# Research Papers — 2026-08 · bundle page 65 of 240 > Full-text markdown of papers #3201–3250 of 11995 published in 2026-08, newest first. AI-readable mirrors of the full 11995-paper 2026-08 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-08/page/65 [← Previous bundle](https://aipod.dev/papers/archive/2026-08/page/64.md) [Next bundle →](https://aipod.dev/papers/archive/2026-08/page/66.md) ### An end-to-end-trained vision-language model for native-language prostate pathology report generation Authors: Christian Grashei, Fabian Gülhan, Maximilian Legnar, Fabian Stögbauer, Cleo-Aron Weis, Carolin Mogler, Peter Schüffler Published: 2026-08-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.23143v1 Prostate cancer is among the most frequently diagnosed malignancies worldwide, and structured reporting of each biopsy core burdens pathologists. Existing tools frame this as classification, leaving pathologists to assemble coherent reports, while many slide-level vision-language models rely on English-centric encoders that transfer poorly to other clinical languages. We present a slide-level framework generating prostate biopsy reports that is language-independent by construction: tokenizer and model are trained from scratch, demonstrated here in German. To address paired-data scarcity, an automated pipeline uses a locally deployed large language model to split composite reports into core-specific image-text pairs, yielding 17,344 pairs from 2,402 historical cases without manual annotation. Evaluated for clinical attributes rather than linguistic similarity, the model achieves 96.2% F1 for malignancy detection and 65.2% for Gleason grading, competitive with an FDA-cleared classifier. Grading is further validated on three external cohorts with latent-space augmentation. Institutions can thus train native-language reporting models on their own archives. --- ### How Merge-Tolerant Are Vision Transformers for Wheat Phenotyping? Authors: Simon Ravé, Pejman Rasti, David Rousseau Published: 2026-08-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.23142v1 Vision-based wheat phenotyping requires repeated measurements under deployment constraints, from growth-stage recognition to wheat-head counting and organ segmentation. Plain Vision Transformers (ViTs) provide a common architecture for these tasks, but quadratic attention limits high-throughput and edge inference. Training-free token merging is attractive because it can be inserted into trained models without retraining. We provide a systematic benchmark of ToMe and Mutual Pair Merging across growth-stage classification, wheat-head detection, and wheat-organ segmentation, measuring task quality, throughput, token count, and peak GPU memory, with additional Raspberry Pi 5 measurements. The benchmark reveals a clear hierarchy: classification is highly merge-tolerant, while detection and segmentation are constrained by repeated instances, thin organs, dense boundaries, reconstruction, and runtime overhead. Optimized attention backends can erase apparent speedups, so deployment value must be profiled on the target runtime rather than inferred from token count. --- ### MIVIFI: Bridging Perspective and Fisheye Domains for Training Multi-View Fisheye Image Generation Models Authors: Matthias Neuwirth-Trapp, Begüm Altunbas, Jiayi Wang, Yan Xia, Maarten Bieshaar, Xinyu Huang, Daniel Cremers Published: 2026-08-24 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2608.23140v1 Achieving 360° coverage is critical for the visual perception systems of autonomous vehicles. Fisheye cameras offer a cost-effective solution by enabling full surround coverage with as few as two sensors. However, existing multi-view fisheye datasets are limited, and synthesizing rare corner cases typically requires computationally expensive 3D simulations, hindering the training. While generative models have achieved significant success in standard perspective imagery, their application to wide-angle distortion remains unexplored. In this work, we formally introduce the novel problem of multi-view fisheye image generation conditioned on volumetric semantic representations and present two distinct methods. We first propose SyntheOcc-FE, which adapts the SyntheOcc architecture to fisheye data. While effective, this method is constrained by the scarcity of fisheye datasets, which limits its generalization. To overcome these limitations, we propose our second method, MIVIFI (multi-view fisheye), which leverages cross-domain learning with Equirectangular Projections. By bridging the gap between dataset domains using KITTI-360 fisheye images alongside nuScenes multi-view standard images, our approach enables high-fidelity manipulation of scene content. This framework enables the structural modification of semantic occupancy inputs to introduce or eliminate specific actors and facilitates the rendering of diverse meteorological conditions and illumination scenarios absent in the limited fisheye datasets. Quantitative and qualitative experiments demonstrate that our methods achieve robust photorealistic multi-view fisheye image generation and highlight the specific advantages of our cross-domain strategy for handling data scarcity. --- ### Pointing-VLA: Typed Spatial Grounding Interfaces for Vision-Language-Action Manipulation Authors: Xiwen Chen, Zelin Li, Zhiruo Zhou, Huiming Chen, Chenwei Wang, Xiaojun Zhu Published: 2026-08-24 Categories: cs.RO, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2608.23138v1 Vision-language-action (VLA) models often expose spatial grounding through autoregressive text coordinates or opaque action tokens, creating brittle interfaces between multimodal reasoning and robot execution. We present Pointing-VLA, a typed hidden-state spatial readout built on Embodied-R1. Geometry-specific heads predict normalized points, object-functional grounding (OFG) heatmaps, and visual trajectories without serializing geometry as text. For the evaluated Bridge/WidowX and physical pick-place deployments, an explicit execution contract assigns PICK to source-conditioned OFG and PLACE to Pointing, providing direct stage-aligned spatial targets. Pointing-VLA achieves SOTA performance on Bridge/WidowX, averaging 72.9\% across the evaluated four-task set without Bridge-specific finetuning under collision-enabled CuRobo execution. Pointing and OFG show complementary strengths across native and cross-dataset evaluations. The OFG/contact readout transfers to NORA-1.5, preserving or improving success while reducing recorded controller time by more than 20$\times$; typed heads are also 6.68--6.90$\times$ faster than Embodied-R1 text decoding on a shared external suite. When integrated as spatial guidance for a $π_{0.5}$ action policy, Pointing-VLA raises autonomous real-robot success from 52.7\% to 80.7\% across three visual contexts. These