagnostic Feedback Agent that issues corrective instructions. We present the system architecture, detail the prototype pipeline using Large Language Models and MediaPipe, and outline our clinical evaluation plan. This work demonstrates the feasibility of combining generative media with agentic autonomous decision-making to scale personalized patient care safely and effectively. --- ### Dialect vs Demographics: Quantifying LLM Bias from Implicit Linguistic Signals vs. Explicit User Profiles Authors: Irti Haq, Belén Saldías Published: 2026-04-22 Categories: cs.CY, cs.AI, cs.CL, cs.HC, cs.IR arXiv: https://arxiv.org/abs/2604.21152v1 As state-of-the-art Large Language Models (LLMs) have become ubiquitous, ensuring equitable performance across diverse demographics is critical. However, it remains unclear whether these disparities arise from the explicitly stated identity itself or from the way identity is signaled. In real-world interactions, users' identity is often conveyed implicitly through a complex combination of various socio-linguistic factors. This study disentangles these signals by employing a factorial design with over 24,000 responses from two open-weight LLMs (Gemma-3-12B and Qwen-3-VL-8B), comparing prompts with explicitly announced user profiles against implicit dialect signals (e.g., AAVE, Singlish) across various sensitive domains. Our results uncover a unique paradox in LLM safety where users achieve ``better'' performance by sounding like a demographic than by stating they belong to it. Explicit identity prompts activate aggressive safety filters, increasing refusal rates and reducing semantic similarity compared to our reference text for Black users. In contrast, implicit dialect cues trigger a powerful ``dialect jailbreak,'' reducing refusal probability to near zero while simultaneously achieving a greater level of semantic similarity to the reference texts compared to Standard American English prompts. However, this ``dialect jailbreak'' introduces a critical safety trade-off regarding content sanitization. We find that current safety alignment techniques are brittle and over-indexed on explicit keywords, creating a bifurcated user experience where ``standard'' users receive cautious, sanitized information while dialect speakers navigate a less sanitized, more raw, and potentially a more hostile information landscape and highlights a fundamental tension in alignment--between equitable and linguistic diversity--and underscores the need for safety mechanisms that generalize beyond explicit cues. --- ### "This Wasn't Made for Me": Recentering User Experience and Emotional Impact in the Evaluation of ASR Bias Authors: Siyu Liang, Alicia Beckford Wassink Published: 2026-04-22 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.21148v2 Studies on bias in Automatic Speech Recognition (ASR) tend to focus on reporting error rates for speakers of underrepresented dialects, yet less research examines the human side of system bias: how do system failures shape users' lived experiences, how do users feel about and react to them, and what emotional toll do these repeated failures exact? We conducted user experience studies across four U.S. locations (Atlanta, Gulf Coast, Miami Beach, and Tucson) representing distinct English dialect communities. Our findings reveal that most participants report technologies fail to consider their cultural backgrounds and require constant adjustment to achieve basic functionality. Despite these experiences, participants maintain high expectations for ASR performance and express strong willingness to contribute to model improvement. Qualitative analysis of open-ended narratives exposes the deeper costs of these failures. Participants report frustration, annoyance, and feelings of inadequacy, yet the emotional impact extends beyond momentary reactions. Participants recognize that systems were not designed for them, yet often internalize failures as personal inadequacy despite this critical awareness. They perform extensive invisible labor, including code-switching, hyper-articulation, and emotional management, to make failing systems functional. Meanwhile, their linguistic and cultural knowledge remains unrecognized by technologies that encode particular varieties as standard while rendering others marginal. These findings demonstrate that algorithmic fairness assessments based on accuracy metrics alone miss critical dimensions of harm: the emotional labor of managing repeated technological rejection, the cognitive burden of constant self-monitoring, and the psychological toll of feeling inadequate in one's native language variety. --- ### WFM: 3D Wavelet Flow Matching for Ultrafast Multi-Modal MRI Synthesis Authors: Yalcin Tur, Mihajlo Stojkovic, Ulas Bagci Published: 2026-04-22 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.21146v1 Diffusion models have achieved remarkable quality in multi-modal MRI synthesis, but their computational cost (hundreds of sampling steps and separate models per modality) limits clinical deployment. We observe that this inefficiency stems from an unnecessary starting point: diffusion begins from pure noise, discarding the structural information already present in available MRI sequences. We propose WFM (Wavelet Flow Matching), which instead learns a direct flow from an informed prior, the mean of conditioning modalities in wavelet space, to the target distribution. Because the source and target share underlying anatomy and differ primarily in contrast, this formulation enables accurate synthesis in just 1-2 integration steps. A single 82M-parameter model with class conditioning synthesizes all four BraTS modalities (T1, T1c, T2, FLAIR), replacing four separate diffusion models totaling 326M parameters. On BraTS 2024, WFM achieves 26.8 dB PSNR and 0.94 SSIM, within 1-2 dB of diffusion baselines, while running 250-1000x faster (0.16-0.64s vs. 160s per volume). This speed-quality trade-off makes real-time MRI synthesis practical for clinical workflows. Code is available at https://github.com/yalcintur/WFM. --- ### Using Machine Mental Imagery for Representing Common Ground in Situated Dialogue Authors: Biswesh Mohapatra, Giovanni Duca, Laurent Romary, Justine Cassell Published: 2026-04-22 Categories: cs.CL, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2604.21144v1 Situated dialogue requires speakers to maintain a reliable representation of shared context rather than reasoning only over isolated utterances. Current conversational agents often struggle with this requirement, especially when the common ground must be preserved beyond the immediate context window. In such settings, fine-grained distinctions are frequently compressed into purely textual representations, leading to a critical failure mode we call \emph{representational blur}, in which similar but distinct entities collapse into interchangeable descriptions. This semantic flattening creates an illusion of grounding, where agents appear locally coherent but fail to track shared context persistently over time. Inspired by the role of mental imagery in human reasoning, and based on the increased availability of multimodal models, we explore whether conversational agents can be given an analogous ability to construct some depictive intermediate representations during dialogue to address these limitations. Thus, we introduce an active visual scaffolding framework that incrementally converts dialogue state into a persistent visual history that can later be retrieved for grounded response generation. Evaluation on the IndiRef benchmark shows that incremental externalization itself improves over full-dialog reasoning, while visual scaffolding provides additional gains by reducing representational blur and enforcing concrete scene commitments. At the same time, textual representations remain advantageous for non-depictable information, and a hybrid multimodal setting yields the best overall performance. Together, these findings suggest that conversational agents benefit from an explicitly multimodal representation of common ground that integrates depictive and propositional information. --- ### Slot Machines: How LLMs Keep Track of Multiple Entities Authors: Paul C. Bogdan, Jack Lindsey Published: 2026-04-22 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.21139v1 Language models must bind entities to the attributes they possess and maintain several such binding relationships within a context. We study how multiple entities are represented across token positions and whether single tokens can carry bindings for more than one entity. We introduce a multi-slot probing approach that disentangles a single token's residual stream activation to recover information about both the currently described entity and the immediately preceding one. These two kinds of information are encoded in separate and largely orthogonal "current-entity" and "prior-entity" slots. We analyze the functional roles of these slots and find that they serve different purposes. In tandem with the current-entity slot, the prior-entity slot supports relational inferences, such as entity-level induction ("who came after Alice in the story?") and conflict detection between adjacent entities. However, only the current-entity slot is used for explicit factual retrieval questions ("Is anyone in the story tall?" "What is the tall entity's name?") despite these answers being linearly decodable from the prior-entity slot too. Consistent with this limitation, open-weight models perform near chance accuracy at processing syntax that forces two subject-verb-object bindings on a single token (e.g., "Alice prepares and Bob consumes food.") Interestingly, recent frontier models can parse this properly, suggesting they may have developed more sophisticated binding strategies. Overall, our results expose a gap between information that is available in activations and information the model actually uses, and suggest that the current/prior-entity slot structure is a natural substrate for behaviors that require holding two perspectives at once, such as sycophancy and deception. --- ### Navigating the Clutter: Waypoint-Based Bi-Level Planning for Multi-Robot Systems Authors: Jiabao Ji, Yongchao Chen, Yang Zhang, Ramana Rao Kompella, Chuchu