# Research Papers — 2026-04 · bundle page 23 of 233

> Full-text markdown of papers #1101–1150 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/23
[← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/22.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-04/page/24.md)

### From World-Gen to Quest-Line: A Dependency-Driven Prompt Pipeline for Coherent RPG Generation

Authors: Dominik Borawski, Marta Szulc, Robert Chudy, Małgorzata Giedrowicz, Piotr Mironowicz
Published: 2026-04-28
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2604.25482v1

Large Language Models (LLMs) have shown strong potential for narrative generation, but their use in complex, multi-layered role-playing game (RPG) worlds is still limited by issues of coherence, controllability, and structural consistency. This paper explores a dependency-aware, multi-stage prompt pipeline for procedural RPG content generation that models narrative dependencies through structured intermediate representations. The approach decomposes generation into sequential stages: world building, non-player character creation, player character creation, campaign-level quest planning, and quest expansion. Each stage conditions on structured JSON outputs from previous stages. By enforcing schemas and explicit data flow, the pipeline reduces narrative drift, limits hallucinations, and supports scalable creation of interconnected narrative elements. The system is evaluated qualitatively through human-centered analysis across multiple independent runs. Outputs are assessed using criteria such as structural completeness, internal consistency, narrative coherence, diversity, and actionability. Results show that the pipeline consistently generates logically sound and structurally valid RPG content, without quality degradation as complexity increases. Separating high-level campaign planning from detailed quest expansion improves both global structure and local storytelling. These findings suggest that dependency-aware prompt pipelines with structured intermediate representations are an effective design pattern for LLM-based procedural content generation. This approach may also generalize to other domains requiring sequential reasoning over evolving contextual states.

---

### Emergent Self-Attention from Astrocyte-Gated Associative Memory Dynamics

Authors: Arnau Vivet, Alex Arenas
Published: 2026-04-28
Categories: physics.data-an, cs.LG, nlin.AO, physics.soc-ph
arXiv: https://arxiv.org/abs/2604.25481v1

We introduce a Hopfield-type associative memory in which effective connectivity is multiplicatively modulated by astrocytic gains evolving under an entropy-regularized replicator equation. The coupled neuron-astrocyte dynamics admit a Lyapunov function, ensuring global convergence. At fixed points, astrocytic gains implement a softmax-normalized allocation over pattern similarity scores, yielding a mechanistic realization of self-attention as emergent routing on the gain simplex. In regimes of high memory load and interference, the model significantly improves retrieval accuracy relative to classical Hopfield dynamics and recent neuron-astrocyte baselines. These results establish a dynamical systems framework linking glial modulation, competitive resource allocation, and attention-like computation.

---

### DDA-Thinker: Decoupled Dual-Atomic Reinforcement Learning for Reasoning-Driven Image Editing

Authors: Hanqing Yang, Qiang Zhou, Yongchao Du, Sashuai Zhou, Zhibin Wang, Jun Song, Tiezheng Ge, Cheng Yu, Bo Zheng
Published: 2026-04-28
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2604.25477v1

Recent image editing models have achieved strong visual fidelity but often struggle with tasks requiring complex reasoning. To investigate and enhance the reasoning-grounded planning for image editing, we propose DDA-Thinker, a Thinker-centric framework designed for the independent optimization of a planning module (Thinker) over a fixed generative model (Editor). This decoupled Thinker-centric paradigm facilitates a controlled analysis of the planning module and makes its contribution under a fixed Editor easier to assess. To effectively guide this Thinker, we introduce a dual-atomic reinforcement learning framework. This framework decomposes feedback into two distinct atomic rewards implemented through verifiable checklists: a cognitive-atomic reward to directly assess the quality of the Thinker's executable plan, which serves as the actionable outcome of the Thinker's reasoning, and a visual-atomic reward to assess the final image quality. To improve checklist quality, our checklist synthesis is grounded not only in the source image and user instruction but also in a rational reference description of the ideal post-edit scene. To support this training, we further develop a two-stage data curation pipeline that first synthesizes a diverse and reasoning-focused dataset, then applies difficulty-aware refinement to curate an effective training curriculum for reinforcement learning. Extensive experiments on reasoning-driven image editing benchmarks, including RISE-Bench and KRIS-Bench, demonstrate that our approach substantially improves overall performance. Our method enables a community model to achieve results competitive with strong proprietary models, highlighting the practical potential of Thinker-centric optimization under a fixed-editor setting.

---

### PSP: An Interpretable Per-Dimension Accent Benchmark for Indic Text-to-Speech

Authors: Venkata Pushpak Teja Menta
Published: 2026-04-28
Categories: cs.SD, cs.CL
arXiv: https://arxiv.org/abs/2604.25476v1

Standard text-to-speech (TTS) evaluation measures intelligibility (WER, CER) and overall naturalness (MOS, UTMOS) but does not quantify accent. A synthesiser may score well on all four yet sound non-native on features that are phonemic in the target language. For Indic languages, these features include retroflex articulation, aspiration, vowel length, and the Tamil retroflex approximant (letter zha). We present PSP, the Phoneme Substitution Profile, an interpretable, per-phonological-dimension accent benchmark for Indic TTS. PSP decomposes accent into six complementary dimensions: retroflex collapse rate (RR), aspiration fidelity (AF), vowel-length fidelity (LF), Tamil-zha fidelity (ZF), Frechet Audio Distance (FAD), and prosodic signature divergence (PSD). The first four are measured via forced alignment plus native-speaker-centroid acoustic probes over Wav2Vec2-XLS-R layer-9 embeddings; the latter two are corpus-level distributional distances. In this v1 we benchmark four commercial and open-source systems (ElevenLabs v3, Cartesia Sonic-3, Sarvam Bulbul, Indic Parler-TTS) on Hindi, Telugu, and Tamil pilot sets, with a fifth system (Praxy Voice) included on all three languages, plus an R5->R6 case study on Telugu. Three findings: (i) retroflex collapse grows monotonically with phonological difficulty Hindi < Telugu < Tamil (~1%, ~40%, ~68%); (ii) PSP ordering diverges from WER ordering -- commercial WER-leaders do not uniformly lead on retroflex or prosodic fidelity; (iii) no single system is Pareto-optimal across all six dimensions. We release native reference centroids (500 clips per language), 1000-clip embeddings for FAD, 500-clip prosodic feature matrices for PSD, 300-utterance golden sets per language, scoring code under MIT, and centroids under CC-BY. Formal MOS-correlation is deferred to v2; v1 reports five internal-consistency signals plus a native-audio sanity check.

---

### SciEval: A Benchmark for Automatic Evaluation of K-12 Science Instructional Materials

Authors: Zhaohui Li, Peng He, Zhiyuan Chen, Honglu Liu, Zeyuan Wang, Tingting Li, Jinjun Xiong
Published: 2026-04-28
Categories: cs.AI
arXiv: https://arxiv.org/abs/2604.25472v1

The need to evaluate instructional materials for K-12 science education has become increasingly important, as more educators use generative AI to create instructional materials. However, the review of instructional materials is time-consuming, expertise-intensive, and difficult to scale, motivating interest in automated evaluation approaches. While large language models (LLMs) have shown strong performance on general evaluation tasks, their performance and reliability on instructional materials remain unclear. To address this gap, we formulate Automatic Instructional Materials Evaluation (AIME) as a generative AI task that predicts scores and evidence using the rubric designed by the educator. We create a benchmark dataset and develop baseline models for AIME. First, we curate the first AIME dataset, SciEval, consisting of instructional materials annotated with pedagogy-aligned evaluation scores and evidence-based rationales. Expert annotations achieve high inter-rater reliability, resulting in a dataset of 273 lesson-level instructional materials evaluated across 13 criteria (N=3549) using the EQuIP rubric. Second, we test mainstream LLMs (GPT, Gemini, Llama, and Qwen) on SciEval and find that none achieve strong performance. Then we fine-tune Qwen3 on SciEval. Results on a held-out test set show that domain-aligned fine-tuning can achieve up to 11 percent performance gains, highlighting the importance of domain-specific fine-tuning for AIME and facilitating the use of LLMs in other educational tasks.

---

### Subspace Optimization for Efficient Federated Learning under Heterogeneous Data

Authors: Shuchen Zhu, Zhengyang Huang, Yuqi Xu, Peijin Li
Published: 2026-04-28
Categories: cs.LG, math.OC
arXiv: https://arxiv.org/abs/2604.25467v1

Federated learning increasingly operates in a large-model regime where communication, memory, and computation are all scarce. Typically, non-IID client data induce drift that degrades the stability and performance of local training. Existing remedies such as SCAFFOLD introduce heterogeneity-correction mechanisms to address this challenge, but they incur substantial extra communication and memory overhead. This paper proposes a subspace optimization method for federated learning (SSF), which performs heterogeneity-corrected optimization in a low-dimensional subspace using only projected quantities, while preserving full-dimensional control information through a backfill-style update that retains residual components whenever the active subspace changes. Under standard smoothness and bounded-variance assumptions, SSF attains a non-asymptotic rate of order $\widetilde{\mathcal{O}}(1/T+1/\sqrt{NKT})$. Experiments show favorable accuracy--efficiency trade-offs under heterogeneous data.