results establish typed spatial readouts as an efficient, inspectable interface between embodied reasoning and robot execution. --- ### Simulate, record, verify: A language-portable framework for muscle-grounded articulatory QA (extended version) Authors: Seungho Eum, Shantong Sun, Unsang Park Published: 2026-08-24 Categories: cs.CV, cs.HC arXiv: https://arxiv.org/abs/2608.23137v6 Articulatory data describe what the tongue looks like, but geometry alone does not explain the muscle-level why behind a configuration or how to move it toward a target posture. We present a simulator-based framework that constructs this supervision from controlled biomechanical inputs. Each simulated configuration is stored with its generating muscle state and geometric properties in a structured fact record. Gold answers are derived before linguistic realization, and every naturalized output is checked against its source record, so the same records support new languages and question types without re-simulation. We instantiate the framework as 3DTongueQA using the ArtiSynth Badin tongue model. From 295,115 meshes, we construct 891,156 record-checked QA instances per language, and the checker detects over 96.9% of injected corruptions. Korean and Spanish realizations and a new question type are added from the same records, confirming portability. A SpiralNet++-Qwen3 probe reaches 62.9 Muscle EM, above 44.0 for nearest-neighbor transfer and 7.2 for a zero-shot commercial LLM given the same mesh, while mesh shuffling drops it to near zero. The supervision is learnable, useful beyond retrieval, and grounded in paired geometry. It reflects simulator-defined states rather than measured physiology. Code, templates, and QA sets are available at https://github.com/esh0504/muscle-grounded-qa. --- ### Bridge Damage Detection from Low-Light UAV Imagery via Degradation-Aware Mixture-of-Experts Enhancement Authors: Hu Wang, Hongxu Pu, Zhiqi Hu, Fangzhou Lin, Wang Wang Published: 2026-08-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.23136v1 Poor illumination obscures small, low-contrast defects in UAV bridge imagery, reducing the reliability and operational flexibility of automated inspection. This paper investigates whether degradation-aware image restoration can improve bridge damage detection under low-light conditions and transfer from synthetic degradations to real inspection scenes. We propose DaL- MoE, a detector-agnostic restoration front end trained with an ISP-aware low-light synthesis pipeline and equipped with degradation-aware guidance estimation and complementary experts for noise suppression, color adjustment, and structural-detail recovery. On paired synthetic data, DaL-MoE achieves 23.12 dB PSNR and 0.8482 SSIM, increasing YOLOv11m box mAP50 from 0.3097 to 0.4923 and mask mAP50 from 0.2281 to 0.3529. On real low-light UAV imagery without paired normal-light references, sim-to-real evaluation shows improved defect visibility and more complete detections than direct inference on raw low-light inputs. Future work will develop low-light-aware bridge damage detectors with stronger cross-scene generalization across bridge sites, imaging conditions, and illumination levels. --- ### LITERARYBIGFIVE: Author-Personalized Text Generation in a Unified Interpretable Space Authors: Jinghui Zhang, Lang Gao, Ao Li, Mingzhe Li, Ruihong Zeng, Zirui Song, Kentaro Inui, Xiuying Chen Published: 2026-08-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.23124v3 Personalized text generation for authors and literary writing is essential for applications such as adaptive writing assistants, creative support tools, and computational literary analysis. However, existing approaches to author modeling and personalization often represent writing behavior as independent labels, requiring large-scale corpus collection or fine-tuning for each author or stylistic category. Such formulations are costly, difficult to interpret, and poorly suited for generalizing across authors. Inspired by the Big Five model's dimensional view of personality, we propose LiteraryBigFive, a framework that reframes authorial writing characteristics as coordinates within a unified and interpretable space. In this space, we derive each interpretable axis (e.g., Classicism, Emotionality) from activation-space contrasts between author-written and neutral passages, yielding distinct stylistic dimensions that allow texts or authors to be positioned within a five-dimensional system. Beyond localizing different authors, we further introduce an interpretable steering mechanism, which adaptively guides text generation toward target coordinates to perform author-personalized writing. Experimental results show that LiteraryBigFive improves authorial expressiveness while preserving semantic fidelity. The derived author per-axis scores strongly correlate with real-world literary consensus, offering transparent and interpretable explanations of author-specific generation behavior: https://github.com/Znull-1220/LiteraryBigFive. --- ### Statistical Machine Translation Systems of English-Pnar Language Pair : Some Insights of the Emperical Study Authors: Edawanbiang Dhar Surmila Thokchom, Thoudam Doren Singh Published: 2026-08-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.23120v1 Pnar, an Austroasiatic language spoken by approximately 0.4 million people in the Jaintia Hills of Meghalaya, lacks the digital corpora and natural language processing (NLP) resources. This paper presents the first machine translation study for the English and Pnar language pair. Using articles collected from the Wyrta newspaper, we built a parallel corpus comprising of 10,234 sentences and trained phrase-based statistical machine translation (SMT) systems the models using 9,563 parallel corpora under three configurations for each direction using Moses, GIZA++ , KenLM, varying lexicalized reordering and minimum error rate training (MERT) tuning. The models are evaluated on a held out test set of 371 sentences, the best performing system achieves a BLEU score of 14.97 (chrF2: 33.42, TER: 77.60) for Pnar to English and 11.16 (chrF2: 31.38, TER: 93.51) for English to Pnar, establishing the first quantitative benchmark for this language pair. Lexicalized reordering improves translation quality by 3.73 BLEU points for Pnar to English, reflecting the structural shift from the source language's SOV word order to the target language's SVO order, whereas MERT tuning degrades BLEU performance under low resource conditions. Finally, we analyze the remaining translation errors, including morphological out of vocabulary (OOV) words, long-distance reordering and Khasi code mixing and discuss future directions toward neural and multilingual machine translation for Pnar. --- ### Quantum Reservoir Computing with Physics-Informed Correction for Reduced-Order PDE