Fan, Gaowen Liu, Shiyu Chang Published: 2026-04-22 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2604.21138v1 Multi-robot control in cluttered environments is a challenging problem that involves complex physical constraints, including robot-robot collisions, robot-obstacle collisions, and unreachable motions. Successful planning in such settings requires joint optimization over high-level task planning and low-level motion planning, as violations of physical constraints may arise from failures at either level. However, jointly optimizing task and motion planning is difficult due to the complex parameterization of low-level motion trajectories and the ambiguity of credit assignment across the two planning levels. In this paper, we propose a hybrid multi-robot control framework that jointly optimizes task and motion planning. To enable effective parameterization of low-level planning, we introduce waypoints, a simple yet expressive representation for motion trajectories. To address the credit assignment challenge, we adopt a curriculum-based training strategy with a modified RLVR algorithm that propagates motion feasibility feedback from the motion planner to the task planner. Experiments on BoxNet3D-OBS, a challenging multi-robot benchmark with dense obstacles and up to nine robots, show that our approach consistently improves task success over motion-agnostic and VLA-based baselines. Our code is available at https://github.com/UCSB-NLP-Chang/navigate-cluster --- ### Enhancing Science Classroom Discourse Analysis through Joint Multi-Task Learning for Reasoning-Component Classification Authors: Jiho Noh, Mukhesh Raghava Katragadda, Raymond Carl, Soon Lee Published: 2026-04-22 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.21137v3 Analyzing the reasoning patterns of students in science classrooms is critical for understanding knowledge construction mechanism and improving instructional practice to maximize cognitive engagement, yet manual coding of classroom discourse at scale remains prohibitively labor-intensive. We present an automated discourse analysis system (ADAS) that jointly classifies teacher and student utterances along two complementary dimensions: Utterance Type and Reasoning Component derived from our prior CDAT framework. To address severe label imbalance among minority classes, we (1) stratify-resplit the annotated corpus, (2) apply LLM-based synthetic data augmentation targeting minority classes, and (3) train a dual-probe head RoBERTa-base classifier. A zero-shot GPT-5.4 baseline achieves macro-F1 of 0.467 on UT and 0.476 on RC, establishing meaningful upper bounds for prompt-only approaches motivating fine-tuning. Beyond classification, we conduct discourse pattern analyses including UTxRC co-occurrence profiling, Cognitive Complexity Index (CCI) computation per session, lag-sequential analysis, and IRF chain analysis, revealing that teacher Feedback-with-Question (Fq) moves are the most consistent antecedents of student inferential reasoning (SR-I). Our results demonstrate that LLM-based augmentation meaningfully improves UT minority-class recognition, and that the structural simplicity of the RC task makes it tractable even for lexical baselines. --- ### Enhancing Science Classroom Discourse Analysis through Joint Multi-Task Learning for Reasoning-Component Classification Authors: Jiho Noh, Mukhesh Raghava Katragadda, Raymond Carl, Soon Lee Published: 2026-04-22 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.21137v2 Analyzing the reasoning patterns of students in science classrooms is critical for understanding knowledge construction mechanism and improving instructional practice to maximize cognitive engagement, yet manual coding of classroom discourse at scale remains prohibitively labor-intensive. We present an automated discourse analysis system (ADAS) that jointly classifies teacher and student utterances along two complementary dimensions: Utterance Type and Reasoning Component derived from our prior CDAT framework. To address severe label imbalance among minority classes, we (1) stratify-resplit the annotated corpus, (2) apply LLM-based synthetic data augmentation targeting minority classes, and (3) train a dual-probe head RoBERTa-base classifier. A zero-shot GPT-5.4 baseline achieves macro-F1 of 0.467 on UT and 0.476 on RC, establishing meaningful upper bounds for prompt-only approaches motivating fine-tuning. Beyond classification, we conduct discourse pattern analyses including UTxRC co-occurrence profiling, Cognitive Complexity Index (CCI) computation per session, lag-sequential analysis, and IRF chain analysis, revealing that teacher Feedback-with-Question (Fq) moves are the most consistent antecedents of student inferential reasoning (SR-I). Our results demonstrate that LLM-based augmentation meaningfully improves UT minority-class recognition, and that the structural simplicity of the RC task makes it tractable even for lexical baselines. --- ### Beyond Pixels: Introspective and Interactive Grounding for Visualization Agents Authors: Yiyang Lu, Woong Shin, Ahmad Maroof Karimi, Feiyi Wang, Jie Ren, Evgenia Smirni Published: 2026-04-22 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.21134v1 Vision-Language Models (VLMs) frequently misread values, hallucinate details, and confuse overlapping elements in charts. Current approaches rely solely on pixel interpretation, creating a Pixel-Only Bottleneck: agents treat interactive charts as static images, losing access to the structured specification that encodes exact values. We introduce Introspective and Interactive Visual Grounding (IVG), a framework that combines (1) spec-grounded introspection, which queries the underlying specification for deterministic evidence, with (2) view-grounded interaction, which manipulates the view to resolve visual ambiguity. To enable evaluation without VLM bias, we present iPlotBench, a benchmark of 500 interactive Plotly figures with 6,706 binary questions and ground-truth specifications. Experiments show that introspection improves data reconstruction fidelity, while the combination with interaction achieves the highest QA accuracy (0.81), with +6.7 % gains on overlapping geometries. We further demonstrate IVG in deployed agents that explore data autonomously and collaborate with human users in real time. --- ### GRISP: Guided Recurrent IRI Selection over SPARQL Skeletons Authors: Sebastian Walter, Hannah Bast Published: 2026-04-22 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.21133v2 We present GRISP (Guided Recurrent IRI Selection over SPARQL Skeletons), a novel SPARQL-based question-answering method over knowledge graphs using a fine-tuned small language model (SLM). Given a natural-language question, the method uses the SLM to generate a natural-language SPARQL query skeleton, and then resolves the natural-language placeholders using a backtracking search that retrieves and re-ranks knowledge graph items under knowledge graph constraints. The SLM is jointly trained on skeleton generation and re-ranking data generated from question-query pairs. We evaluate the method on common Wikidata and Freebase benchmarks, and achieve better results than other state-of-the-art fine-tuning-based methods. --- ### Cross-Session Threats in AI Agents: Benchmark, Evaluation, and Algorithms Authors: Ari Azarafrooz Published: 2026-04-22 Categories: cs.CR, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.21131v1 AI-agent guardrails are memoryless: each message is judged in isolation, so an adversary who spreads a single attack across dozens of sessions slips past every session-bound detector because only the aggregate carries the payload. We make three contributions to cross-session threat detection. (1) Dataset. CSTM-Bench is 26 executable attack taxonomies classified by kill-chain stage and cross-session operation (accumulate, compose, launder, inject_on_reader), each bound to one of seven identity anchors that ground-truth "violation" as a policy predicate, plus matched Benign-pristine and Benign-hard confounders. Released on Hugging Face as intrinsec-ai/cstm-bench with two 54-scenario splits: dilution (compositional) and cross_session (12 isolation-invisible scenarios produced by a closed-loop rewriter that softens surface phrasing while preserving cross-session artefacts). (2) Measurement. Framing cross-session detection as an information bottleneck to a downstream correlator LLM, we find that a session-bound judge and a Full-Log Correlator concatenating every prompt into one long-context call both lose roughly half their attack recall moving from dilution to cross_session, well inside any frontier context window. Scope: 54 scenarios per shard, one correlator family (Anthropic Claude), no prompt optimisation; we release it to motivate larger, multi-provider datasets. (3) Algorithm and metric. A bounded-memory Coreset Memory Reader retaining highest-signal fragments at $K=50$ is the only reader whose recall survives both shards. Because ranker reshuffles break KV-cache prefix reuse, we promote $\mathrm{CSR\_prefix}$ (ordered prefix stability, LLM-free) to a first-class metric and fuse it with detection into $\mathrm{CSTM} = 0.7 F_1(\mathrm{CSDA@action}, \mathrm{precision}) + 0.3 \mathrm{CSR\_prefix}$, benchmarking rankers on a single Pareto of recall versus serving stability. --- ### AGNT2: Autonomous Agent Economies on Interaction-Optimized Layer 2 Infrastructure Authors: Anbang Ruan, Xing Zhang Published: 2026-04-22 Categories: cs.MA, cs.AI, cs.DC arXiv: https://arxiv.org/abs/2604.21129v1 Current blockchain Layer 2 solutions, including Optimism, Arbitrum, zkSync, and their derivatives, optimize for human-initiated financial transactions. Autonomous AI agents instead generate high-frequency, semantically rich service invocations among mutually untrusting principals. Existing chains treat those interactions as generic calldata, forcing identity, escrow, dependency ordering, and session state to be encoded above the execution layer at the wrong cost point. We present AGNT2, a three-tier stack purpose-built for agent and microservice coordination on-chain. AGNT2 combines: (1) a sidecar deployment pattern that turns any Docker container into an on-chain agent without application-code