---

### Generalizable Human Gaussian Splatting via Multi-view Semantic Consistency

Authors: Jingi Kim, Wonjun Kim
Published: 2026-04-28
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.25466v1

Recently, generalizable human Gaussian splatting from sparse-view inputs has been actively studied for the photorealistic human rendering. Most existing methods rely on explicit geometric constraints or predefined structural representations to accurately position 3D Gaussians. Although these approaches have shown the remarkable progress in this field, they still suffer from inconsistent feature representations across multi-view inputs due to complex articulations of the human body and limited overlaps between different views. To address this problem, we propose a novel method to accurately localize 3D Gaussians and ultimately improve the quality of human rendering. The key idea is to unproject latent embeddings encoded from each viewpoint into a shared 3D space through predicted depth maps and recalibrate them belonging to the same body part based on cross-view attention. This helps the model resolve the spatial ambiguity occurring in highly textured regions as well as occluded body parts, thus leading to the accurate localization of 3D Gaussians. Experimental results on benchmark datasets show that the proposed method efficiently improves the performance of generalizable human Gaussian splatting from sparse-view inputs.

---

### Image Compression with Bubble-Aware Frame Rate Adaptation for Energy-Efficient Video Capsule Endoscopy

Authors: Oliver Bause, Jörg Gamerdinger, Julia Werner, Oliver Bringmann
Published: 2026-04-28
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.25464v2

Video Capsule Endoscopy (VCE) is a promising method for improving the medical examination of the small intestine in the gastrointestinal tract. A key challenge is their limited size, resulting in a short battery lifetime which conflicts with high energy consumption for image capturing and transmission to an on-body device. Thus, we propose an image compression pipeline that substantially reduces the transmitted data while preserving diagnostic image quality. Furthermore, we exploit characteristics of the compression process to identify frames with low diagnostic value mainly caused by bubbles, without requiring additional image analysis. For low-visibility frames, a dynamic bubble-aware frame rate adaptation strategy reduces image acquisition and transmission during these phases while preserving sensitivity to potential anomalies. The proposed compression and frame rate adaptation are evaluated on a RISC-V platform using the Kvasir-Capsule and Galar datasets. The compression method achieves a compression ratio of 5.748 (82.6%) at a peak signal-to-noise ratio of 40.3 dB, indicating negligible loss of visual quality. The compression accomplished a mean energy reduction of the whole system by 20.58%. Additionally, the proposed bubble-aware frame rate adaptation reduced the energy consumption by up to 40%. These results demonstrate the potential of our method to increase the applicability of VCE.

---

### Benchmarking Stopping Criteria for Evolutionary Multi-objective Optimization

Authors: Kenji Kitamura, Ryoji Tanabe
Published: 2026-04-28
Categories: cs.NE
arXiv: https://arxiv.org/abs/2604.25458v1

Stopping criteria automatically determine when to stop an evolutionary algorithm, so as not to waste function evaluations on a stagnant population. Although stopping criteria play an important role in real-world applications, they have attracted little attention in the evolutionary multi-objective optimization (EMO) community. In fact, new stopping criteria for EMO have been rarely developed in recent years. One reason for the stagnation in developing stopping criteria for EMO is a lack of effective benchmarking methodologies. To address this issue, this paper proposes (i) a performance measure of stopping criteria for EMO and (ii) a file-based benchmarking approach. This paper also proposes (iii) a data representation method that effectively stores population states in text files. (i) The proposed measure represents the performance of stopping criteria as a single scalar value, making comparison easy. (ii) The proposed file-based approach not only simplifies the benchmarking process but also facilitates reproducibility. (iii) The proposed data representation method addresses the issue of file size in (ii). We demonstrate the effectiveness of our three contributions (i)--(iii) by benchmarking five representative stopping criteria for EMO.

---

### GramSR: Visual Feature Conditioning for Diffusion-Based Super-Resolution

Authors: Fabio D'Oronzio, Federico Putamorsi, Leonardo Zini, Marcella Cornia, Lorenzo Baraldi
Published: 2026-04-28
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.25457v1

Despite recent advances, single-image super-resolution (SR) remains challenging, especially in real-world scenarios with complex degradations. Diffusion-based SR methods, particularly those built on Stable Diffusion, leverage strong generative priors but commonly rely on text conditioning derived from semantic captioning. Such textual descriptions provide only high-level semantics and lack the spatially aligned visual information required for faithful restoration, leading to a representation gap between abstract semantics and spatially aligned visual details. To address this limitation, we propose GramSR, a one-step diffusion-based SR framework that replaces text conditioning with dense visual features extracted from the low-resolution input using a pre-trained DINOv3 encoder. GramSR adopts a three-stage LoRA architecture, where pixel-level, semantic-level, and texture-level LoRA modules are trained sequentially. The pixel-level module focuses on degradation removal using $\ell_2$ loss, the semantic-level module enhances perceptual details via LPIPS and CSD losses, and the texture-level module enforces feature correlation consistency through a Gram matrix loss computed from DINOv3 features. At inference, independent guidance scales enable flexible control over degradation removal, semantic enhancement, and texture preservation. Extensive experiments on standard SR benchmarks demonstrate that GramSR consistently outperforms existing one-step diffusion-based methods, achieving superior structural fidelity and texture realism. The code for this work is available at: https://github.com/aimagelab/GramSR.

---

### An Investigation of Linguistic Biases in LLM-Based Recommendations

Authors: Nitin Venkateswaran, Jason Ang, Deep Adhikari, Tarun Krishna Dasari
Published: 2026-04-28
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2604.25456v1

We investigate linguistic biases in LLM-based restaurant and product recommendations given prompts varying across Southern American English (AE), Indian English (IE), and Code-Switched Hindi-English dialects, using the Yelp Open dataset (Yelp Inc., 2023) and Walmart product reviews dataset (PromptCloud,2020). We add lists of restaurant and product names balanced by cuisine type and product category to the prompts given to the LLM, and we zero-shot prompt the LLMs in a cold-start setting to select the top-20 restaurant and product recommendations from these lists for each of the dialect-varied prompts. We prompt LLMs using different list samples across 20 seeds for better generalization, and aggregate per cuisine-type and per category response counts for each seed, question/prompt, and LLM model. We run mixed-effects regression models for each model family and topic (restaurant/product) with the aggregate response counts as the dependent, and conduct likelihood ratio tests for the fixed effects with post-hoc pairwise testing of estimated marginal means differences, to investigate group-level differences in recommendation counts by model size and dialect type. Results show that dialect plays a role in the type of restaurant selected across the models tested with the mistral-small-3.1 model and both the llama-3.1 family models tested showing more sensitivity to Indian English and Code-Switched prompts. In terms of product recommendations, the llama-3.1-70B-model is particularly sensitive to Code-Switched prompts in four out of seven categories, and more beauty and home category recommendations are seen when using the Indian English and Code-Switched prompts for larger and smaller models, respectively. No broad trends are seen in the model-size based differences, with differing recommendations based on model sizes conditioned by the type of dialect.

---

### Generative UI as an Accessibility Bridge: Lessons from C2C E-Commerce

Authors: Bektur Ryskeldiev
Published: 2026-04-28
Categories: cs.HC, cs.AI, cs.CY
arXiv: https://arxiv.org/abs/2604.25455v1

Web accessibility rests on static standards and developer compliance. That model frays in platforms where content is user-generated: photos arrive blurry or off-frame, descriptions skip size and condition, and page structure shifts from listing to listing. Drawing on six studies conducted between 2022 and 2025 with blind, low-vision, and older adult users of customer-to-customer (C2C) marketplaces, I argue that generative UI can produce adapted interfaces at the point of use, addressing barriers that static design cannot anticipate. Three interventions from this program -- HTML regeneration for screen readers, conversational guidance for older sellers, and audio-guided photo framing for blind sellers -- demonstrate how runtime generation can bridge gaps that standards leave open. I outline what these findings imply for HCI practice: generative UI extends beyond the screen, complements rather than replaces ability-based design, and shifts the designer's role from specifying layouts to specifying policies. This is an expanded arXiv version of a position paper accepted at the CHI 2026 workshop "What does Generative UI mean for HCI Practice?"