Forecasting Authors: Krishna Bhatia, Harsh, Shalini Devendrababu Published: 2026-08-24 Categories: quant-ph, cs.LG arXiv: https://arxiv.org/abs/2608.23119v1 We study a hybrid proposal--correction architecture for reduced-order PDE forecasting in which a pure-state quantum reservoir computer (QRC) predicts latent coefficient dynamics and a PINN-based physics-informed corrector (PIC) refines local rollout windows. The method is evaluated on Burgers and Kuramoto--Sivashinsky (KS), with KS as the primary chaotic benchmark. On KS, QRC+PIC consistently improves over QRC alone in RMSE, NRMSE, and PDE residual, while Burgers highlights a regime in which simple baselines remain strong. These results suggest that QRC proposals with local physics-informed correction are a viable benchmark-dependent reduced-order forecasting strategy. --- ### DeMixPert: Decomposed Response Modeling with Gaussian Mixtures for OOD Single-Cell Perturbation Prediction Authors: Jiawen Liu, Xuechenxiao Cao, Yutong Li, Bing Liu, Jiaming Liang, Tinghe Zhang, Xiaoqi Sheng, Hongmin Cai Published: 2026-08-24 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.23114v1 Predicting transcriptome-wide responses to unseen genetic perturbations remains a major computational challenge because accurate prediction requires recovering both perturbation-specific transcriptional shifts and heterogeneous cellular responses. Existing methods often entangle deterministic response structure with stochastic population-level variation, causing dominant shared patterns to mask weaker perturbation-specific signals and impair distributional modeling. To address these challenges, we propose \textbf{DeMixPert}, an approach for Decomposed response Modeling with Gaussian Mixtures for Out-Of-Distribution (OOD) single-cell Perturbation prediction. DeMixPert decomposes perturbation-induced changes into a basal-state-dependent systematic response, a perturbation-specific response, and population-level variation. The systematic component is derived from the basal state encoded from control-cell expression, whereas the perturbation-specific component is inferred from pretrained target embeddings for unseen-target generalization. DeMixPert models population-level variation using a Gaussian prototype Invertible Network and adaptively combines reusable Gaussian prototypes according to the basal state and perturbation condition. The resulting mixture is mapped to a condition-specific variation distribution. Sampled variations are integrated with the systematic and perturbation-specific components, followed by joint decoding with the basal state to reconstruct perturbed-cell gene expression. Experimental results show that DeMixPert effectively captures heterogeneous single-cell perturbation responses and achieves superior performance across unseen-perturbation settings. The source code is made publicly available upon publication. --- ### Molecular LLM Agents: From Architectural Design to Scientific Autonomy Authors: Jiatong Li, Wengyu Zhang, Weida Wang, Yuxuan Ren, Wei Liu, Chenyang Mao, Yuqiang Li, Yatao Bian, Changmeng Zheng, Xiaoyong Wei, Qing Li Published: 2026-08-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.23104v2 Molecular science represents an important frontier for LLM-based agents. Unlike general agents that mainly operate over natural language, code, or web environments, molecular LLM agents must perceive, reason about, and act upon chemical objects across symbolic strings, molecular graphs, 3D conformations, spectra, simulations, and wet-lab measurements. Their capabilities depend on chemically faithful molecular perception, an LLM-centered agent framework, domain-specific tool grounding, and computational or experimental feedback, in addition to planning and tool use. This work develops a conceptual framework for molecular LLM agents from two complementary perspectives. First, we introduce an architectural view of molecular-agent design, covering molecular representation and perception, the agent framework, domain-specific toolboxes, and learning and optimization. Second, we propose a scientific autonomy ladder inspired by staged autonomy in engineering systems, categorizing agents into four levels: L1 assistive or fixed workflows, L2 adaptive computational agents, L3 feedback-aware physical experiment agents, and L4 scientific-agenda agents. Together, these two perspectives establish a comprehensive framework for comparing existing molecular LLM agents, identifying missing capabilities and deployment risks, and guiding the design, evaluation, and deployment of future agents in molecular discovery workflows. --- ### Training-Free Pseudo-Fusion for Composed Image Retrieval with Diffusion Models and Multimodal Large Language Models Authors: Fan Xu, Luis A. Leiva Published: 2026-08-24 Categories: cs.CV, cs.IR arXiv: https://arxiv.org/abs/2608.23102v1 Composed Image Retrieval (CIR) is an emerging paradigm in content-based image retrieval that enables users to formulate compositional queries by combining a reference image with an auxiliary modality, usually text-based. This approach supports fine-grained search where the target image shares structural elements with the user-provided image while incorporating the modifications specified by the auxiliary text. Conventional CIR methods rely on multimodal fusion to combine visual and textual features into a joint query embedding, which requires training modules that align composed queries with the targets. In this work, we propose PeFuse (for pseudo-fusion), a training-free framework that leverages pretrained Diffusion Models and Multimodal Large Language Models to bridge modalities via generative conversion. We introduce two novel strategies: uni-directional and bi-directional conversion, which convert CIR into four single-modality retrieval problems. These methods reformulate CIR as either intra-modal or cross-modal single-query retrieval tasks, bypassing the need for dedicated task-specific training. Extensive experiments on standard benchmarks demonstrate that converting CIR into text-to-image retrieval tasks is more effective than alternative conversion strategies, achieving competitive or superior performance compared with state-of-the-art methods, while maintaining high flexibility thanks to replaceable components of the conversion pipeline. These results highlight the effectiveness of the pseudo-fusion paradigm for zero-shot CIR. Our code is publicly available at: https://github.com/StevenXuf/PeFuse4CIR. --- ### PolyChirp: Multi-Species Birdsong Classification Using TinyML on Low-Power Acoustic Sensors Authors: Nathan Duboisset, Zhaolan Huang, Felix Bießmann, Roudy Dagher, Antoine Lavandier, Emmanuel Baccelli Published: 2026-08-24 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.23101v2 Recent progress in the field of TinyML has demonstrated that