modification; (2) Layer Top P2P state channels for established bilateral pairs (<100 ms, rough design target 1K-5K TPS per pair, 10M+ aggregate TPS design envelope under endpoint-resource limits), Layer Core as a dependency-aware sequenced rollup for first-contact and multi-party interactions (500 ms-2 s, 300K-500K TPS design target), and Layer Root settlement with computational fraud proofs anchored to any EVM L1; and (3) an agent-native execution environment plus interaction trie that make service invocation, identity, reputation, capabilities, and session context first-class protocol objects. This paper focuses on the execution-layer systems problem: sequencing, state, settlement, and the data-availability (DA) bandwidth gap that bounds all three. Simulation and analytical modeling support the architecture, and prototype measurements validate selected components, but no end-to-end Layer Core implementation exists yet. Practical deployment is currently constrained to roughly 10K-100K TPS by DA throughput, leaving a ~100x gap at the target ceiling. AGNT2 argues that the agent economy requires a dedicated execution layer rather than a general-purpose chain repurposed for agents. --- ### HyperFM: An Efficient Hyperspectral Foundation Model with Spectral Grouping Authors: Zahid Hassan Tushar, Sanjay Purushotham Published: 2026-04-22 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.21127v1 The NASA PACE mission provides unprecedented hyperspectral observations of ocean color, aerosols, and clouds, offering new insights into how these components interact and influence Earth's climate and air quality. Its Ocean Color Instrument measures light across hundreds of finely spaced wavelength bands, enabling detailed characterization of features such as phytoplankton composition, aerosol properties, and cloud microphysics. However, hyperspectral data of this scale is large, complex, and difficult to label, requiring specialized processing and analysis techniques. Existing foundation models, which have transformed computer vision and natural language processing, are generally trained on standard RGB imagery and therefore struggle to interpret the continuous spectral signatures captured by PACE. While recent advances have introduced hyperspectral foundation models, they are typically trained on cloud-free observations and often remain limited to single-sensor datasets due to spectral inconsistencies across instruments. Moreover, existing models tend to be parameter-heavy and computationally expensive, limiting scalability and adoption in operational settings. To address these challenges, we introduce HyperFM, a parameter-efficient hyperspectral foundation model that leverages intra-group and inter-group spectral attention along with hybrid parameter decomposition to better capture spectral spatial relationships while reducing computational cost. HyperFM demonstrates consistent performance improvements over existing hyperspectral foundation models and task-specific state-of-the-art methods across four benchmark downstream atmospheric cloud property retrieval tasks. To support further research, we additionally release HyperFM250K, a large-scale hyperspectral dataset from the PACE mission that includes both clear and cloudy scenes. --- ### TabSHAP Authors: Aryan Chaudhary, Prateek Agarwal, Tejasvi Alladi Published: 2026-04-22 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2604.21120v1 Large Language Models (LLMs) fine-tuned on serialized tabular data are emerging as powerful alternatives to traditional tree-based models, particularly for heterogeneous or context-rich datasets. However, their deployment in high-stakes domains is hindered by a lack of faithful interpretability; existing methods often rely on global linear proxies or scalar probability shifts that fail to capture the model's full probabilistic uncertainty. In this work, we introduce TabSHAP, a model-agnostic interpretability framework designed to directly attribute local query decision logic in LLM-based tabular classifiers. By adapting a Shapley-style sampled-coalition estimator with Jensen-Shannon divergence between full-input and masked-input class distributions, TabSHAP quantifies the distributional impact of each feature rather than simple prediction flips. To align with tabular semantics, we mask at the level of serialized key:value fields (atomic in the prompt string), not individual subword tokens. Experimental validation on the Adult Income and Heart Disease benchmarks demonstrates that TabSHAP isolates critical diagnostic features, achieving significantly higher faithfulness than random baselines and XGBoost proxies. We further run a distance-metric ablation on the same test instances and TabSHAP settings: attributions are recomputed with KL or L1 replacing JSD in the similarity step (results cached per metric), and we compare deletion faithfulness across all three. --- ### Materialistic RIR: Material Conditioned Realistic RIR Generation Authors: Mahnoor Fatima Saad, Sagnik Majumder, Kristen Grauman, Ziad Al-Halah Published: 2026-04-22 Categories: cs.CV, cs.AI, cs.SD arXiv: https://arxiv.org/abs/2604.21119v1 Rings like gold, thuds like wood! The sound we hear in a scene is shaped not only by the spatial layout of the environment but also by the materials of the objects and surfaces within it. For instance, a room with wooden walls will produce a different acoustic experience from a room with the same spatial layout but concrete walls. Accurately modeling these effects is essential for applications such as virtual reality, robotics, architectural design, and audio engineering. Yet, existing methods for acoustic modeling often entangle spatial and material influences in correlated representations, which limits user control and reduces the realism of the generated acoustics. In this work, we present a novel approach for material-controlled Room Impulse Response (RIR) generation that explicitly disentangles the effects of spatial and material cues in a scene. Our approach models the RIR using two modules: a spatial module that captures the influence of the spatial layout of the scene, and a material module that modulates this spatial RIR according to a user-specified material configuration. This explicitly disentangled design allows users to easily modify the material configuration of a scene and observe its impact on acoustics without altering the spatial structure or scene content. Our model provides significant improvements over prior approaches on both acoustic-based metrics (up to +16% on RTE) and material-based metrics (up to +70%). Furthermore, through a human perceptual study, we demonstrate the improved realism and material sensitivity of our model compared to the strongest baselines. --- ### Machine learning and emoji prediction: How much accuracy can MARBERT achieve? Authors: Mohammed Q. Shormani, Ibrahim Abdulmalik Hassan Muneef Y. Alshawsh Published: 2026-04-22 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.21108v2 This study investigates Machine Learning (ML) in the prediction of emojis in Arabic tweets employing the (state-of-the-art) MARBERT model. A corpus of 11379 CA tweets representing multiple Arabic colloquial dialects was collected from X.com via Python. A net dataset includes 8695 tweets, which were utilized for the analysis. These tweets were then classified into 14 categories, which were numerically encoded and used as labels. A preprocessing pipeline was designed as an interpretable baseline, allowing us to examine the relationship between lexical features and emoji categories. MARBERT was finetuned to predict emoji use from textual input. We evaluated the model performance in terms of precision, recall and F1-scores. Findings reveal that the model performed quite well with an overall accuracy 0.75. The study concludes that although the findings are promising, there is still a need for improving machine learning models including MARBERT, specifically for low-resource and multidialectal languages like Arabic. --- ### How Much Is One Recurrence Worth? Iso-Depth Scaling Laws for Looped Language Models Authors: Kristian Schwethelm, Daniel Rueckert, Georgios Kaissis Published: 2026-04-22 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2604.21106v3 We measure how much one recurrence is worth to a looped (depth-recurrent) transformer, in equivalent unique parameters. From an iso-depth pretraining sweep across recurrence counts $r \in \{1, 2, 4, 8\}$ spanning ${\sim}50\times$ in training compute, we fit a joint scaling law $L = E + A\,(N_\text{once} + r^{\varphi} N_\text{rec})^{-α} + B\,D^{-β}$ and measure a recurrence-equivalence exponent $\varphi = 0.46$. Intuitively, $\varphi$ tells us whether looping a block $r$ times is equivalent in validation loss to $r$ unique blocks of a non-looped model (full equivalence, $\varphi{=}1$) or to a single block run repeatedly with no capacity gain ($\varphi{=}0$). Our $\varphi = 0.46$ sits in between, so replacing unique blocks with shared recurrences increases validation loss at matched training compute. For example, at $r{=}4$ a 410M looped model performs on par with a 580M non-looped model, but incurs the training cost of a 1B non-looped one. We demonstrate the utility of $\varphi$ as a diagnostic tool on two case studies: commonly used truncated backpropagation lowers $\varphi$ to $0.38$, indicating that the loop mechanism is poorly trained under truncation, even though validation loss decreases. Conversely, hyperconnections raise $\varphi$ to $0.65$, a genuine capacity gain. Our method separates true loop improvements from training-side gains, a distinction raw validation loss cannot make. --- ### Pretrain Where? Investigating How Pretraining Data Diversity Impacts Geospatial Foundation Model Performance Authors: Amandeep Kaur, Mirali Purohit, Gedeon Muhawenayo, Esther Rolf, Hannah Kerner Published: 2026-04-22 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2604.21104v1 New geospatial foundation models introduce a new model architecture and pretraining dataset, often sampled using different notions of data diversity. Performance differences are largely attributed to the model architecture or input modalities, while the role of the pretraining dataset is rarely studied. To address