---

### Benchmarking Logistic Regression, SVM, and LightGBM Against BiLSTM with Attention for Sentiment Analysis on Indonesian Product Reviews

Authors: Razin Hafid Hamdi, Ivana Margareth Hutabarat, Hanna Gresia Sinaga, Luluk Muthoharoh, Ardika Satria, Martin C. T. Manullang
Published: 2026-04-28
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.25452v1

Sentiment analysis of product reviews on e-commerce platforms plays a critical role in automatically understanding customer satisfaction and providing actionable insights for sellers seeking to improve product quality. This paper presents a comprehensive benchmarking study comparing a Machine Learning (ML) approach via the PyCaret AutoML framework against a Deep Learning (DL) approach based on a Bidirectional Long Short-Term Memory (BiLSTM) architecture with an Attention mechanism for binary sentiment classification on Indonesian product reviews. The dataset comprises 19,728 samples balanced equally between positive and negative reviews. For the ML approach, three prominent algorithms were evaluated via 10-fold stratified cross-validation: Logistic Regression (LR), Support Vector Machine (SVM) with a linear kernel, and Light Gradient Boosting Machine (LightGBM). Logistic Regression achieved the best ML performance with an accuracy of 97.26\% and an F1-score of 97.26\%. The BiLSTM with Attention model, evaluated on 3,946 held-out test samples, achieved an accuracy of 97.24\% and an F1-score of 97.24\%. These comparative results demonstrate that traditional ML algorithms with proper preprocessing and feature extraction can compete closely with, and even marginally outperform, more complex sequential DL architectures on high-dimensional datasets, while simultaneously offering greater computational efficiency.

---

### Navigating Global AI Regulation: A Multi-Jurisdictional Retrieval-Augmented Generation System

Authors: Courtney Ford, Ojas Rane, Susan Leavy
Published: 2026-04-28
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.25448v1

Navigating AI regulation across jurisdictions is increasingly difficult for policymakers, legal professionals, and researchers. To address this, we present a multi-jurisdictional Retrieval-Augmented Generation system for global AI regulation. Our corpus includes 242 documents across 68 jurisdictions, ranging from formal legislation like the EU AI Act to unstructured policy documents such as national AI strategies. The system makes three technical contributions: type-specific chunking that preserve legal structure across heterogenous documents; conditional retrieval routing with entity detection and metadata for legal citations; and priority-based re-ranking to boost enacted legislation over policy and secondary sources. Evaluation of 50 queries reveals strong performance across both single-entity and multi-jurisdictional questions, achieving 0.87 average faithfulness and 0.84 average answer relevancy. Single-entity queries achieve 0.86 average faithfulness and 0.92 average answer relevancy, while multi-jurisdictional comparison queries achieve 0.88 average faithfulness and 0.75 average answer relevancy. These findings highlight the effectiveness of domain-specific retrieval strategies for navigating complex, heterogenous regulatory corpora.

---

### One Refiner to Unlock Them All: Inference-Time Reasoning Elicitation via Reinforcement Query Refinement

Authors: Yixiao Zhou, Dongzhou Cheng, zhiliang wu, Yi Yang, Yu Cheng, Hehe Fan
Published: 2026-04-28
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.25444v1

Large Language Models (LLMs) often fail to utilize their latent reasoning capabilities due to a distributional mismatch between ambiguous human inquiries and the structured logic required for machine activation. Existing alignment methods either incur prohibitive $O(N)$ costs by fine-tuning each model individually or rely on static prompts that fail to resolve query-level structural complexity. In this paper, we propose ReQueR (\textbf{Re}inforcement \textbf{Que}ry \textbf{R}efinement), a modular framework that treats reasoning elicitation as an inference-time alignment task. We train a specialized Refiner policy via Reinforcement Learning to rewrite raw queries into explicit logical decompositions, treating frozen LLMs as the environment. Rooted in the classical Zone of Proximal Development from educational psychology, we introduce the Adaptive Solver Hierarchy, a curriculum mechanism that stabilizes training by dynamically aligning environmental difficulty with the Refiner's evolving competence. ReQueR yields consistent absolute gains of 1.7\%--7.2\% across diverse architectures and benchmarks, outperforming strong baselines by 2.1\% on average. Crucially, it provides a promising paradigm for one-to-many inference-time reasoning elicitation, enabling a single Refiner trained on a small set of models to effectively unlock reasoning in diverse unseen models. Code is available at https://github.com/newera-xiao/ReQueR.

---

### Praxy Voice: Voice-Prompt Recovery + BUPS for Commercial-Class Indic TTS from a Frozen Non-Indic Base at Zero Commercial-Training-Data Cost

Authors: Venkata Pushpak Teja Menta
Published: 2026-04-28
Categories: cs.SD, cs.CL, eess.AS
arXiv: https://arxiv.org/abs/2604.25441v1

Commercial TTS systems produce near-native Indic audio, but the best open-source bases (Chatterbox, Indic Parler-TTS, IndicF5) trail them on measured phonological dimensions, and the most widely adopted multilingual base (Chatterbox, 23 languages) does not even tokenise Telugu or Tamil. We ask: what is the minimum intervention that brings such a non-Indic-native base to commercial-class output on Telugu, Tamil, and Hindi, without training a new acoustic decoder and without any commercial TTS training data? We combine three pieces: (1) BUPS, a Brahmic Unified Phoneme Space that deterministically romanises seven Indic scripts to ISO-15919 so Chatterbox's Latin tokeniser can process them; (2) a LoRA adapter on only the text-token predictor (Chatterbox's t3), trained on ~1,220h of licensed Indic audio with a Hindi-proxy language_id; (3) a voice-prompt recovery recipe -- an 8-11s same-language reference clip plus three sampling overrides (exaggeration 0.7, temperature 0.6, min_p 0.1; "Config B") -- that recovers commercial-class acoustic output with no acoustic-decoder training. On Hindi, the LoRA regresses accuracy and we instead use vanilla Chatterbox + Config B, giving a two-branch deployment. Evaluated on 10-utterance pilot sets with the companion PSP benchmark, Praxy Voice matches or slightly leads commercial baselines: 26.7% retroflex collapse on Telugu (vs Sarvam Bulbul 33.3%), 71% Tamil-zha collapse (vs commercial trio's 86%), 0.025 LLM-WER on Hindi (tied with Cartesia Sonic-3). For intra-sentential code-mix we add a third branch (IndicF5 + native-script transliteration) that drops code-mix LLM-WER from 0.80-0.85 to 0.14-0.27 across Hi/Te/Ta. We release R6 LoRA weights (Apache-2.0), inference code and router (MIT), and a Gradio demo.

---

### PI-TTA: Physics-Informed Source-Free Test-Time Adaptation for Robust Human Activity Recognition on Mobile Devices

Authors: Changyu Li, Lu Wang, Ming Lei, Jiashen Liu, Yichen Zhang, Kaishun Wu, Fei Luo
Published: 2026-04-28
Categories: cs.AI
arXiv: https://arxiv.org/abs/2604.25435v1

Source-free test-time adaptation (TTA) is appealing for mobile and wearable sensing because it enables on-device personalization from unlabeled test streams without centralizing private data. However, sensor-based human activity recognition (HAR) poses challenges that are less pronounced in standard vision benchmarks: behavioral inertial streams are temporally correlated and often exhibit within-session shifts caused by sensor rotation, placement change, and sampling-rate drift. Under this streaming non-i.i.d. setting, widely used vision-style TTA objectives can become unstable, leading to overconfident errors, representation collapse, and catastrophic forgetting. We propose PI-TTA, a lightweight source-free adaptation framework that stabilizes online updates through three physics-consistent constraints: gravity consistency, short-horizon temporal continuity, and spectral stability. PI-TTA updates the same small parameter subset as strong source-free baselines and incurs only modest overhead, making it suitable for on-device deployment. Experiments on USCHAD, PAMAP2, and mHealth under long-sequence stress tests and factorized shift protocols show that PI-TTA mitigates the severe degradation observed in confidence-driven baselines and preserves stable adaptation under sustained streaming conditions. It improves long-sequence accuracy by up to 9.13% and reduces physical-violation rates by 27.5%, 24.1%, and 45.4% on USCHAD, PAMAP2, and mHealth, respectively. These results demonstrate that physics-informed adaptation can improve accuracy, stability, and deployment reliability for real-world mobile sensing systems.