low-power hardware based on microcontrollers can achieve bird species monitoring in real time based on acoustic sensor data for an entire breeding period on a single battery charge. However, the state of the art on low-power microcontrollers was so far limited to binary classification of a single species. In contrast, real fauna monitoring deployments often target multiple species simultaneously. To address this challenge we develop PolyChirp, an approach combining biological domain expertise, automated dataset curation, neural architecture optimization and novel hardware to achieve multiclass bird species detection in the wild. PolyChirp is based on newly designed tiny multiclass models that leverage recent microcontrollers and hardware acceleration with a neural processing unit (NPU). We evaluate the predictive performance of these models, and we measure their computational performance -- flash footprint, latency, energy consumption -- on common microcontroller hardware. Our results demonstrate that PolyChirp matches or exceeds the TinyChirp architectures retrained under our protocol on single-species detection, and further achieves robust classification of up to 10 species simultaneously (macro F2 up to 0.97), while still fitting the flash, latency and energy budget of a low-power microcontroller sensor. A data-driven front-end redesign additionally makes on-device mel feature extraction 7x to 11x cheaper. --- ### Shaping the Evolutionary Dynamics of Robot Morphology via Adaptive Control Learning Authors: Junru Song, Yang Yang, Yaqing Xu, Ying Wen, Wei Peng, Guozhen Li, Wei'en Zhou, Wen Yao Published: 2026-08-24 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2608.23100v1 Robot co-design via bi-level optimization couples within-lifetime controller learning for fitness evaluation with cross-generational morphological evolution. Prior work has established that well-adapted morphology facilitates faster control learning, a property termed morphological intelligence. Yet how control learning reciprocally shapes morphological evolution remains unexplored. This paper examines both directions for a holistic account of brain-body interplay. We first show that morphological contributions to control learning decouple into two orthogonal dimensions. We formalize the convergence speed as morphological intelligence and identify the performance ceiling as a complementary quantity termed true potential. A concise functional relation is then established to jointly characterize both quantities from individual learning curves, which, when aggregated at the population level, capture evolutionary profiles. Through extensive experiments on simulated voxel-based soft robots, we reveal that premature fitness evaluation systematically underestimates true potential and biases selection towards fast learners. This restricts design space exploration, compromising both optimization efficiency and morphological diversity. Notably, the widely recognized morphological Baldwin effect emerges as an artifact of this bias rather than a general evolutionary tendency. We therefore propose AdaControl, which monitors disproportionate selection for morphological intelligence during evolution and allocates minimally sufficient control learning for unbiased fitness evaluation. With AdaControl, a simple genetic algorithm rivals state-of-the-art generative-model-based co-design methods in discovering diverse high-performing designs while cutting computation by up to 80% versus exhaustive control. --- ### Jiuge-Tuiqiao: An Interpretable Human-AI System for Classical Chinese Poetry Refinement Authors: Yufeng Han, Lifan Deng, Cunliang Kong, Wenhao Li, Xin Cong, Yuzhuo Bai, Kangyang Luo, Maosong Sun Published: 2026-08-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.23098v2 Classical Chinese poetry composition has long valued Tuiqiao, the iterative refinement of words, imagery, and prosody. However, many current AI poetry systems follow a one-shot generation paradigm, which reduces users to prompt providers and weakens their creative agency. We present Jiuge-Tuiqiao, an interactive human-AI collaborative system for classical Chinese poetry composition. The system is designed around a triadic model: user-driven control, ancient-guided evidence, and AI-assisted generation. Users can lock characters or lines, receive real-time prosody feedback, and obtain interpretable refinement suggestions grounded in high-frequency collocations, PPL-ranked classical lines, and structured knowledge extracted from classical encyclopedias. This design turns AI from an autonomous generator into a background assistant that supports the user's own process of poetic refinement. Preliminary experiments and user feedback suggest that Jiuge-Tuiqiao improves controllability, interpretability, and user engagement in classical poetry composition. --- ### Definitional Sensitivity in Media Bias Detection: A Multi-Definition Dataset and Benchmark Authors: Martin Wessel, Timo Spinde, Jürgen Pfeffer, Gianluca Demartini Published: 2026-08-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.23095v2 Media bias detection relies on definitions and examples that specify what counts as bias, yet these specifications often vary across datasets or remain implicit, even when given the same name. Such variation makes it unclear whether models trained for the same bias category learn the same construct or different phenomena, a problem largely overlooked in prior work. We examine how definition choice affects bias annotation in a between-subjects experiment with 354 participants and a parallel evaluation with four LLMs. Participants and models rate six news articles across four bias categories using definitions that vary in conceptual framing and elaboration. Across 8,496 human and 28,800 LLM ratings, we find that the conceptual target of a definition drives annotation divergence, while construct-preserving elaboration does not: conceptual framing significantly shifts annotations for humans and does so even more strongly for LLMs. We discuss implications for construct specification in annotation protocols and prompt-based measurement, and consider how definitional sensitivity may propagate to downstream classification beyond media bias. We also release MUDD, the Multi-Definition Bias Detection Dataset. --- ### One Inverse Step is a Convex Program: Bayes-Limit Calibration of Diffusion Inversion Authors: Gordei Verbii Published: 2026-08-24 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2608.23094v1 One implicit DDIM inversion step is the cheapest probe of whether a pretrained diffusion model encodes local manifold geometry. It is the stationarity condition of an explicit potential, $x-G(x)=\nablaΨ_t(x)$, strongly convex at the Bayes limit with modulus exactly $e^{-h_t}$ for the step's log-SNR gap $h_t$ $-$ for every data law, schedule