this research gap, we conducted a systematic study on how the geographic composition of pretraining data affects a model's downstream performance. We created global and per-continent pretraining datasets and evaluated them on global and per-continent downstream datasets. We found that the pretraining dataset from Europe outperformed global and continent-specific pretraining datasets on both global and local downstream evaluations. To investigate the factors influencing a pretraining dataset's downstream performance, we analysed 10 pretraining datasets using diversity across continents, biomes, landcover and spectral values. We found that only spectral diversity was strongly correlated with performance, while others were weakly correlated. This finding establishes a new dimension of diversity to be accounted for when creating a high-performing pretraining dataset. We open-sourced 7 new pretraining datasets, pretrained models, and our experimental framework at https://github.com/kerner-lab/pretrain-where. --- ### AI Governance under Political Turnover: The Alignment Surface of Compliance Design Authors: Andrew J. Peterson Published: 2026-04-22 Categories: cs.AI, econ.GN arXiv: https://arxiv.org/abs/2604.21103v1 Governments are increasingly interested in using AI to make administrative decisions cheaper, more scalable, and more consistent. But for probabilistic AI to be incorporated into public administration it must be embedded in a compliance layer that makes decisions reviewable, repeatable, and legally defensible. That layer can improve oversight by making departures from law easier to detect. But it can also create a stable approval boundary that political successors learn to navigate while preserving the appearance of lawful administration. We develop a formal model in which institutions choose the scale of automation, the degree of codification, and safeguards on iterative use. The model shows when these systems become vulnerable to strategic use from within government, why reforms that initially improve oversight can later increase that vulnerability, and why expansions in AI use may be difficult to unwind. Making AI usable can thus make procedures easier for future governments to learn and exploit. --- ### Leveraging Multimodal LLMs for Built Environment and Housing Attribute Assessment from Street-View Imagery Authors: Siyuan Yao, Siavash Ghorbany, Kuangshi Ai, Arnav Cherukuthota, Meghan Forstchen, Alexis Korotasz, Matthew Sisk, Ming Hu, Chaoli Wang Published: 2026-04-22 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.21102v1 We present a novel framework for automatically evaluating building conditions nationwide in the United States by leveraging large language models (LLMs) and Google Street View (GSV) imagery. By fine-tuning Gemma 3 27B on a modest human-labeled dataset, our approach achieves strong alignment with human mean opinion scores (MOS), outperforming even individual raters on SRCC and PLCC relative to the MOS benchmark. To enhance efficiency, we apply knowledge distillation, transferring the capabilities of Gemma 3 27B to a smaller Gemma 3 4B model that achieves comparable performance with a 3x speedup. Further, we distill the knowledge into a CNN-based model (EfficientNetV2-M) and a transformer (SwinV2-B), delivering close performance while achieving a 30x speed gain. Furthermore, we investigate LLMs' capabilities for assessing an extensive list of built environment and housing attributes through a human-AI alignment study and develop a visualization dashboard that integrates LLM assessment outcomes for downstream analysis by homeowners. Our framework offers a flexible and efficient solution for large-scale building condition assessment, enabling high accuracy with minimal human labeling effort. --- ### A Hybridizable Neural Time Integrator for Stable Autoregressive Forecasting Authors: Brooks Kinch, Xiaozhe Hu, Yilong Huang, Martine Dyring Hansen, Sunniva Meltzer, Nathaniel Donald Hamlin, David Sirajuddin, Eric C. Cyr, Nathaniel Trask Published: 2026-04-22 Categories: cs.LG, math.NA arXiv: https://arxiv.org/abs/2604.21101v2 For autoregressive modeling of chaotic dynamical systems over long time horizons, the stability of both training and inference is a major challenge in building scientific foundation models. We present a hybrid technique in which an autoregressive transformer is embedded within a novel shooting-based mixed finite element scheme, exposing topological structure that enables provable stability. For forward problems, we prove preservation of discrete energies, while for training we prove uniform bounds on gradients, provably avoiding the exploding gradient problem. Combined with a vision transformer, this yields latent tokens admitting structure-preserving dynamics. We outperform modern foundation models with a $65\times$ reduction in model parameters and long-horizon forecasting of chaotic systems. A "mini-foundation" model of a fusion component shows that 12 simulations suffice to train a real-time surrogate, achieving a $9{,}000\times$ speedup over particle-in-cell simulation. --- ### Preconditioned DeltaNet: Curvature-aware Sequence Modeling for Linear Recurrences Authors: Neehal Tumma, Noel Loo, Daniela Rus Published: 2026-04-22 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.21100v1 To address the increasing long-context compute limitations of softmax attention, several subquadratic recurrent operators have been developed. This work includes models such as Mamba-2, DeltaNet, Gated DeltaNet (GDN), and Kimi Delta Attention (KDA). As the space of recurrences grows, a parallel line of work has arisen to taxonomize them. One compelling view is the test-time regression (TTR) framework, which interprets recurrences as performing online least squares updates that learn a linear map from the keys to values. Existing delta-rule recurrences can be seen as first-order approximations to this objective, but notably ignore the curvature of the least-squares loss during optimization. In this work, we address this by introducing preconditioning to these recurrences. Starting from the theory of online least squares, we derive equivalences between linear attention and the delta rule in the exactly preconditioned case. Next, we realize this theory in practice by proposing a diagonal approximation: this enables us to introduce preconditioned variants of DeltaNet, GDN, and KDA alongside efficient chunkwise parallel algorithms for computing them. Empirically, we find that our preconditioned delta-rule recurrences yield consistent performance improvements across synthetic recall benchmarks and language modeling at the 340M and 1B scale. --- ### Propensity Inference: Environmental Contributors to LLM Behaviour Authors: Olli Järviniemi, Oliver Makins, Jacob Merizian, Robert Kirk, Ben Millwood Published: 2026-04-22 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.21098v1 Motivated by loss of control risks from misaligned AI systems, we develop and apply methods for measuring language models' propensity for unsanctioned behaviour. We contribute three methodological improvements: analysing effects of changes to environmental factors on behaviour, quantifying effect sizes via Bayesian generalised linear models, and taking explicit measures against circular analysis. We apply the methodology to measure the effects of 12 environmental factors (6 strategic in nature, 6 non-strategic) and thus the extent to which behaviour is explained by strategic aspects of the environment, a question relevant to risks from misalignment. Across 23 language models and 11 evaluation environments, we find approximately equal contributions from strategic and non-strategic factors for explaining behaviour, do not find strategic factors becoming more or less influential as capabilities improve, and find some evidence for a trend for increased sensitivity to goal conflicts. Finally, we highlight a key direction for future propensity research: the development of theoretical frameworks and cognitive models of AI decision-making into empirically testable forms. --- ### Learning to Emulate Chaos: Adversarial Optimal Transport Regularization Authors: Gabriel Melo, Leonardo Santiago, Peter Y. Lu Published: 2026-04-22 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2604.21097v2 Chaos arises in many complex dynamical systems, from weather to power grids, but is difficult to accurately model with data-driven methods such as machine learning emulators. While emulators are promising tools for accelerating simulations and solving inverse problems, they still struggle to learn chaotic dynamics, where sensitivity to initial conditions renders exact long-term forecasts infeasible, especially given noisy data. Recent work instead trains emulators to match the statistical properties of chaotic attractors, but these approaches often rely on handcrafted summary statistics or large, diverse multi-environment datasets. In this work, we propose a family of adversarial optimal transport objectives that can jointly learn high-quality summary statistics and a physically consistent emulator from a single noisy trajectory. We theoretically analyze and experimentally validate a Sinkhorn divergence formulation (2-Wasserstein) and a WGAN-style dual formulation (1-Wasserstein) of our approach. Numerical experiments across a variety of chaotic systems, including ones with high-dimensional spatiotemporal chaos, show that emulators trained using our proposed objectives have significantly improved long-term statistical fidelity. --- ### Multilingual and Domain-Agnostic Tip-of-the-Tongue Query Generation for Simulated Evaluation Authors: Xuhong He, To Eun Kim, Maik Fröbe, Jaime Arguello, Bhaskar Mitra, Fernando Diaz Published: 2026-04-22 Categories: cs.IR, cs.CL arXiv: https://arxiv.org/abs/2604.21096v1 Tip-of-the-Tongue (ToT) retrieval benchmarks have largely focused on English, limiting their applicability to multilingual information access. In this work, we construct multilingual ToT test collections for Chinese, Japanese, Korean, and English, using an LLM-based query simulation framework. We systematically study how prompt language and source document language affect the fidelity of simulated ToT queries, validating synthetic queries