---

### SARU: A Shadow-Aware and Removal Unified Framework for Remote Sensing Images with New Benchmarks

Authors: Zi-Yang Bo, Wei Lu, Hongruixuan Chen, Si-Bao Chen, Bin Luo
Published: 2026-04-28
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.25432v2

Shadows are a prevalent problem in remote sensing imagery (RSI), degrading visual quality and severely limiting the performance of downstream tasks like object detection and semantic segmentation. Most prior works treat shadow detection and removal as separate, cascaded tasks, which can lead to cumbersome process and error accumulation. Furthermore, many deep learning methods rely on paired shadow and non-shadow images for training, which are often unavailable in practice. To address these challenges, we propose Shadow-Aware and Removal Unified (SARU) Framework , a cohesive two-stage framework. First, its dual-branch detection module (DBCSF-Net) fuses multi-color space and semantic features to generate high-fidelity shadow masks, effectively distinguishing shadows from dark objects. Then, leveraging these masks, a novel, training-free physical algorithm (N$^2$SGSR) restores illumination by transferring properties from adjacent non-shadow regions within the single input image. To facilitate rigorous evaluation and foster future work, we also introduce two new benchmark datasets: the RSI Shadow Detection (RSISD) dataset and the Single-image Shadow Removal Benchmark (SiSRB). Extensive experiments on the AISD and RSISD datasets demonstrate that SARU achieves SOTA shadow detection performance. For shadow removal, our training-free N$^2$SGSR algorithm attains an average processing speed of approximately $1.3$s, which is over $10$ times faster than the SOTA MAOSD while maintains an SRI value close to 0.9 on both the AISD and SiSRB datasets, a level comparable to the advanced RS-GSSR method. By holistically integrating shadow detection and removal to mitigate error propagation and eliminating the dependency on paired training data, SARU establishes a robust, practical framework for real-world RSI analysis. The code and datasets are publicly available at: https://github.com/AeroVILab-AHU/SARU

---

### A Systematic Post-Train Framework for Video Generation

Authors: Zeyue Xue, Siming Fu, Jie Huang, Shuai Lu, Haoran Li, Yijun Liu, Yuming Li, Xiaoxuan He, Mengzhao Chen, Haoyang Huang, Nan Duan, Ping Luo
Published: 2026-04-28
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.25427v1

While large-scale video diffusion models have demonstrated impressive capabilities in generating high-resolution and semantically rich content, a significant gap remains between their pretraining performance and real-world deployment requirements due to critical issues such as prompt sensitivity, temporal inconsistency, and prohibitive inference costs. To bridge this gap, we propose a comprehensive post-training framework that systematically aligns pretrained models with user intentions through four synergistic stages: we first employ Supervised Fine-Tuning (SFT) to transform the base model into a stable instruction-following policy, followed by a Reinforcement Learning from Human Feedback (RLHF) stage that utilizes a novel Group Relative Policy Optimization (GRPO) method tailored for video diffusion to enhance perceptual quality and temporal coherence; subsequently, we integrate Prompt Enhancement via a specialized language model to refine user inputs, and finally address system efficiency through Inference Optimization. Together, these components provide a systematic approach to improving visual quality, temporal coherence, and instruction following, while preserving the controllability learned during pretraining. The result is a practical blueprint for building scalable post-training pipelines that are stable, adaptable, and effective in real-world deployment. Extensive experiments demonstrate that this unified pipeline effectively mitigates common artifacts and significantly improves controllability and visual aesthetics while adhering to strict sampling cost constraints.

---

### Do LLMs Capture Embodied Cognition and Cultural Variation? Cross-Linguistic Evidence from Demonstratives

Authors: Yu Wang, Emmanuele Chersoni, Chu-Ren Huang
Published: 2026-04-28
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2604.25423v1

Do large language models (LLMs) truly acquire embodied cognition and cultural conventions from text? We introduce demonstratives, fundamental spatial expressions like "this/that" in English and "zhè/nà" in Chinese, as a novel probe for grounded knowledge. Using 6,400 responses from 320 native speakers, we establish a human baseline: English speakers reliably distinguish proximal-distal referents but struggle with perspective-taking, while Chinese speakers switch perspectives fluently but tolerate distal ambiguity. In contrast, five state-of-the-art LLMs fail to inherently understand the proximal-distal contrast and show no cultural differences, defaulting to English-centric reasoning. Our study contributes (i) a new task, based on demonstratives, as a new lens for evaluating embodied cognition and cultural conventions; (ii) empirical evidence of cross-cultural asymmetries in human interpretation; (iii) a new perspective on the egocentric-sociocentric debate, showing both orientations coexist but vary across languages; and (iv) a call to address individual variation in future model design.

---

### FED-FSTQ: Fisher-Guided Token Quantization for Communication-Efficient Federated Fine-Tuning of LLMs on Edge Devices

Authors: Changyu Li, Shuanghong Huang, Jiashen Liu, Ming Lei, Jidu Xing, Kaishun Wu, Lu Wang, Fei Luo
Published: 2026-04-28
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2604.25421v3

Federated fine-tuning provides a practical route to adapt large language models (LLMs) on edge devices without centralizing private data. However, in mobile deployments, the training wall-clock is often dominated by straggler-limited uplink communication under heterogeneous bandwidth, intermittent participation, and non-IID client data. Although parameter-efficient fine-tuning (PEFT) methods such as LoRA and QLoRA reduce local memory and trainable parameters, repeated transmission of adapter updates remains a major bottleneck. We propose Fed-FSTQ, a semantic-sensitivity-aware communication-control primitive for communication-efficient federated LLM fine-tuning. Fed-FSTQ uses a lightweight token-level Fisher proxy to estimate semantic sensitivity, couples token-guided sparsification with mixed-precision adapter-update quantization, and allocates higher communication fidelity to semantically load-bearing evidence while suppressing redundant transmission. The method is drop-in compatible with standard federated PEFT pipelines and requires no change to the server aggregation rule. Experiments on multilingual QA and medical QA under non-IID partitions show that Fed-FSTQ reduces cumulative uplink traffic required to reach a fixed quality threshold by 46-fold relative to a Fed-LoRA baseline and improves straggler-limited wall-clock time-to-accuracy by 52%. Under the corrected Controlled LTE-20Mbps accounting, Fed-FSTQ reduces per-round time from 414.60s to 67.29s and reduces per-round energy from 839.20J to 146.28J, yielding a 6.16-fold speedup. On NVIDIA Jetson-class edge devices, Fisher-guided token reduction also yields up to a 1.55-fold inference speedup, demonstrating deployability under tight resource constraints.

---

### JURY-RL: Votes Propose, Proofs Dispose for Label-Free RLVR

Authors: Xinjie Chen, Biao Fu, Jing Wu, Guoxin Chen, Xinggao Liu, Dayiheng Liu, Minpeng Liao
Published: 2026-04-28
Categories: cs.AI
arXiv: https://arxiv.org/abs/2604.25419v1

Reinforcement learning with verifiable rewards (RLVR) enhances the reasoning of large language models (LLMs), but standard RLVR often depends on human-annotated answers or carefully curated reward specifications. In machine-checkable domains, label-free alternatives such as majority voting or LLM-as-a-judge remove annotation cost but can introduce false positives that destabilize training. We introduce JURY-RL, a label-free RLVR framework that decouples answer proposal from reward disposal: votes from model rollouts propose a candidate answer, and a formal verifier determines whether that candidate can receive positive reward. Concretely, only rollouts matching the plurality-voted answer are rewarded when that answer is successfully verified in Lean. When verification is inconclusive, we invoke ResZero (Residual-Zero), a fallback reward that discards the unverified plurality proposal and redistributes a zero-mean, variance-preserving signal over the residual answers. This design maintains a stable optimization gradient without reinforcing unverifiable consensus. Across three backbone models trained on mathematical data, JURY-RL consistently outperforms other label-free baselines on mathematical reasoning benchmarks and transfers competitively to code generation and general benchmarks. It attains pass@1 performance comparable to supervised ground-truth training, with superior generalization demonstrated by higher pass@k and response diversity.

---

### Biased Dreams: Limitations to Epistemic Uncertainty Quantification in Latent Dynamics Models

Authors: Julia Berger, Bernd Frauenknecht, Sebastian Trimpe, Bastian Leibe
Published: 2026-04-28
Categories: cs.LG
arXiv: https://arxiv.org/abs/2604.25416v2

Model-based reinforcement learning distinguishes between dynamics models operating on proprioceptive states and latent dynamics models typically operating on high-dimensional image observations. Among the latter, Dreamer's Recurrent State Space Model (RSSM) has emerged as a dominant architecture. While ensemble-based epistemic uncertainty has proven effective in proprioceptive dynamics for mitigating model exploitation, guiding exploration, or promoting caution, its behavior in latent dynamics remains largely unexplored. Our experiments reveal that although ensemble disagreement captures local epistemic uncertainty, it does not reliably reflect global compounding model error accumulated over prolonged RSSM latent rollouts. We provide evidence for an attractor behavior that draws rollouts toward well-supported latent regions, where uncertainty diminishes despite increasing discrepancies from the true environment dynamics. This can cause the model to overestimate returns when attractor regions correspond to high-reward behaviors. Our findings reveal a structural limitation of epistemic uncertainty estimation in RSSMs and challenge the assumption that epistemic uncertainty estimation transfers directly from proprioceptive to latent dynamics.