and point, with no manifold, reach or unimodality hypothesis. Three consequences must be kept apart. (i) The solution is unique at the Bayes limit; a second one requires the trained score to violate the posterior-covariance bound by $1/(1-e^{-h_t})$, a hypothesis-free certificate of model error; the same bound makes contraction a schedule constant, $ρ_g^{\star}=1-e^{-h_t}<0.326$ throughout the standard DDPM schedule. (ii) The solver can still fail: Picard iteration is unit-step gradient descent on $Ψ_t$, unstable wherever $λ_{\max}(\nabla^2Ψ_t)>2$, so oscillation certifies nothing; damping below $2/λ_{\max}$ cures it. (iii) The geometry lives in the convergence domain: on the scale-free depth $w=rκ_{\max}$ the oscillation shell sits at $w=\tfrac12$, schedule-free, and the divergence shell at $w=1/(1+ρ_g^{\star})$, with a measured finite-noise correction in $\|\mathrm{II}\|^2$. Exact scores reproduce both to within $0.54\%$ on three classes; no trained score we probe shows a shell $-$ a derived limitation, not a null result: the Fermi window conflicts with the model's own training support by $3.6$-$5.6\times$, and the trained Hessian-Lipschitz constant is $2$-$12\%$ of the curvature the law reads, $0$ on a ReLU net. Finally the unconditional ceiling $σ_tλ_{\max}(\mathrm{sym}\,J)\le1$, from $\mathrm{Cov}(x_0\mid x_t)\succeq0$ alone, holds for the exact score to $3\times10^{-7}$ but is violated in all DDPM CIFAR-10/CelebA-HQ-256 settings, by $1.26$-$4.66\times$. --- ### Partial-Moment PINNs for Caldeira--Leggett Parameter Learning in Quantum Brownian Motion Authors: Krishna Bhatia Published: 2026-08-24 Categories: quant-ph, cs.LG arXiv: https://arxiv.org/abs/2608.23093v1 We study parameter recovery in the Caldeira--Leggett (quantum Brownian) oscillator from partial moment traces. Our model is a moment-level PINN that predicts the five first/second moments and enforces the linear CL/HPZ ODEs by automatic differentiation. Physical structure is imposed through a PSD (Cholesky) covariance head, high-temperature CL assumptions with $D_{xp}\approx0$, and fluctuation--dissipation ties between $D_{pp}$ and $γ$. On synthetic CL data with channels ${μ_x,σ_{xx},σ_{xp}}$, the constrained variant recovers $(ω,γ)$ accurately, stabilizes $D_{pp}$, and achieves low rollout error compared to finite differences and Kalman--EM (expectation--maximization) with exact Van Loan discretization. Fisher-style checks confirm that diffusion needs at least one variance observable, and sparse $σ_{pp}$ ``anchors'' restore conditioning. We also show that the same PINN can learn time-varying HPZ coefficients. --- ### Loopy: Seamless Video Loop Generation via Anchored Looping Shift of Positional Embedding Authors: Haotian Dong, Wenjing Wang, Chen Li, Jing Lyu, Xin Wang, Di Lin Published: 2026-08-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.23090v1 Looping videos are essential for practical applications such as web graphics, game development, and social media. However, existing approaches typically fail to generate high-quality looping videos due to the neglect of how video generation models perceive temporal order and how this relates to the looping behavior. In this work, we are the first to reveal that position embedding at different attention layers within DiT exhibits varying levels of positional control, with the most pronounced layer acting as an anchor. We formulate this anchored layer as the reference point of the looping video, offering strong contextual priors for the remaining layers to facilitate the generation of seamless and coherent video content. Based on this insight, we propose an anchored position embedding shifting strategy that applies layer-specific shift lengths according to each layer's temporal control effect, effectively transforming DiT's temporal perception from a straight line to a circle. Leveraging this strategy, we develop a general framework, Loopy, for high-quality looping video generation, supporting both RGB and RGBA videos, while also enabling advanced AIGC features such as identity control and style transfer. Experiments demonstrate that our approach significantly improves temporal consistency and visual fidelity in generated looping videos. The released model is available on our website: https://donghaotian123.github.io/Loopy. --- ### POOL: Propagated Uncertainty Over Lookalikes Authors: Rounak Sharma, Ananya B. Sai, Soumyabrata Pal Published: 2026-08-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.23086v1 Black-box large language models need confidence scores that can separate likely-correct from likely-incorrect outputs, enabling systems to prioritize human review, route uncertain cases to stronger models, or choose abstention thresholds on development data. Yet existing confidence estimators face a cost-quality trade-off: verbal confidence is cheap but is often overconfident, while sampling-based uncertainty is more informative but scales linearly with the number of samples per query. We propose \textsc{POOL} (\emph{Propagated Uncertainty Over Lookalikes}),a cost-efficient framework that addresses this trade-off taking inspiration from group-testing.\textsc{POOL} clusters query stems with overlaps, evaluates a base estimator on representative medoids, softly propagates confidence scores to nearby queries, and selectively evaluates high-disagreement cases. We instantiate this framework with \textsc{Hy@}$p$, a hybrid estimator that combines verbal confidence with spectral answer diversity computed from the negative von Neumann entropy of sampled answer embeddings.Across six domains from three datasets and five black-box LLMs, \textsc{Hy@}5 achieves higher average AUROC than verbal confidence and \textsc{Vn@}10 sampling while using half as many samples as \textsc{Vn@}10. \textsc{POOL}-\textsc{Hy@}5 retains 93.5--97.9\% of its AUROC while saving 19.3--39.3\% of generations. On paraphrase-dense workloads, generation savings rise to 73-76\%, showing that semantic redundancy can be leveraged to lower confidence-estimation costs. --- ### AgentWeave: Routing Before Reasoning for Efficient Function Calling in Tool-Rich Language Models Authors: Saurav Singla, Aarav Singla, Advik Gupta, Parnika Gupta Published: 2026-08-24 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2608.23078v1 Large language models increasingly operate over large collections of tools, functions, APIs, and specialized agents. As the candidate action space grows, a function-calling model must process more schemas, consume more prompt tokens, and distinguish among increasingly similar or irrelevant alternatives. We study a complementary systems strategy: reduce the candidate set before language-model inference while leaving the downstream model unchanged. We introduce AgentWeave, a deterministic pre-inference routing layer that constructs a bounded model-visible