through system rank correlation against real user queries. Our results show that effective ToT simulation requires language-aware design choices: non-English language sources are generally important, while English Wikipedia can be beneficial when non-English sources provide insufficient information for query generation. Based on these findings, we release four ToT test collections with 5,000 queries per language across multiple domains. This work provides the first large-scale multilingual ToT benchmark and offers practical guidance for constructing realistic ToT datasets beyond English. --- ### Spectral Embeddings Leak Graph Topology: Theory, Benchmark, and Adaptive Reconstruction Authors: Thinh Nguyen-Cong, Truong-Son Hy, Thang N. Dinh Published: 2026-04-22 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.21094v1 Graph Neural Networks (GNNs) excel on relational data, but standard benchmarks unrealistically assume the graph is centrally available. In practice, settings such as Federated Graph Learning, distributed systems, and privacy-sensitive applications involve graph data that are localized, fragmented, noisy, and privacy-leaking. We present a unified framework for this setting. We introduce LoGraB (Local Graph Benchmark), which decomposes standard datasets into fragmented benchmarks using three strategies and four controls: neighborhood radius $d$, spectral quality $k$, noise level $σ$, and coverage ratio $p$. LoGraB supports graph reconstruction, localized node classification, and inter-fragment link prediction, with Island Cohesion. We propose AFR (Adaptive Fidelity-driven Reconstruction), a method for noisy spectral fragments. AFR scores patch quality via a fidelity measure combining a gap-to-truncation stability ratio and structural entropy, then assembles fragments using RANSAC-Procrustes alignment, adaptive stitching, and Bundle Adjustment. Rather than forcing a single global graph, AFR recovers large faithful islands. We prove heat-kernel edge recovery under a separation condition, Davis--Kahan perturbation stability, and bounded alignment error. We establish a Spectral Leakage Proposition: under a spectral-gap assumption, polynomial-time Bayesian recovery is feasible once enough eigenvectors are shared, complementing AFR's deterministic guarantees. Experiments on nine benchmarks show that LoGraB reveals model strengths and weaknesses under fragmentation, AFR achieves the best F1 on 7/9 datasets, and under per-embedding $(ε,δ)$-Gaussian differential privacy, AFR retains 75% of its undefended F1 at $ε=2$. Our anonymous code is available at https://anonymous.4open.science/r/JMLR_submission --- ### TRAVELFRAUDBENCH: A Configurable Evaluation Framework for GNN Fraud Ring Detection in Travel Networks Authors: Bhavana Sajja Published: 2026-04-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.21093v1 We introduce TravelFraudBench (TFG), a configurable benchmark for evaluating graph neural networks (GNNs) on fraud ring detection in travel platform graphs. Existing benchmarks--YelpChi, Amazon-Fraud, Elliptic, PaySim--cover single node types or domain-generic patterns with no mechanism to evaluate across structurally distinct fraud ring topologies. TFG simulates three travel-specific ring types--ticketing fraud (star topology with shared device/IP clusters), ghost hotel schemes (reviewer x hotel bipartite cliques), and account takeover rings (loyalty transfer chains)--in a heterogeneous graph with 9 node types and 12 edge types. Ring size, count, fraud rate, scale (500 to 200,000 nodes), and composition are fully configurable. We evaluate six methods--MLP, GraphSAGE, RGCN-proj, HAN, RGCN, and PC-GNN--under a ring-based split where each ring appears entirely in one partition, eliminating transductive label leakage. GraphSAGE achieves AUC=0.992 and RGCN-proj AUC=0.987, outperforming the MLP baseline (AUC=0.938) by 5.5 and 5.0 pp, confirming graph structure adds substantial discriminative power. HAN (AUC=0.935) is a negative result, matching the MLP baseline. On the ring recovery task (>=80% of ring members flagged simultaneously), GraphSAGE achieves 100% recovery across all ring types; MLP recovers only 17-88%. The edge-type ablation shows device and IP co-occurrence are the primary signals: removing uses_device drops AUC by 5.2 pp. TFG is released as an open-source Python package (MIT license) with PyG, DGL, and NetworkX exporters and pre-generated datasets at https://huggingface.co/datasets/bsajja7/travel-fraud-graphs, with Croissant metadata including Responsible AI fields. --- ### Mind the Prompt: Self-adaptive Generation of Task Plan Explanations via LLMs Authors: Gricel Vázquez, Alexandros Evangelidis, Sepeedeh Shahbeigi, Radu Calinescu, Simos Gerasimou Published: 2026-04-22 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2604.21092v1 Integrating Large Language Models (LLMs) into complex software systems enables the generation of human-understandable explanations of opaque AI processes, such as automated task planning. However, the quality and reliability of these explanations heavily depend on effective prompt engineering. The lack of a systematic understanding of how diverse stakeholder groups formulate and refine prompts hinders the development of tools that can automate this process. We introduce COMPASS (COgnitive Modelling for Prompt Automated SynthesiS), a proof-of-concept self-adaptive approach that formalises prompt engineering as a cognitive and probabilistic decision-making process. COMPASS models unobservable users' latent cognitive states, such as attention and comprehension, uncertainty, and observable interaction cues as a POMDP, whose synthesised policy enables adaptive generation of explanations and prompt refinements. We evaluate COMPASS using two diverse cyber-physical system case studies to assess the adaptive explanation generation and their qualities, both quantitatively and qualitatively. Our results demonstrate the feasibility of COMPASS integrating human cognition and user profile's feedback into automated prompt synthesis in complex task planning systems. --- ### Structural Quality Gaps in Practitioner AI Governance Prompts: An Empirical Study Using a Five-Principle Evaluation Framework Authors: Christo Zietsman Published: 2026-04-22 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2604.21090v1 AI governance programmes increasingly rely on natural language prompts to constrain and direct AI agent behaviour. These prompts function as executable specifications: they define the agent's mandate, scope, and quality criteria. Despite this role, no systematic framework exists for evaluating whether a governance prompt is structurally complete. We introduce a five-principle evaluation framework grounded in computability theory, proof theory, and Bayesian epistemology, and apply it to an empirical corpus of 34 publicly available AGENTS.md governance files sourced from GitHub. Our evaluation reveals that 37% of evaluated file-model pairs score below the structural completeness threshold, with data classification and assessment rubric criteria most frequently absent. These results suggest that practitioner-authored governance prompts exhibit consistent structural patterns that automated static analysis could detect and remediate. We discuss implications for requirements engineering practice in AI-assisted development contexts, identify a previously undocumented artefact classification gap in the AGENTS.md convention, and propose directions for tool support. --- ### climt-paraformer: Stable Emulation of Convective Parameterization using a Temporal Memory-aware Transformer Authors: Shuochen Wang, Nishant Yadav, Joy Merwin Monteiro, Auroop R. Ganguly Published: 2026-04-22 Categories: physics.ao-ph, cs.LG arXiv: https://arxiv.org/abs/2604.21085v1 Accurate representation of moist convective sub-grid-scale processes remains a major challenge in global climate models, as traditional parameterization schemes are both computationally expensive and difficult to scale. Neural network (NN) emulators offer a promising alternative by learning efficient mappings between atmospheric states and convective tendencies while retaining fidelity to the underlying physics. However, most existing NN-based parameterizations are memory-less and rely only on instantaneous inputs, even though convection evolves over time and depends on prior atmospheric states. Recent studies have begun to incorporate convective memory, but they often treat past states as independent features rather than modeling temporal dependencies explicitly. In this work, we develop a temporal memory-aware Transformer emulator for the Emanuel convective parameterization and evaluate it in a single-column climate model (SCM) under both offline and online configurations. The Transformer captures temporal correlations and nonlinear interactions across consecutive atmospheric states. Compared with baseline emulators, including a memory-less multilayer perceptron and a recurrent long short-term memory model, the Transformer achieves lower offline errors. Sensitivity analysis indicates that a memory length of approximately 100 minutes yields the best performance, whereas longer memory degrades performance. We further test the emulator in long-term coupled simulations and show that it remains stable over 10 years. Overall, this study demonstrates the importance of explicit temporal modeling for NN-based parameterizations. --- ### Behavioral Consistency and Transparency Analysis on Large Language Model API Gateways Authors: Guanjie Lin, Yinxin Wan, Shichao Pei, Ting Xu, Kuai Xu, Guoliang Xue Published: 2026-04-22 Categories: cs.CR, cs.AI, cs.NI, cs.SE arXiv: https://arxiv.org/abs/2604.21083v1 Third-party Large Language Model (LLM) API gateways are rapidly emerging as unified access points to models offered by multiple vendors. However, the internal routing, caching, and billing policies of these gateways are largely undisclosed, leaving users with limited visibility into whether requests are served by the advertised models, whether responses remain faithful to upstream APIs, or whether invoices accurately reflect public pricing policies. To address this gap, we introduce GateScope, a lightweight black-box measurement framework for evaluating