---

### One-shot emergency psychiatric triage across 15 frontier AI chatbots

Authors: Veith Weilnhammer, Lennart Luettgau, Christopher Summerfield, Viknesh Sounderajah, Elise Wilkinson, Virginia Corno, Matthew M Nour
Published: 2026-04-28
Categories: q-bio.NC, cs.AI, cs.HC
arXiv: https://arxiv.org/abs/2604.25415v1

AI chatbots are increasingly used for health advice, but their performance in psychiatric triage remains undercharacterized. Psychiatric triage is particularly challenging because urgency must often be inferred from thoughts, behavior, and context rather than from objective findings. We evaluated the performance of 15 frontier AI chatbots on psychiatric triage from realistic single-message disclosures using 112 clinical vignettes, each paired with 1 of 4 original benchmark triage labels: A, routine; B, assessment within 1 week; C, assessment within 24 to 48 hours; and D, emergency care now. Vignettes covered 9 psychiatric presentation clusters and 9 focal risk dimensions, organized into 28 presentation-by-risk groups. Each group contributed 4 distinct vignettes, with 1 vignette at each triage level. Each vignette was rendered as a realistic human-authored conversational query, and the AI chatbots were tasked with assigning a triage label from that disclosure. Emergency under-triage occurred in 23 of 410 level D trials (5.6%), and all under-triaged emergencies were reassigned to level C urgency. Across target models, average accuracy ranged from 42.0% to 71.8%. Accuracy was highest for level D vignettes (94.3%) and lowest for level B vignettes (19.7%). Mean signed ordinal error was positive (+0.47 triage levels), indicating net over-triage. Dispersion was highest around the middle triage levels. All results were confirmed relative to clinician consensus labels from 50 medical doctors. When presented with user messages containing sufficient clinical information, frontier AI chatbots thus recognized psychiatric emergencies as requiring urgent medical assessment with near-zero error rates, yet showed marked over-triage for low and intermediate risk presentations.

---

### Scaling Probabilistic Transformer via Efficient Cross-Scale Hyperparameter Transfer

Authors: Penghao Kuang, Haoyi Wu, Kewei Tu
Published: 2026-04-28
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.25409v1

Probabilistic Transformer (PT), a white-box probabilistic model for contextual word representation, has demonstrated substantial similarity to standard Transformers in both computational structure and downstream task performance on small models and small to medium sized datasets. However, PT is less robust to hyperparameter choices than standard Transformers, making it harder to scale efficiently. In this work, we follow Maximal Update Parametrization (muP) to rescale PT's parameters, so that hyperparameters optimized on small models can be transferred to larger models without additional tuning. With this approach, we successfully scale PT to models with up to 0.4B parameters. Experiments show that PT consistently outperforms standard transformer under the same parameter budget on Masked Language Modeling (MLM) tasks. We hope this work will contribute to the practical deployment of probabilistic models at substantially larger scales in the future.

---

### Beyond Fidelity: Semantic Similarity Assessment in Low-Level Image Processing

Authors: Runjie Wang, Weiling Chen, Tiesong Zhao, Chang Wen Chen
Published: 2026-04-28
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.25408v1

Low-level image processing has long been evaluated mainly from the perspective of visual fidelity. However, with the rise of deep learning and generative models, processed images may preserve perceptual quality while altering semantic content, making conventional Image Quality Assessment (IQA) insufficient for semantic-level assessment. In this paper, we formalize \textit{Semantic Similarity} as a new evaluation task for low-level image processing, aimed at measuring whether semantic content is preserved after processing. We further present a structured formulation of image semantics based on semantic entities and their relations, and discuss the desired properties and constraints of a valid semantic similarity index. Based on this formulation, we propose Triplet-based Semantic Similarity Score (T3S), which models image semantics through foreground entities, background entities, and relations. T3S combines semantic entity extraction, foreground-background disentanglement, and open-world class/relation modeling. Experiments on COCO and SPA-Data show that T3S consistently outperforms existing fidelity-oriented metrics and representative semantic-level baselines, while better reflecting progressive semantic changes under diverse degradations. These results highlight the importance of semantic assessment in modern low-level vision.

---

### Leveraging Previous-Traversal Point Cloud Map Priors for Camera-Based 3D Object Detection and Tracking

Authors: Markus Käppeler, Özgün Çiçek, Yakov Miron, Abhinav Valada
Published: 2026-04-28
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2604.25405v1

Camera-based 3D object detection and tracking are central to autonomous driving, yet precise 3D object localization remains fundamentally constrained by depth ambiguity when no expensive, depth-rich online LiDAR is available at inference. In many deployments, however, vehicles repeatedly traverse the same environments, making static point cloud maps from prior traversals a practical source of geometric priors. We propose DualViewMapDet, a camera-only inference framework that retrieves such map priors online and leverages them to mitigate the absence of a LiDAR sensor during deployment. The key idea is a dual-space camera-map fusion strategy that avoids one-sided view conversion. Specifically, we (i) project the map into perspective view (PV) and encode multi-channel geometric cues to enrich image features and support BEV lifting, and (ii) encode the map directly in bird's-eye view (BEV) with a sparse voxel backbone and fuse it with lifted camera features in a shared metric space. Extensive evaluations on nuScenes and Argoverse 2 demonstrate consistent improvements over strong camera-only baselines, with particularly strong gains in object localization. Ablations further validate the contributions of PV/BEV fusion and prior-map coverage. We make the code and pre-trained models available at https://dualviewmapdet.cs.uni-freiburg.de .

---

### Benchmarking PyCaret AutoML Against IndoBERT Fine-Tuning for Sentiment Analysis on Indonesian IKN Twitter Data

Authors: Mutia Alfi Mayzaroh, Dwi Fitria Ningsih, Nindi Destriani, Martin C. T. Manullang
Published: 2026-04-28
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.25392v1

This paper benchmarks a classical machine learning approach based on PyCaret AutoML against a deep learning approach based on IndoBERT fine-tuning for binary sentiment analysis of Indonesian-language Twitter comments related to Ibu Kota Nusantara (IKN). The dataset contains 1,472 manually labeled samples, consisting of 780 negative and 692 positive comments. In the machine learning setting, Logistic Regression, Naive Bayes, and Support Vector Machine were evaluated using 10-fold cross-validation, with Logistic Regression achieving the best performance among the classical models at 77.57% accuracy and 77.17% F1-score. In the deep learning setting, the indobenchmark/indobert-base-p1 model was fine-tuned for five epochs and achieved 89.59% test accuracy and 89.37% F1-score. The results show that IndoBERT substantially outperforms the machine learning baselines, highlighting the effectiveness of Transformer-based contextual representations for informal Indonesian social media text.

---

### GeoSearch: Augmenting Worldwide Geolocalization with Web-Scale Reverse Image Search and Image Matching

Authors: Tung-Duong Le-Duc, Hoang-Quoc Nguyen-Son, Minh-Son Dao
Published: 2026-04-28
Categories: cs.IR, cs.CV
arXiv: https://arxiv.org/abs/2604.25390v1

Worldwide image geolocalization, which aims to predict the GPS coordinates of any image on Earth, remains challenging due to global visual diversity. Recent generative approaches based on Retrieval-Augmented Generation (RAG) and Large Multimodal Models (LMMs) leverage candidates retrieved from fixed databases for reasoning, but often struggle with scenes that are absent from the reference set. In this work, we propose GeoSearch, an open-world geolocation framework that integrates web-scale reverse image search into the RAG pipeline. GeoSearch augments LMM prompts with database-retrieved coordinates and textual evidence extracted from web pages. To mitigate noise from irrelevant content, we introduce a two-layer filtering mechanism consisting of image matching, followed by confidence-based gating. Experiments on standard benchmarks Im2GPS3k and YFCC4k demonstrate the superiority of GeoSearch under leakage-aware evaluation. Our code and data are publicly available to support reproducibility.

---

### Co-Writing with AI: An Empirical Study of Diverse Academic Writing Workflows

Authors: Silvia Bodei, Duncan P. Brumby, Katie Fisher, Jon Mella
Published: 2026-04-28
Categories: cs.HC, cs.AI
arXiv: https://arxiv.org/abs/2604.25389v1

Despite AI tools becoming increasingly embedded in academic practice, little is known about how university students integrate them into their writing processes. We examine how students engage with AI across different writing tasks, and how this engagement is shaped by individual factors including AI literacy, writing confidence, trust, authorship concerns, and motivation. Study~1 surveys 107 UK university students to map task-specific and co-occurring patterns of AI use across five writing stages (ideation, sourcing, planning, drafting, and reviewing) and their associations with individual factors. Study~2 complements this by exploring how these patterns can be assembled in practice, through interviews with 12 postgraduates reflecting on their established use of AI in assessed writing. Together, the studies suggest that AI integration is selective and heterogeneous, forming three recurring and value-oriented configurations: (1) early-stage (learning-oriented), where tools support exploration and understanding; (2) late-stage (quality-oriented), where tools support drafting and refinement; and (3) peripheral (productivity-oriented), where tools are used to reduce friction and sustain momentum across the process. We offer a workflow-level account of AI-supported academic writing, showing how students navigate competing priorities of learning, quality, productivity, and authorship, and how they evaluate and take responsibility for AI-generated outputs.