action space using eligibility, requirement, capability, and routing signals. We evaluate AgentWeave with a frozen BFCL-derived routing-pressure protocol using the public MadeAgents/Hammer2.1-1.5b model. On 48 fresh BFCL V4 multiple-function tasks, AgentWeave achieves 6/48 (12.5%) native BFCL successes, whereas all-tools, deterministic random top-8, and semantic top-8 baselines each achieve 0/48. The paired success difference is +12.5 percentage points with a 10,000-resample paired bootstrap 95% confidence interval of +4.17 to +22.92 points and exact McNemar p=0.03125. Relative to all-tools exposure, AgentWeave presents 70.18% fewer tools, uses 61.70% fewer input tokens, and exhibits 50.95% lower mean local-model latency. The result is deliberately narrow: this is a BFCL-derived routing-pressure study rather than an official full BFCL leaderboard score, and absolute task success remains low. The evidence nevertheless shows that candidate-space construction can materially affect a fixed model's function-calling behavior and motivates evaluating routing as a distinct stage before model reasoning. --- ### When a neural surrogate cannot accelerate a solver: runtime share, closed-loop drift, and the economics of uncertainty gating in a stiff coupled simulation Authors: L. Thümmler, T. Kuroda Published: 2026-08-24 Categories: astro-ph.IM, astro-ph.HE, cs.LG, physics.comp-ph arXiv: https://arxiv.org/abs/2608.23075v1 Learned surrogates for expensive inner solver blocks are a widely pursued route to faster multiphysics simulation. We report a controlled, end-to-end negative result and identify three structural barriers, none of them a deficiency of the network we trained. The testbed is the implicit Newton solve coupling energy-dependent neutrino radiation to matter in a general-relativistic radiation-hydrodynamics code, its most expensive physics routine per call. First, per-call cost and share of runtime are different quantities, and only the second bounds acceleration. An exclusive self-time profile puts the target block at 16.9% of critical-rank wall clock, capping any surrogate at ~1.2x by Amdahl's law. A surrogate 5.8x cheaper per call merely ties the solver, and the configuration stable enough to run without fallback reaches only parity. Second, offline accuracy cannot rank surrogates for deployment: across fourteen networks the pooled Spearman error-versus-survival correlation (rho = +0.73) is a between-family confound that vanishes under control (rho = -0.04). Third, a correct out-of-distribution gate cannot accelerate a loop that leaves its training distribution. We give the break-even deferral fraction in closed form: because the visited states sit 73x off the data manifold, the gate defers 96.8 to 99.7% of cells, almost invariant to surrogate quality. Including its own cost, the gated loop is a 0.94 to 0.96x slowdown. We further separate stability from fidelity: a never-crashing gated run accumulates a linear -19.9% density bias over 6000 steps. The error is a directed, ballistically accumulating bias, not the variance-driven divergence the autoregressive literature targets. --- ### Grounding Isn't Knowing: Do VLMs Need Object Localization for Spatial Reasoning? Authors: Xiwei Liu, Yulong Li, Xinlin Zhuang, Xuhui Li, Zhixiang Lu, Haolin Yang, Imran Razzak, Yutong Xie Published: 2026-08-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.23074v1 Vision-language models (VLMs) can answer spatial questions, yet the mechanisms connecting object grounding to spatial reasoning remain poorly understood. It is underexplored whether spatial reasoning internally requires precise objects localization, or can bypass explicit localization through global layout cues. In this work, we investigate two representative model families, LLaVA-1.5 and Qwen2.5-VL, using a suite of mechanistic interpretability tools, including token ablation, layer-wise probing, attention knockout, and causal mediation analysis. We find that spatial relation prediction follows a staged grounding-to-reasoning process in which object-aligned tokens establish coarse target-reference anchors, while precise bounding-box boundaries are not required. Positional information becomes decodable before relation decisions emerge, and a small set of attention heads mediates the causal effects of both localization and spatial reasoning. The two tasks share early grounding-related processing but ultimately rely on partially distinct specialized pathways. Through rigorous experiments, we provide a token-, layer-, and head-level account of how VLMs transform object grounding into spatial relations, showing that knowing where objects are is not equivalent to knowing how they relate. --- ### From Generation to Simulation: How Far Are World Models from Being True Simulators? Authors: Tong Wang, Huan Deng, Mucheng Yang, Yang He, Xiaohui Kuang, Gang Zhao Published: 2026-08-24 Categories: cs.AI, cs.CV arXiv: https://arxiv.org/abs/2608.23070v1 With the rapid progress of diffusion models and large-scale video generation, generative world models are increasingly expected to replace traditional simulators, including physics engines, game engines, and reinforcement-learning environments. Yet the remaining distance from generation to simulation lacks a systematic assessment. We present a capability-based study using an external yardstick: eight capabilities of a traditional simulator, namely asset construction, physics engine, interaction, controllability, stability, state feedback, diversity, and evaluation metrics. We trace three main technical routes--latent dynamics, video generation, and joint-embedding prediction--and map exactly 200 representative works published from 2018 to June 2026 onto these capabilities. Our analysis shows that world models have achieved functional substitution in interaction and controllability for specific scenarios, but remain short of traditional simulators in formal guarantees of physical laws, structured state feedback, and reproducible long-horizon evolution. State feedback is the most neglected cross-route shortcoming: only 6 of 163 implementation papers expose a runtime interface for querying entity states or physical parameters. We identify six research directions: formalized physics, a unified action interface, first-class state feedback, long-horizon stability, downstream-utility evaluation, and cross-route hybridization. Project page: https://github.com/AtongWang/world-model-simulators --- ### Signal or Noise? A Benchmark Study of Agent Skills in Web Development Authors: Ziyue Yang, Fan Ding Published: 2026-08-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.23067v1 Agent Skills are reusable procedural modules that are increasingly injected into coding-agent sessions to encode framework conventions, anti-patterns, and reusable tools. However, because each injected Skill