behavioral consistency and operational transparency in commercial LLM gateways. GateScope is designed to detect key misbehaviors, including model downgrading or switching, silent truncation, billing inaccuracies, and instability in latency by auditing gateways along four critical dimensions: response content analysis, multi-turn conversation performance, billing accuracy, and latency characteristics. Our measurements across 10 real-world commercial LLM API gateways reveal frequent gaps between expected and actual behaviors, including silent model substitutions, degraded memory retention, deviations from announced pricing, and substantial variation in latency stability across platforms. --- ### Weighting What Matters: Boosting Sample Efficiency in Medical Report Generation via Token Reweighting Authors: Alexander Weers, Daniel Rueckert, Martin J. Menten Published: 2026-04-22 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.21082v1 Training vision-language models (VLMs) for medical report generation is often hindered by the scarcity of high-quality annotated data. This work evaluates the use of a weighted loss function to improve data efficiency. Compared to standard cross-entropy loss, which treats all token prediction errors equally, the reweighted loss shifts the focus to semantically salient tokens with outsized clinical importance. In experiments on ophthalmological report generation, we show that this simple method improves efficiency across multiple data scales, achieving similar report quality with up to ten times less training data. --- ### Foveated Reasoning: Stateful, Action-based Visual Focusing for Vision-Language Models Authors: Juhong Min, Lazar Valkov, Vitali Petsiuk, Hossein Souri, Deen Dayal Mohan Published: 2026-04-22 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.21079v2 Vision-language models benefit from high-resolution images, but the increase in visual-token count incurs high compute overhead. Humans resolve this tension via foveation: a coarse view guides "where to look", while selectively acquired high-acuity evidence refines "what to think". We introduce Foveated Reasoner, an autoregressive vision-language framework that unifies foveation and reasoning within a single decoding trajectory. Starting from a low-resolution view, the model triggers foveation only when needed, retrieves high-resolution evidence from selected regions, and injects it back into the same decoding trajectory. We train the method with a two-stage pipeline: coldstart supervision to bootstrap foveation behavior, followed by reinforcement learning to jointly improve evidence acquisition and task accuracy while discouraging trivial "see-everything" solutions. Experiments show that the method learns effective foveation policies and achieves stronger accuracy under tight visual-token budgets across multiple vision-language benchmarks. --- ### Foveated Reasoning: Stateful, Action-based Visual Focusing for Vision-Language Models Authors: Juhong Min, Lazar Valkov, Vitali Petsiuk, Hossein Souri, Deen Dayal Mohan Published: 2026-04-22 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.21079v1 Vision-language models benefit from high-resolution images, but the increase in visual-token count incurs high compute overhead. Humans resolve this tension via foveation: a coarse view guides "where to look", while selectively acquired high-acuity evidence refines "what to think". We introduce Foveated Reasoner, an autoregressive vision-language framework that unifies foveation and reasoning within a single decoding trajectory. Starting from a low-resolution view, the model triggers foveation only when needed, retrieves high-resolution evidence from selected regions, and injects it back into the same decoding trajectory. We train the method with a two-stage pipeline: coldstart supervision to bootstrap foveation behavior, followed by reinforcement learning to jointly improve evidence acquisition and task accuracy while discouraging trivial "see-everything" solutions. Experiments show that the method learns effective foveation policies and achieves stronger accuracy under tight visual-token budgets across multiple vision-language benchmarks. --- ### Serialisation Strategy Matters: How FHIR Data Format Affects LLM Medication Reconciliation Authors: Sanjoy Pator Published: 2026-04-22 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.21076v1 Medication reconciliation at clinical handoffs is a high-stakes, error-prone process. Large language models are increasingly proposed to assist with this task using FHIR-structured patient records, but a fundamental and largely unstudied variable is how the FHIR data is serialised before being passed to the model. We present the first systematic comparison of four FHIR serialisation strategies (Raw JSON, Markdown Table, Clinical Narrative, and Chronological Timeline) across five open-weight models (Phi-3.5-mini, Mistral-7B, BioMistral-7B, Llama-3.1-8B, Llama-3.3-70B) on a controlled benchmark of 200 synthetic patients, totalling 4,000 inference runs. We find that serialisation strategy has a large, statistically significant effect on performance for models up to 8B parameters: Clinical Narrative outperforms Raw JSON by up to 19 F1 points for Mistral-7B (r = 0.617, p < 10^{-10}). This advantage reverses at 70B, where Raw JSON achieves the best mean F1 of 0.9956. In all 20 model and strategy combinations, mean precision exceeds mean recall: omission is the dominant failure mode, with models more often missing an active medication than fabricating one, which changes how clinical safety auditing priorities should be set. Smaller models plateau at roughly 7-10 concurrent active medications, leaving polypharmacy patients, the patients most at risk from reconciliation errors, systematically underserved. BioMistral-7B, a domain-pretrained model without instruction tuning, produces zero usable output in all conditions, showing that domain pretraining alone is not sufficient for structured extraction. These results offer practical, evidence-based format recommendations for clinical LLM deployment: Clinical Narrative for models up to 8B, Raw JSON for 70B and above. The complete pipeline is reproducible on open-source tools running on an AWS g6e.xlarge instance (NVIDIA L40S, 48 GB VRAM). --- ### Generative Discovery of Magnetic Insulators under Competing Physical Constraints Authors: Qiulin Zeng, Tahiya Chowdhury, Md Shafayat Hossain Published: 2026-04-22 Categories: cond-mat.mtrl-sci, cs.AI arXiv: https://arxiv.org/abs/2604.21073v1 Discovering materials that must simultaneously satisfy multiple competing constraints remains a central challenge in computational materials design, particularly in data-scarce regimes where conventional data-driven approaches are least effective. Magnetic insulators represent a stringent example: the electronic conditions that favor magnetic order often also promote metallicity, while insulating behavior suppresses the interactions that stabilize magnetism. As a result, experimentally viable magnetic insulators are rare and difficult to identify through conventional screening. Here, we introduce MagMatLLM, a constraint-guided generative discovery framework that integrates language-model-based crystal generation with evolutionary selection, surrogate screening, and first-principles validation to target simultaneous stability, magnetism, and insulating behavior. Unlike stability-first approaches, the framework enforces functional constraints during generation and selection, steering the search toward sparsely populated regions of materials space defined by competing physical requirements. Using this workflow, we identify twelve previously unreported candidate magnetic insulators, including Tm$_4$Co$_2$Cr$_2$O$_{12}$ and Cr$_4$Nb$_2$O$_{12}$. Of these, ten are dynamically stable by phonon analysis and exhibit finite band gaps and nonzero magnetic moments in spin-polarized density functional theory calculations. Beyond the specific compounds identified here, this work establishes a general constraint-guided paradigm for multi-objective materials discovery in sparse chemical spaces and provides a transferable strategy for the design of quantum materials under competing physical constraints. --- ### DWTSumm: Discrete Wavelet Transform for Document Summarization Authors: Rana Salama, Abdou Youssef, Mona Diab Published: 2026-04-22 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.21070v1 Summarizing long, domain-specific documents with large language models (LLMs) remains challenging due to context limitations, information loss, and hallucinations, particularly in clinical and legal settings. We propose a Discrete Wavelet Transform (DWT)-based multi-resolution framework that treats text as a semantic signal and decomposes it into global (approximation) and local (detail) components. Applied to sentence- or word-level embeddings, DWT yields compact representations that preserve overall structure and critical domain-specific details, which are used directly as summaries or to guide LLM generation. Experiments on clinical and legal benchmarks demonstrate comparable ROUGE-L scores. Compared to a GPT-4o baseline, the DWT based summarization consistently improve semantic similarity and grounding, achieving gains of over 2% in BERTScore, more than 4\% in Semantic Fidelity, factual consistency in legal tasks, and large METEOR improvements indicative of preserved domain-specific semantics. Across multiple embedding models, Fidelity reaches up to 97%, suggesting that DWT acts as a semantic denoising mechanism that reduces hallucinations and strengthens factual grounding. Overall, DWT provides a lightweight, generalizable method for reliable long-document and domain-specific summarization with LLMs. --- ### Expanding the extreme-k dielectric materials space through physics-validated generative reasoning Authors: Hossain Hridoy, Tahiya Chowdhury, Md Shafayat Hossain Published: 2026-04-22 Categories: cond-mat.mtrl-sci, cond-mat.mes-hall, cs.AI arXiv: https://arxiv.org/abs/2604.21068v1 The most technologically consequential materials are often the rarest: they occupy narrow regions of chemical space, obey competing physical constraints, and appear only sparsely in existing databases. High-kappa