---

### COMPASS: COmpact Multi-channel Prior-map And Scene Signature for Floor-Plan-Based Visual Localization

Authors: Muhammad Shaheer, Miguel Fernandez-Cortizas, Asier Bikandi-Noya, Holger Voos, Jose Luis Sanchez-Lopez
Published: 2026-04-28
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2604.25388v1

Architectural floor plans are widely available priors which contain not only geometry but also the semantic information of the environment, yet existing localization methods largely ignore this semantic information. To address this, we present COMPASS, an algorithm that exploits both geometric and semantic priors from floor plans to estimate the pose of a robot equipped with dual fisheye cameras. Inspired by scan context descriptor from LiDAR-based place recognition, we design a multi-channel radial descriptor that encodes the geometric layout surrounding a position. From the floor plan, rays are cast in 360 azimuth bins and the results are encoded into five channels: normalized range, structural hit type (wall, window, or opening), range gradient, inverse range, and local range variance. From the image side, the same descriptor structure is populated by detecting structural elements in the fisheye imagery. As a first step toward full cross-modal matching, we present a window detection algorithm for fisheye images that uses a line segment detector to identify window frames via vertical edge clustering and brightness verification. Detected windows are projected to azimuthal bearings through the fisheye camera model, producing the hit-type channel of the visual descriptor. As a proof of concept, we generate both descriptors at a single known pose from the Hilti-Trimble SLAM Challenge 2026 dataset and demonstrate that the wall-window pattern extracted from the first frame of each camera closely matches the floor plan descriptor, validating the feasibility of cross-modal structural matching.

---

### Wiki Dumps to Training Corpora: South Slavic Case

Authors: Mihailo Škorić, Cosimo Palma
Published: 2026-04-28
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.25384v2

This paper presents a pipeline designed to transform raw Wikimedia dumps into quality textual corpora for seven South Slavic languages. The work is divided into two major phases. The first involves extracting and cleaning text from raw dumps of Wikipedia, Wikisource, Wikibooks, Wikinews, and Wikiquote. This step requires careful handling of raw wiki markup to isolate, first of all, textual articles, and then usable natural language text within them. The second phase addresses the challenge of questionable or low-quality articles, which are often generated from databases or structured knowledge bases. These articles are characterised by repetitive patterns, generic phrasing, and minimal to no original content. To mitigate their impact, a n-gram-based filtering strategy was employed to detect high levels of textual redundancy between articles and then remove such articles from the corpora entirely. The resulting datasets aim to provide linguistically rich texts suitable for training language models or conducting comparative research across South Slavic languages. By combining systematic extraction with quality control, this work contributes to the creation of reliable, high-information corpora that reflect the authentic cultural contexts of languages. While focused on the South Slavic case in the paper, the approach is mostly language-agnostic and can be generalised to other languages.

---

### ML-SAN: Multi-Level Speaker-Adaptive Network for Emotion Recognition in Conversations

Authors: Kexue Wang, Yinfeng Yu, Liejun Wang
Published: 2026-04-28
Categories: cs.SD, cs.AI, eess.AS
arXiv: https://arxiv.org/abs/2604.25383v1

To establish empathy with machines, it is essential to fully understand human emotional changes. However, research in multimodal emotion recognition often overlooks one problem: individual expressive traits vary significantly, which means that different people may express emotions differently. In our daily lives, we can see this. When communicating with different people, some express "happiness" through their facial expressions and words, while others may hide their happiness or express it through their actions. Both are expressions of 'happiness,' but such differences in emotional expression are still too difficult for machines to distinguish. Current emotion recognition remains at a 'static' level, using a single recognition model to identify all emotional styles. This "simplification" often affects the recognition results, especially in multi-turn dialogues. To address this problem, this paper introduces a novel Multi-Level Speaker Adaptive Network (ML-SAN), which, specifically, effectively addresses the challenge of speaker identity information confusion. ML-SAN does not simply assign a speaker's ID after recognition; instead, it employs a three-stage adaptive process: First, Input-level Calibration uses Feature-Level Linear Modulation (FiLM) to adjust the raw audio and visual features into a neutral space unrelated to the speaker. Then, Interaction-level Gating re-adjusts the trust level for each modality (e.g., voice or facial features) based on the speaker's identity information. Finally, Output-level Regularization maintains the consistency of speaker features in the latent space. Tests on the MELD and IEMOCAP datasets show that our model (ML-SAN) achieves better results, performs exceptionally well in handling challenging tail sentiment categories, and better addresses the diversity of speakers in real-world scenarios.

---

### Benchmarking and Improving GUI Agents in High-Dynamic Environments

Authors: Enqi Liu, Liyuan Pan, Zhi Gao, Yan Yang, Chenrui Shi, Yang Liu, Jingrong Wu, Qing Li
Published: 2026-04-28
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.25380v2

Recent advancements in Graphical User Interface (GUI) agents have predominantly focused on training paradigms like supervised fine-tuning (SFT) and reinforcement learning (RL). However, the challenge of high-dynamic GUI environments remains largely underexplored. Existing agents typically rely on a single screenshot after each action for decision-making, leading to a partially observable (or even unobservable) Markov decision process, where the key GUI state including important information for actions is often inadequately captured. To systematically explore this challenge, we introduce DynamicGUIBench, a comprehensive online GUI benchmark spanning ten applications and diverse interaction scenarios characterized by important interface changes between actions. Furthermore, we present DynamicUI, an agent designed for dynamic interfaces, which takes screen-recording videos of the interaction process as input and consists of three components: a dynamic perceiver, a refinement strategy, and a reflection. Specifically, the dynamic perceiver clusters frames of the GUI video, generates captions for the centroids, and iteratively selects the most informative frames as the salient dynamic context. Considering that there may be inconsistencies and noise between the selected frames and the textual context of the agent, the refinement strategy employs an action-conditioned filtering to refine thoughts to mitigate thought-action inconsistency and redundancy. Based on the refined agent trajectories, the reflection module provides effective and accurate guidance for further actions. Experiments on DynamicGUIBench demonstrate that DynamicUI significantly improves the performance in dynamic GUI environments, while maintaining competitive performance on other public benchmarks.

---

### Safe-Support Q-Learning: Learning without Unsafe Exploration

Authors: Yeeun Lim, Narim Jeong, Donghwan Lee
Published: 2026-04-28
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2604.25379v1

Ensuring safety during reinforcement learning (RL) training is critical in real-world applications where unsafe exploration can lead to devastating outcomes. While most safe RL methods mitigate risk through constraints or penalization, they still allow exploration of unsafe states during training. In this work, we adopt a stricter safety requirement that eliminates unsafe state visitation during training. To achieve this goal, we propose a Q-learning-based safe RL framework that leverages a behavior policy supported on a safe set. Under the assumption that the induced trajectories remain within the safe set, this policy enables sufficient exploration within the safe region without requiring near-optimality. We adopt a two-stage framework in which the Q-function and policy are trained separately. Specifically, we introduce a KL-regularized Bellman target that constrains the Q-function to remain close to the behavior policy. We then derive the policy induced from the trained Q-values and propose a parametric policy extraction method to approximate the optimal policy. Our approach provides a unified framework that can be adapted to different action spaces and types of behavior policies. Experimental results demonstrate that the proposed method achieves stable learning and well-calibrated value estimates and yields safer behavior with comparable or better performance than existing baselines.

---

### CoRE: Concept-Reasoning Expansion for Continual Brain Lesion Segmentation

Authors: Qianqian Chen, Anglin Liu, Jingyang Zhang, Yudong Zhang
Published: 2026-04-28
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2604.25376v1

Accurate brain lesion segmentation in MRI is vital for effective clinical diagnosis and treatment planning. Due to high annotation costs and strict data privacy regulations, universal models require employing Continual Learning (CL) to adapt to evolving clinical tasks without losing previously acquired knowledge. However, existing CL paradigms often suffer from capacity limits or redundant parameter growth, and even advanced dynamic methods rely mostly on image-perception strategies that struggle to handle the substantial pathological and multimodal heterogeneity inherent in brain imaging. To address this issue, we propose Concept-Reasoning Expansion (CoRE) framework, which establishes a joint decision-making mechanism by integrating visual features with structured concepts. Through the alignment of image tokens with a hierarchical concept library, CoRE simulates clinical reasoning to guide both interpretable expert routing and demand-based model growth. This collaborative process ensures model evolution is grounded in clinical priors, preventing redundant parameter expansion while maximizing knowledge reuse. Extensive evaluations across 12 sequential brain lesion MRI tasks demonstrate that CoRE achieves state-of-the-art performance and provides a high knowledge starting point for efficient future adaptation. Its superior few-shot transferability and clinical interpretability further validate its effectiveness in managing non-stationary clinical data streams. Our code will be released soon.