expands the prompt of every query, an effective Skill benchmark must determine not only whether an agent can solve a task, but whether the Skill should have been injected at all. We introduce WebDev-Skills-Bench and use it for a controlled empirical study of 31 public WebDev Skills on 50 Web-Bench projects and 1,000 ordered tasks. The benchmark compares four matched conditions, including a length-matched irrelevant control and leave-one-out component ablations. To isolate Skill effects from prompt-length artifacts, we place only SKILL.md in the prompt while mounting auxiliary files into the agent workspace. Across four models, target Skill injection reduces mean Pass@2 by 1.3% to 4.2%, lowers task completion depth, and increases token cost by 72% to 394%, with gains in only 17% to 36% of Skill-project pairs. Length-matched controls reveal two failure modes: some models are length-distracted, where an equally long irrelevant Skill reproduces most of the loss, while others are content-misled, where prompt length is neutral but Skill content still lowers Pass@2 by 1.1% to 1.4%. Further analysis shows that losses concentrate on easy early tasks, Skill rankings transfer weakly across models, and anti-pattern rules outperform example-heavy content within helpful Skills. These findings recast a matched Skill as a hypothesis about a particular Skill-project-model triple rather than a portable asset, reframing injection as a per-deployment routing decision and making length-matched controls and per-model audits a minimum standard for Agent-Skill evaluation. --- ### Cultural Moment Benchmark: Evaluating Video Cultural Reasoning and Grounding in Southeast Asia Authors: Burak Satar, Zhixin Ma, Cheng Yu-Tong, Huy Hoang Tran, Phuong Anh Nguyen, Chong-Wah Ngo Published: 2026-08-24 Categories: cs.CV, cs.AI, cs.CL, cs.IR, cs.MM arXiv: https://arxiv.org/abs/2608.23065v1 Cultural understanding in video means more than recognizing what is visible; it requires grasping the symbolic and temporal significance of cultural concepts. We decompose this into three abilities: naming what a concept symbolizes, visually recognizing it on video, and locating its sub-events in time. Existing video-cultural benchmarks tend to test what is seen, collapsing these three abilities into a single score that hides the bottleneck. We introduce the Cultural Moment Benchmark (CMB): 306 expert-curated concepts from seven countries in Southeast Asia across five categories. We evaluate each concept through three stages, one per ability. Given a description, Stage 1 (S1) selects from four candidate concept names, Stage 2 (S2) selects from four candidate video moments, and Stage 3 (S3) predicts the start and end times of the moment in a video. To keep each stage focused on a distinct ability, we use three design choices: semantic-similarity distractors (S1, S2), unlabeled video moments (S2), and free-form localization on a different example video (S3). Across six vision-language models, failure modes vary by ability and modality. i) Even the strongest closed-source models score below 30% when all three stages must be correct; ii) The three abilities do not fully cascade: naming a concept correctly helps half the models recognize it on video, but recognizing it has little effect on locating the sub-event in time; iii) Audio is complementary, redundant, or distracting depending on the concept, more often distracting in non-Latin-script countries; removing both audio and subtitles hurts Games and Music the most. Our 14-rater human study shows that even Expert raters score below chance on concepts from a neighboring country, indicating that CMB requires country-specific cultural knowledge. CMB acts as a diagnostic harness, attributing failures to a specific ability or modality. --- ### Improving O-RADS Risk Stratification from Ultrasound Reports: A Comparative Evaluation of Hybrid versus End-to-End LLM Reasoning Strategies Authors: Xiaotong Tan, Chunli Qiu, Xin Liu, Qing Huang, Guangli Zhou, Bo Gao, Xiaoyan Song, Shuyan Wang, Xiuqin Wang, Wufeng Xue, Ruobing Huang, Dong Ni et al. Published: 2026-08-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.23061v1 Background: Automating clinical guideline-based decision-making with large language models (LLMs) remains challenging because of reliability, hallucination, and limited interpretability. We compared the performance of LLMs and reasoning strategies for automated Ovarian-Adnexal Reporting and Data System (O-RADS) classification from free-text pelvic ultrasound reports. Methods: In this retrospective study, consecutive patients with ovarian masses who underwent pelvic ultrasound were included. Eight LLMs were tested with three reasoning strategies: implicit-knowledge end-to-end, rule-informed end-to-end, and a feature-based hybrid architecture that decoupled feature extraction from rule-based classification. The reference standard was O-RADS categorization established by expert consensus. Results: A total of 310 women with 390 ovarian masses were evaluated. The feature-based hybrid architecture using Gemini 3.6 Flash demonstrated the best performance, achieving an accuracy of 99.2% (387 of 390) and almost perfect agreement with the reference standard (weighted kappa = 1.00; 95% CI: 0.99-1.00). Its performance surpassed that of original clinical reports (accuracy, 87.7% (342 of 390); weighted kappa = 0.94; 95% CI: 0.91-0.96) and end-to-end LLM strategies (accuracy range, 65.6% (256 of 390) to 95.9% (374 of 390)). For structured feature extraction, Gemini 3.6 Flash demonstrated higher overall accuracy than Claude Fable 5 (98.9% vs 97.8%; P < 0.001). The hybrid architecture reduced misclassification errors and mitigated the overstaging tendency observed in original reports. Conclusion: The feature-based hybrid LLM architecture that separates clinical feature extraction from deterministic guideline execution enables highly accurate, reliable, and interpretable automated O-RADS classification, providing a promising approach for standardized, guideline-based clinical decision-making. --- ### LLM-based Agents for Forecasting and Prediction: Methods, Training, Evaluation, and Applications Authors: Xiaogang Xu, Jiaqi Tang, Jianmin Chen, Yingying Yan, Zhenchao Tang, Xiangxin Zhou, Xiaobin Hu, Wei Wei, Jinfeng Wu, Qifeng Chen, Lu Zhou, Jiafei Wu et al. Published: 2026-08-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.23058v1 Large language models (LLMs) now support forecasting systems that combine language-based reasoning with temporal data, evidence retrieval, external tools, and iterative prediction. We investigate LLM-based forecasting agents, meaning systems in which a language model contributes to a scored prediction about a future or currently unobserved target. We organize architectures into three groups. Standalone LLM workflows operate on encoded