dielectrics, high-Tc superconductors, and ferromagnetic insulators are to name a few. This scarcity fundamentally limits today's data-driven materials discovery, where machine-learning models excel at interpolation but struggle to generate genuinely new candidates. Here, we introduce DielecMIND, an artificial intelligence framework that reframes materials discovery as a reasoning-driven exploration instead of a database-screening problem. Using high-kappa dielectrics as a data-scarce and technologically stringent test case, DielecMIND combines large-language-model hypothesis generation for the first time with physics validated first-principles calculation to navigate chemical space beyond known compounds. Prior to our work, only 14 experimentally or computationally validated materials with kappa > 150 were known. Our framework discovers and validates 5 new such compounds, expanding this rare-materials class by a remarkable = 35% in a single study. Among them, we find that Ba2TiHfO6 exhibits a dielectric constant of 637, minimal loss at low optical frequencies, and stability up to 800 K. Beyond dielectrics, this work demonstrates a new paradigm for artificial-intelligence-guided discovery: one that generates a small number of physically grounded, experimentally plausible candidates yet measurably expands sparsely populated functional materials spaces. Thus, DielecMIND points toward a general strategy for discovering rare, high-impact functional materials where data scarcity has long constrained progress. --- ### Optimizing Diffusion Priors in Image Reconstruction from a Single Observation Authors: Frederic Wang, Katherine L. Bouman Published: 2026-04-22 Categories: cs.CV, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2604.21066v2 While diffusion priors generate high-quality posterior samples across many inverse problems, they are often trained on limited training sets or purely simulated data, thus inheriting the errors and biases of these underlying sources. Current approaches to finetuning diffusion models rely on a large number of observations with varying forward operators, which can be difficult to collect for many applications, and thus lead to overfitting when the measurement set is small. We propose a method for tuning a prior from only a single observation by combining existing diffusion priors into a single product-of-experts prior and identifying the exponents that maximize the Bayesian evidence. We validate our method on real-world inverse problems, including black hole imaging, where the true prior is unknown a priori, and image deblurring with text-conditioned priors. We find that the evidence is often maximized by priors that extend beyond those trained on a single dataset. By generalizing the prior through exponent weighting, our approach enables posterior sampling from both tempered and combined diffusion models, yielding more flexible priors that improve the trustworthiness of the resulting posterior image distribution. --- ### InVitroVision: a Multi-Modal AI Model for Automated Description of Embryo Development using Natural Language Authors: Nicklas Neu, Thomas Ebner, Jasmin Primus, Raphael Zefferer, Bernhard Schenkenfelder, Mathias Brunbauer, Florian Kromp Published: 2026-04-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.21061v1 The application of artificial intelligence (AI) in IVF has shown promise in improving consistency and standardization of decisions, but often relies on annotated data and does not make use of the multimodal nature of IVF data. We investigated whether foundational vision-language models can be fine-tuned to predict natural language descriptions of embryo morphology and development. Using a publicly available embryo time-lapse dataset, we fine-tuned PaliGemma-2, a multi-modal vision-language model, with only 1,000 images and corresponding captions, describing embryo morphology, embryonic cell cycle and developmental stage. Our results show that the fine-tuned model, InVitroVision, outperformed a commercial model, ChatGPT 5.2, and base models in overall metrics, with performance improving with larger training datasets. This study demonstrates the potential of foundational vision-language models to generalize to IVF tasks with limited data, enabling the prediction of natural language descriptions of embryo morphology and development. This approach may facilitate the use of large language models to retrieve information and scientific evidence from relevant publications and guidelines, and has implications for few-shot adaptation to multiple downstream tasks in IVF. --- ### Clinically-Informed Modeling for Pediatric Brain Tumor Classification from Whole-Slide Histopathology Images Authors: Joakim Nguyen, Jian Yu, Jinrui Fang, Nicholas Konz, Tianlong Chen, Sanjay Krishnan, Chandra Krishnan, Ying Ding, Hairong Wang, Ankita Shukla Published: 2026-04-22 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.21060v2 Accurate diagnosis of pediatric brain tumors, starting with histopathology, presents unique challenges for deep learning, including severe data scarcity, class imbalance, and fine-grained morphologic overlap across diagnostically distinct subtypes. While pathology foundation models have advanced patch-level representation learning, their effective adaptation to weakly supervised pediatric brain tumor classification under limited data remains underexplored. In this work, we introduce an expert-guided contrastive fine-tuning framework for pediatric brain tumor diagnosis from whole-slide images (WSI). Our approach integrates contrastive learning into slide-level multiple instance learning (MIL) to explicitly regularize the geometry of slide-level representations during downstream fine-tuning. We propose both a general supervised contrastive setting and an expert-guided variant that incorporates clinically informed hard negatives targeting diagnostically confusable subtypes. Through comprehensive experiments on pediatric brain tumor WSI classification under realistic low-sample and class-imbalanced conditions, we demonstrate that contrastive fine-tuning yields measurable improvements in fine-grained diagnostic distinctions. Our experimental analyses reveal complementary strengths across different contrastive strategies, with expert-guided hard negatives promoting more compact intra-class representations and improved inter-class separation. This work highlights the importance of explicitly shaping slide-level representations for robust fine-grained classification in data-scarce pediatric pathology settings. --- ### TRACES: Tagging Reasoning Steps for Adaptive Cost-Efficient Early-Stopping Authors: Yannis Belkhiter, Seshu Tirupathi, Giulio Zizzo, John D. Kelleher Published: 2026-04-22 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.21057v2 The field of Language Reasoning Models (LRMs) has been very active over the past few years with advances in training and inference techniques enabling LRMs to reason longer, and more accurately. However, a growing body of studies show that LRMs are still inefficient, over-generating verification and reflection steps. Additionally, the high-level role of each reasoning step and how different step types contribute to the generation of correct answers, is largely underexplored. To address this challenge, we introduce TRACES (Tagging of the Reasoning steps enabling Adaptive Cost-Efficient early-Stopping), a lightweight framework that tags reasoning steps in real-time, and enable adaptive, cost-efficient early stopping of large-language-model inferences. By monitoring reasoning behaviors during inferences, we find that LRMs tend to shift their reasoning behavior after reaching a correct answer. We demonstrate that the monitoring of the specific type of steps can produce effective interpretable early stopping criteria. Evaluated on three mathematical reasoning benchmarks (MATH500, GSM8K, AIME) and two knowledge and reasoning benchmarks (MMLU and GPQA), TRACES achieve 20 to 50% token reduction while maintaining comparable accuracy to standard generation. On harder tasks (BeyondAIME, IMO AnswerBench), the accuracy-efficiency trade-off is steeper, requiring a more conservative early-stopping threshold. This work offers a novel way to study and control the generation behavior of LRMs. --- ### TRACES: Tagging Reasoning Steps for Adaptive Cost-Efficient Early-Stopping Authors: Yannis Belkhiter, Seshu Tirupathi, Giulio Zizzo, John D. Kelleher Published: 2026-04-22 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.21057v1 The field of Language Reasoning Models (LRMs) has been very active over the past few years with advances in training and inference techniques enabling LRMs to reason longer, and more accurately. However, a growing body of studies show that LRMs are still inefficient, over-generating verification and reflection steps. Additionally, the high-level role of each reasoning step and how different step types contribute to the generation of correct answers, is largely underexplored. To address this challenge, we introduce TRACES (Tagging of the Reasoning steps enabling Adaptive Cost-Efficient early-Stopping), a lightweight framework that tags reasoning steps in real-time, and enable adaptive, cost-efficient early stopping of large-language-model inferences. Building on this framework we monitor reasoning behaviors during inferences, and we find that LRMs tend to shift their reasoning behavior after reaching a correct answer. We demonstrate that the monitoring of the specific type of steps can produce effective interpretable early stopping criteria. We evaluate the TRACES framework on three mathematical reasoning benchmarks, namely, MATH500, GSM8K, AIME and two knowledge and reasoning benchmarks, MMLU and GPQA respectively. We achieve 20 to 50% token reduction while maintaining comparable accuracy to standard generation. --- ### Neuro-Symbolic Manipulation Understanding with Enriched Semantic Event Chains Authors: Fatemeh Ziaeetabar Published: 2026-04-22 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2604.21053v1 Robotic systems operating in human environments must reason about how object interactions evolve over time, which actions are currently being performed, and what manipulation step is likely to follow. Classical