---

### Language corpora for the Dutch medical domain

Authors: B. van Es
Published: 2026-04-28
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2604.25374v2

Background: Dutch medical corpora are scarce, limiting NLP development. Methods: We translated English datasets, identified medical text in generic corpora, and extracted open Dutch medical resources. Results: The resulting corpus comprises +- 36 billion tokens across the medical domain in about 105 million documents, freely available on Hugging Face. Conclusion: This work establishes the first large-scale Dutch medical language corpus for pre-training and downstream NLP tasks.

---

### Language corpora for the Dutch medical domain

Authors: B. van Es
Published: 2026-04-28
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2604.25374v1

\textbf{Background:} Dutch medical corpora are scarce, limiting NLP development. \\ \textbf{Methods:} We translated English datasets, identified medical text in generic corpora, and extracted open Dutch medical resources. \\ \textbf{Results:} The resulting corpus comprises $\pm$ 35 billion tokens across the medical domain in about 100 million documents, freely available on Hugging Face. \\ \textbf{Conclusion:} This work establishes the first large-scale Dutch medical language corpus for pre-training and downstream NLP tasks.

---

### From Cursed to Competitive: Closing the ZO-FO Gap via Input-to-State Stability

Authors: Amir Ali Farzin, Philipp Braun, Iman Shames
Published: 2026-04-28
Categories: math.OC, cs.LG, eess.SY, math.NA
arXiv: https://arxiv.org/abs/2604.25372v2

While it is generally understood that zeroth-order (ZO) algorithms have an extra dependency on their number of iterations for any choice of parameters, compared to their first-order (FO) counterparts, in this work, we show that under several conditions, in expectation, ZO methods do not suffer from extra dimension dependencies in their convergence rates with respect to their FO counterparts. We look at optimisation algorithms from the dynamical systems perspective and analyse the conditions under which one can formulate the average of a ZO algorithm as the average of its FO counterpart with bounded perturbations with values dependent on design parameters. Then, using input-to-state stability properties, we show ZO methods follow the same decay rate as their FO counterparts and converge to a neighbourhood of the fixed point of FO methods, where its radius depends on the bound of the norm of the perturbations, which can be made arbitrarily small. The theoretical findings are illustrated via numerical examples.

---

### PhyloSDF: Phylogenetically-Conditioned Neural Generation of 3D Skull Morphology via Residual Flow Matching

Authors: Kaikwan Lau, Gary P. T. Choi
Published: 2026-04-28
Categories: q-bio.QM, cs.CV
arXiv: https://arxiv.org/abs/2604.25371v2

Generating novel, biologically plausible three-dimensional morphological structures is a fundamental challenge in computational evolutionary biology, hampered by extreme data scarcity and the requirement that generated shapes respect phylogenetic relationships among species. In this work, we present PhyloSDF, a phylogenetically-conditioned neural generative model for 3D biological morphology that integrates two innovations: (1) a DeepSDF auto-decoder regularized by a novel Phylogenetic Consistency Loss that structures the latent space to correlate with evolutionary distances (Pearson r=0.993); (2) a Residual Conditional Flow Matching (Residual CFM) architecture that factorizes generation into analytic species-centroid lookup and learned residual prediction, enabling generation from as few as ~4 specimens per species. We evaluate PhyloSDF on 100 micro-CT-scanned skulls of Darwin's Finches and their relatives across 24 species. The model generates novel meshes achieving 88-129% of real intra-species variation at the code level, with all 180 generated meshes verified as non-memorized. Residual CFM surpasses denoising diffusion (which fails entirely at this scale), standard flow matching (which mode-collapses to 3-6% variation), and a Gaussian mixture baseline in both fidelity (Chamfer Distance 0.00181 vs. 0.00190) and morphometric Fréchet distance (10,641 vs. 13,322). Leave-one-species-out experiments across 18 species demonstrate phylogenetic extrapolation capability, and smooth latent interpolations produce biologically plausible ancestral skull reconstructions.

---

### GPT-Image-2 in the Wild: A Twitter Dataset of Self-Reported AI-Generated Images from the First Week of Deployment

Authors: Kidus Zewde, Simiao Ren, Xingyu Shen, Jiaqi Wu, Yuchen Zhou, Tommy Duong, Zikang Zhang, Ethan Traister, Kewen Xie
Published: 2026-04-28
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2604.25370v2

The release of GPT-image-2 by OpenAI marks a watershed moment in AI-generated imagery: the boundary between photographic reality and synthetic content has never been more difficult to discern. We introduce the GPT-Image-2 Twitter Dataset, the first published dataset of GPT-image-2 generated images, sourced from publicly available Twitter/X posts in the immediate aftermath of the model's April 21, 2026 release. Leveraging the Twitter API v2 and a multi-stage curation pipeline spanning multilingual text heuristics (English, Japanese, and Chinese), browser-automated Twitter "Made with AI" badge verification, and model name variant matching, we curate 10,217 confirmed GPT-image-2 images from 27,662 collected records over a six-day window. We characterize the dataset across four analyses: CLIP-based zero-shot subject taxonomy, OCR text legibility (82.0% of images contain detectable text), face detection (59.2% of images, 22,583 total faces), and semantic clustering (137 CLIP ViT-L/14 clusters). A key negative result is that C2PA content credentials are systematically stripped by Twitter's CDN on upload, rendering cryptographic provenance verification infeasible for social-media-sourced AI images. The dataset and all curation code are released publicly.

---

### Multi-action Tangled Program Graphs for Multi-task Reinforcement Learning with Continuous Control

Authors: Quentin Vacher, Nicolas Beuve, Mickaël Dardaillon, Karol Desnos
Published: 2026-04-28
Categories: cs.AI
arXiv: https://arxiv.org/abs/2604.25369v1

Over the past few decades, machine learning has been widely used to learn complex tasks. Reinforcement Learning (RL), inspired by human behavior, is a great example, as it involves developing specific behaviours for specific tasks. To further challenge algorithms, Multi-Task RL (MTRL) environments have been introduced, requiring a single model to learn multiple behaviors. The Tangled Program Graph (TPG) algorithm is a Genetic Programming (GP) algorithm designed for discrete MTRL environments. Recently, the MAPLE algorithm has been proposed, as another GP algorithm that achieves high results in single task continuous RL environments. A variation of the TPG is proposed alongside MAPLE, named Multi-Action TPG (MATPG) that aggregates MAPLE agents, and creates a control flow to activate them. Initially tested on single task RL environments only, MATPG achieved similar results to MAPLE. In this work, we present a new benchmark based on the MuJoCo Half Cheetah from Gymnasium. This benchmark features five distinct obstacles that are randomly positioned in front of the agent, each of which demands a unique behavior. This benchmark serves as a use case for MATPG, to prove its ability as a GP solution for continuous MTRL environments. Our experiments demonstrate its superiority in this multi-task use case when combined with lexicase selection. Furthermore, we examine the interpretability of the evolved graph, revealing that the decision flow of the model is fully interpretable.

---

### Self-DACE++: Robust Low-Light Enhancement via Efficient Adaptive Curve Estimation

Authors: Jianyu Wen, Jun Xie, Feng Chen, Zhepeng Wang, Chenhao Wu, Tong Zhang, Yixuan Yu, Piotr Swierczynski
Published: 2026-04-28
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.25367v2

In this paper, we present Self-DACE++, an improved unsupervised and lightweight framework for Low-Light Image Enhancement (LLIE), building upon our previous Self-Reference Deep Adaptive Curve Estimation (Self-DACE). To better address the trade-off between computational efficiency and restoration quality, Self-DACE++ introduces enhanced Adaptive Adjustment Curves (AACs). These curves, governed by minimal trainable parameters, flexibly adjust the dynamic range while preserving the color fidelity, structural integrity, and naturalness of the enhanced images. To achieve an extremely lightweight architecture without sacrificing performance, we propose a randomized order training strategy coupled with a network fusion mechanism, which compresses the model into an efficient iterative inference structure. Furthermore, we formulate a physics-grounded objective function based on Retinex theory and incorporate a dedicated denoising module to effectively estimate and suppress latent noise in dark regions. Extensive qualitative and quantitative evaluations on multiple real-world benchmark datasets demonstrate that Self-DACE++ outperforms existing state-of-the-art methods, delivering superior enhancement quality with real-time inference capability. The code is available at https://github.com/John-Wendell/Self-DACE.