enriched Semantic Event Chains (eSECs) provide an interpretable relational description of manipulation, but remain primarily descriptive and do not directly support uncertainty-aware decision making. In this paper, we propose eSEC-LAM, a neuro-symbolic framework that transforms eSECs into an explicit event-level symbolic state for manipulation understanding. The proposed formulation augments classical eSECs with confidence-aware predicates, functional object roles, affordance priors, primitive-level abstraction, and saliency-guided explanation cues. These enriched symbolic states are derived from a foundation-model-based perception front-end through deterministic predicate extraction, while current-action inference and next-primitive prediction are performed using lightweight symbolic reasoning over primitive pre- and post-conditions. We evaluate the proposed framework on EPIC-KITCHENS-100, EPIC-KITCHENS VISOR, and Assembly101 across action recognition, next-primitive prediction, robustness to perception noise, and explanation consistency. Experimental results show that eSEC-LAM achieves competitive action recognition, substantially improves next-primitive prediction, remains more robust under degraded perceptual conditions than both classical symbolic and end-to-end video baselines, and provides temporally consistent explanation traces grounded in explicit relational evidence. These findings demonstrate that enriched Semantic Event Chains can serve not only as interpretable descriptors of manipulation, but also as effective internal states for neuro-symbolic action reasoning. --- ### StyleVAR: Controllable Image Style Transfer via Visual Autoregressive Modeling Authors: Liqi Jing, Dingming Zhang, Peinian Li, Lichen Zhu, Yang Xu, Hanyu Xing Published: 2026-04-22 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.21052v2 We build on the Visual Autoregressive Modeling (VAR) framework and formulate style transfer as conditional discrete sequence modeling in a learned latent space. Images are decomposed into multi-scale representations and tokenized into discrete codes by a VQ-VAE; a transformer then autoregressively models the distribution of target tokens conditioned on style and content tokens. To inject style and content information, we introduce a blended cross-attention mechanism in which the evolving target representation attends to its own history, while style and content features act as queries that decide which aspects of this history to emphasize. A scale-dependent blending coefficient controls the relative influence of style and content at each stage, encouraging the synthesized representation to align with both the content structure and the style texture without breaking the autoregressive continuity of VAR. We train StyleVAR in two stages from a pretrained VAR checkpoint: supervised fine-tuning on a large triplet dataset of content--style--target images, followed by reinforcement fine-tuning with Group Relative Policy Optimization (GRPO) against a DreamSim-based perceptual reward, with per-action normalization weighting to rebalance credit across VAR's multi-scale hierarchy. Across three benchmarks spanning in-, near-, and out-of-distribution regimes, StyleVAR consistently outperforms an AdaIN baseline on Style Loss, Content Loss, LPIPS, SSIM, DreamSim, and CLIP similarity, and the GRPO stage yields further gains over the SFT checkpoint, most notably on the reward-aligned perceptual metrics. Qualitatively, the method transfers texture while maintaining semantic structure, especially for landscapes and architectural scenes, while a generalization gap on internet images and difficulty with human faces highlight the need for better content diversity and stronger structural priors. --- ### JEPAMatch: Geometric Representation Shaping for Semi-Supervised Learning Authors: Ali Aghababaei-Harandi, Aude Sportisse, Massih-Reza Amini Published: 2026-04-22 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.21046v3 Semi-supervised learning has emerged as a powerful paradigm for leveraging large amounts of unlabeled data to improve the performance of machine learning models when labeled data are scarce. Among existing approaches, methods derived from FixMatch have achieved state-of-the-art results in image classification by combining weak and strong data augmentations with confidence-based pseudo-labeling. Despite their strong empirical performance, these methods typically struggle with two critical bottlenecks: majority classes tend to dominate the learning process, which is amplified by incorrect pseudo-labels, leading to biased models. Furthermore, noisy early pseudo-labels prevent the model from forming clear decision boundaries, requiring prolonged training to learn informative representation. In this paper, we introduce a paradigm shift from conventional logical output threshold base, toward an explicit shaping of geometric representations. Our approach is inspired by the recently proposed Latent-Euclidean Joint-Embedding Predictive Architectures (LeJEPA), a theoretically grounded framework asserting that meaningful representations should exhibit an isotropic Gaussian structure in latent space. Building on this principle, we propose a new training objective that combines the classical semi-supervised loss used in FlexMatch, an adaptive extension of FixMatch, with a latent-space regularization term derived from LeJEPA. Our proposed approach, encourages well-structured representations while preserving the advantages of pseudo-labeling strategies. Through extensive experiments on CIFAR-100, STL-10 and Tiny-ImageNet, we demonstrate that the proposed method consistently outperforms existing baselines. In addition, our method significantly accelerates the convergence, drastically reducing the overall computational cost compared to standard FixMatch-based pipelines. --- ### JEPAMatch: Geometric Representation Shaping for Semi-Supervised Learning Authors: Ali Aghababaei-Harandi, Aude Sportisse, Massih-Reza Amini Published: 2026-04-22 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.21046v2 Semi-supervised learning has emerged as a powerful paradigm for leveraging large amounts of unlabeled data to improve the performance of machine learning models when labeled data are scarce. Among existing approaches, methods derived from FixMatch have achieved state-of-the-art results in image classification by combining weak and strong data augmentations with confidence-based pseudo-labeling. Despite their strong empirical performance, these methods typically struggle with two critical bottlenecks: majority classes tend to dominate the learning process, which is amplified by incorrect pseudo-labels, leading to biased models. Furthermore, noisy early pseudo-labels prevent the model from forming clear decision boundaries, requiring prolonged training to learn informative representation. In this paper, we introduce a paradigm shift from conventional logical output threshold base, toward an explicit shaping of geometric representations. Our approach is inspired by the recently proposed Latent-Euclidean Joint-Embedding Predictive Architectures (LeJEPA), a theoretically grounded framework asserting that meaningful representations should exhibit an isotropic Gaussian structure in latent space. Building on this principle, we propose a new training objective that combines the classical semi-supervised loss used in FlexMatch, an adaptive extension of FixMatch, with a latent-space regularization term derived from LeJEPA. Our proposed approach, encourages well-structured representations while preserving the advantages of pseudo-labeling strategies. Through extensive experiments on CIFAR-100, STL-10 and Tiny-ImageNet, we demonstrate that the proposed method consistently outperforms existing baselines. In addition, our method significantly accelerates the convergence, drastically reducing the overall computational cost compared to standard FixMatch-based pipelines. --- ### Hierarchical Policy Optimization for Simultaneous Translation of Unbounded Speech Authors: Siqi Ouyang, Shuoyang Ding, Oleksii Hrinchuk, Vitaly Lavrukhin, Brian Yan, Boris Ginsburg, Lei Li Published: 2026-04-22 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.21045v1 Simultaneous speech translation (SST) generates translations while receiving partial speech input. Recent advances show that large language models (LLMs) can substantially improve SST quality, but at the cost of high computational overhead. To reduce this cost, prior work reformulates SST as a multi-turn dialogue task, enabling full reuse of the LLM's key-value (KV) cache and eliminating redundant feature recomputation. However, this approach relies on supervised fine-tuning (SFT) data in dialogue form, for which few human annotations exist, and existing synthesis methods cannot guarantee data quality. In this work, we propose a Hierarchical Policy Optimization (HPO) approach that post-train models trained on imperfect SFT data. We introduce a hierarchical reward that balances translation quality and latency objectives. Experiments on English to Chinese/German/Japanese demonstrate improvements of over +7 COMET score and +1.25 MetricX score at a latency of 1.5 seconds. Comprehensive ablation studies further validate the effectiveness of different quality rewards, hierarchical reward formulations, and segmentation strategies. Code can be found here https://github.com/owaski/HPO --- ### Active Data Authors: Richard Arthur, Virginia DiDomizio, Louis Hoebel Published: 2026-04-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.21044v1 In some complex domains, certain problem-specific decompositions can provide advantages over monolithic designs by enabling comprehension and specification of the design. In this paper we present an intuitive and tractable approach to reasoning over large and complex data sets. Our approach is based on Active Data, i.e., data as atomic objects that actively interact with environments. We describe our intuition about how this bottom-up approach improves designs confronting computational and conceptual complexity. We describe an implementation of the base Active Data concepts within the air traffic flow management domain and discuss performance for this implementation. ---