---

### HuM-Eval: A Coarse-to-Fine Framework for Human-Centric Video Evaluation

Authors: Bingzi Zhang, Kaisi Guan, Ruihua Song
Published: 2026-04-28
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.25361v1

Video generation models have developed rapidly in recent years, where generating natural human motion plays a pivotal role. However, accurately evaluating the quality of generated human motion video remains a significant challenge. Existing evaluation metrics primarily focus on global scene statistics, often overlooking fine-grained human details and consequently failing to align with human subjective preference. To bridge this gap, we propose HuM-Eval, a novel human-centric evaluation framework that adopts a coarse-to-fine strategy. Specifically, our framework first utilizes a Vision Language Model to perform a coarse assessment of global video quality. It then proceeds to a fine-grained analysis, using 2D pose to verify anatomical correctness and 3D human motion to evaluate motion stability. Extensive experiments demonstrate that HuM-Eval achieves an average human correlation of 58.2%, outperforming state-of-the-art baselines. Furthermore, we introduce HuM-Bench, a comprehensive benchmark comprising 1,000 diverse prompts, and conduct a detailed evaluation of existing text-to-video models, paving the way for next-generation human motion generation.

---

### The Structured Output Benchmark: A Multi-Source Benchmark for Evaluating Structured Output Quality in Large Language Models

Authors: Abhinav Kumar Singh, Harsha Vardhan Khurdula, Yoeven D Khemlani, Vineet Agarwal
Published: 2026-04-28
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2604.25359v1

Large Language Models are increasingly being deployed to extract structured data from unstructured and semi-structured sources: parsing invoices, medical records, and converting PDF documents to database entries. Yet existing benchmarks for structured output generation either focus on schema compliance alone, or evaluate value correctness within a single source domain. We introduce SOB (The Structured Output Benchmark), a multi-source benchmark spanning three source modalities: native text, images, and audio conversations. All models receive a text-normalized representation of their context regardless of source modality; this deliberate design isolates structured-output capability from raw vision or speech-processing quality, ensuring a fair, source-agnostic comparison. Our benchmark comprises 5,000 text evaluation records derived from multi-hop QA drawn from a 25,091-record full corpus, 209 image records from OCR-processed PDFs across seven document types including multi-column layouts, dense tables, scanned historical documents, small-print text, and mathematical typesetting, and 115 audio records from the AMI corpus. Each record pairs a natural-language question with a JSON schema that the model must follow and a ground-truth answer verified against the source context. We evaluate 21 frontier and open-weight models across three source domains and seven metrics. Our results reveal a consistent pattern: models achieve near-perfect schema compliance, yet the best Value Accuracy, measured by exact leaf-value match, reaches only 83.0% on text, 67.2% on images, and 23.7% on audio, where longer context makes extraction substantially harder. We release the dataset, evaluation pipeline, and all related code.

---

### Benchmarking Layout-Guided Diffusion Models through Unified Semantic-Spatial Evaluation in Closed and Open Settings

Authors: Luca Parolari, Nicla Faccioli, Lamberto Ballan
Published: 2026-04-28
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.25358v1

Evaluating layout-guided text-to-image generative models requires assessing both semantic alignment with textual prompts and spatial fidelity to prescribed layouts. Assessing layout alignment requires collecting fine-grained annotations, which is costly and labor-intensive. Consequently, current benchmarks rarely provide comprehensive layout evaluation and often remain limited in scale or coverage, making model comparison, ranking, and interpretation difficult. In this work, we introduce a closed-set benchmark (C-Bench) designed to isolate key generative capabilities while providing varying levels of complexity in both prompt structure and layout. To complement this controlled setting, we propose an open-set benchmark (O-Bench) that evaluates models using real-world prompts and layouts, offering a measure of semantic and spatial alignment in the wild. We further develop a unified evaluation protocol that combines semantic and spatial accuracy into a single score, ensuring consistent model ranking. Using our benchmarks, we conduct a large-scale evaluation of six state-of-the-art layout-guided diffusion models, totaling 319,086 generated and evaluated images. We establish a model ranking based on their overall performance and provide detailed breakdowns for text and layout alignment to enhance interpretability. Fine-grained analyses across scenarios and prompt complexities highlight the strengths and limitations of current models. Code is available at https://github.com/lparolari/cobench.

---

### GraphPL: Leveraging GNN for Efficient and Robust Modalities Imputation in Patchwork Learning

Authors: Xingjian Hu, Zuoyu Yan, Jianhua Zhu, Liangcai Gao, Fei Wang, Tengfei Ma
Published: 2026-04-28
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2604.25352v1

Current research on distributed multi-modal learning typically assumes that clients can access complete information across all modalities, which may not hold in practice. In this paper, we explore patchwork learning, in which the modalities available to different clients vary, and the objective is to impute the missing modalities for each client in an unsupervised manner. Existing methods are shown not to fully utilize the modality information as they tend to rely on only a subset of the observed modalities. To address this issue, we propose GraphPL, which combines graph neural networks with patchwork learning to flexibly integrate all observed modalities and remains robust with noisy inputs. Experimental results show that GraphPL achieves SOTA performance on benchmark datasets. Our results on real-world distributed electronic health record dataset show GraphPL learns strong downstream features and enables tasks like disease prediction via superior modality imputation.

---

### A Faceted Proposal for Transparent Attribution of AI-Assisted Text Production

Authors: Geraldo Xexéo
Published: 2026-04-28
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2604.25346v1

Artificial intelligence systems are increasingly integrated into writing processes, challenging traditional notions of authorship, responsibility, and intellectual contribution. Current disclosure practices usually indicate whether AI was used, but rarely explain how it was used, where it intervened, or how its output was reviewed. This paper proposes a faceted model for representing AI-assisted text production at the levels of documents, chapters, sections, and paragraphs. The proposal introduces a core model based on Form, Generation, and Evaluation, and an extended model that adds Intent, Control, and Traceability. The model is positioned as a minimal operational baseline with extensibility toward higher-fidelity representations. A worked example based on the production of this article demonstrates applicability.

---

### Plausible but Wrong: A case study on Agentic Failures in Astrophysical Workflows

Authors: Shivam Rawat, Lucie Flek
Published: 2026-04-28
Categories: cs.AI, astro-ph.IM
arXiv: https://arxiv.org/abs/2604.25345v2

Agentic AI systems are increasingly being integrated into scientific workflows, yet their behavior under realistic conditions remains insufficiently understood. We evaluate CMBAgent across two workflow paradigms and eighteen astrophysical tasks. In the One-Shot setting, access to domain-specific context yields an approximately ~6x performance improvement (0.85 vs. ~0 without context), with the primary failure mode being silent incorrect computation - syntactically valid code that produces plausible but inaccurate results. In the Deep Research setting, the system frequently exhibits silent failures across stress tests, producing physically inconsistent posteriors without self-diagnosis. Overall, performance is strong on well-specified tasks but degrades on problems designed to probe reasoning limits, often without visible error signals. These findings highlight that the most concerning failure mode in agentic scientific workflows is not overt failure, but confident generation of incorrect results. We release our evaluation framework to facilitate systematic reliability analysis of scientific AI agents.

---

### VAE-Inf: A statistically interpretable generative paradigm for imbalanced classification

Authors: Hongfei Wu, Ruijian Han, Yancheng Yuan
Published: 2026-04-28
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2604.25334v1

Imbalanced classification remains a pervasive challenge in machine learning, particularly when minority samples are too scarce to provide a robust discriminative boundary. In such extreme scenarios, conventional models often suffer from unstable decision boundaries and a lack of reliable error control. To bridge the gap between generative modeling and discriminative classification, we propose a two-stage framework \textbf{VAE-Inf} that integrates deep representation learning with statistically interpretable hypothesis testing. In the first stage, we adopt a one-class modeling perspective by training a variational autoencoder (VAE) exclusively on majority-class data to capture the underlying reference distribution. The resulting latent posteriors are aggregated via a Wasserstein barycenter to construct a global Gaussian reference model, providing a geometrically principled baseline for the majority class. In the second stage, we transform this generative foundation into a discriminative classifier by fine-tuning the encoder with limited minority samples. This is achieved through a novel distribution-aware loss that enforces probabilistic separation between classes based on variance-normalized projection statistics. For inference, we introduce a projection-based score that admits a natural hypothesis testing interpretation, allowing for a distribution-free calibration procedure. This approach yields exact finite-sample control of the Type-I error (false positive rate) without relying on restrictive parametric assumptions. Extensive experiments on diverse real-world benchmarks demonstrate that our framework achieves competitive performance against other approaches. The codes are available upon request.

---