time series or event context. Tool- and retrieval-augmented agents incorporate external evidence. Hybrid systems pair LLMs with statistical or foundation models. We then review training methods and evaluation protocols. We examine negative as well as positive evidence, including sensitivity to small input perturbations, ablations in which the LLM component does not improve accuracy, and benchmark gains that may reflect contamination instead of temporal reasoning. We cover applications in finance, weather, health, energy, and operations, and we summarize the benchmarks and datasets used for evaluation. The evidence indicates that measurement is a central limitation. Future work requires calibration under distribution shift, contamination-resistant live evaluation, explicit reporting of cost and accuracy together, and methods for handling feedback between deployed forecasts and the outcomes being forecast. --- ### Macro-Action Topological Navigation under Noisy Localization using Reinforcement Learning Authors: Simon Hakenes, Tobias Glasmachers Published: 2026-08-24 Categories: cs.LG, cs.RO arXiv: https://arxiv.org/abs/2608.23055v1 Navigating large, photorealistic 3D apartments from raw pixels is widely considered infeasible for plain reinforcement learning. We build an agent that does it anyway, estimating its own pose from the camera alone. The agent has to reach several target objects in sequence, and their positions change between episodes, so it must explore to find them. It builds on our earlier object-centric topological controller, which still read the agent's true pose and its object detections from the simulator. Here we replace that true pose with an onboard, object-centric estimate. For each object we keep a bank of ORB features that, when the object is seen again, yield a rough pose measurement, which a minimal Extended Kalman Filter (EKF) fuses with a motion model. As on a real robot, the executed motions are noisy. The estimate drifts, but the agent and the nearby objects drift together, so a locally consistent pose is enough to follow each short edge and then home in visually on the target, which lets us replace full SLAM with a much smaller model, closer to how biological navigation appears to work. In the photorealistic Habitat simulator, the agent reaches its target objects from vision alone, with a pose that only needs to be locally consistent. --- ### Graph Representation Learning of Lightweight IoT Ciphers Authors: Jonathan Cook, Sabih ur Rehman, M. Arif Khan Published: 2026-08-24 Categories: cs.LG, cs.CR arXiv: https://arxiv.org/abs/2608.23054v1 SIMON and SIMECK belong to a family of Lightweight Cryptographic Algorithms (LCAs) based on the Feistel block cipher, designed for Internet of Things (IoT) devices. As with all Feistel ciphers, they are susceptible to differential cryptanalysis, necessitating rigorous resilience evaluations. While state-of-the-art techniques leverage heuristics and sampling to improve efficiency, little work has applied Machine Learning (ML) guided Graph Representation Learning (GRL) to efficiently identify and visualise high-probability differential clusters. We address this gap by introducing an efficient feature engineering strategy that extracts four differential attributes from a partial Difference Distribution Table (pDDT), revealing structural information concealed in raw differential data. Utilising the enriched features, we construct and compare three ML-guided directed graphs for SIMON$32$ and SIMECK$32$ using K-Nearest Neighbour (KNN), Decision Trees (DT), and Random Forests (RF). To the best of our knowledge, our framework produces the first graph-based visualisation of the differential clustering effect, in which high-probability single-bit differentials form geometrically close clusters in the learned embedding. All three models achieve a precision of $1.0$ in identifying high-probability differentials, confirming zero false positives. KNN achieves the strongest cluster separation, the highest F1 score and the lowest graph construction time of approximately $2.3$ seconds, while DT and RF produce optimal paths with near-perfect regression. The results are consistent across both LCAs, demonstrating the applicability of the framework to other AND-rotation LCA families. --- ### Reservoir of Importance: Learning Semi-Structured Sparsity with Differentiable Subset Sampling Authors: Ha Dinh, Xuan Duy Ta, Khoat Than, Khac-Hoai Nam Bui Published: 2026-08-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.23048v1 Semi-structured $N$:$M$ sparsity has emerged as a practical direction for accelerating large language models (LLMs). However, existing learnable-mask approaches incur substantial parameter and memory overhead, limiting their scalability to large models and aggressive sparsity regimes. In this work, we revisit semi-structured pruning from a perspective that reconciles efficiency with scalability. We propose Reservoir of Importance (RoI), a lightweight semi-structured pruning framework that learns sparsity masks through differentiable subset sampling. Unlike prior methods that model full categorical distributions over all feasible $N$:$M$ patterns, RoI introduces a compact-logit parameterization for sparsity mask learning and performs sampling without replacement to select masks, thereby reducing trainable parameters from combinatorial complexity to $\mathcal{O}({M})$. As a result, RoI requires 1.5-8.75$\times$ fewer learnable parameters and significantly lower memory cost, while remaining fully aligned with hardware-friendly sparsity patterns. Extensive evaluations across multiple scales of the Qwen2.5 LLM family (0.5-7B parameters) demonstrate that RoI achieves competitive performance with strong memory efficiency, stability, and scalability to more aggressive $N$:$M$ sparsity patterns, offering a practical path toward efficient LLM deployment. --- ### Beyond Verdicts: A Graph-Based Analysis of Human and LLM Reasoning in Scientific Fact-Checking Authors: Abdul Ghafoor, Muhammad Arslan Manzoor, Yufang Hou Published: 2026-08-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.23047v1 Misinformation that cites legitimate papers can be especially harmful when it distorts what those studies actually report. While existing automatic fact-checking systems based on large language models (LLMs) can assess whether a model assigns an Incorrect verdict and can gen- erate explanations for that decision, they typi- cally do not indicate whether the model follows the same reasoning path as human experts or arrives at the verdict through a different but still valid path. In this work, we introduce a graph- based framework (typed reasoning graph) for comparing human and LLM reasoning paths in scientific fact-checking. Building on prior work on fallacious reasoning in biomedical misinformation, MISSCIPLUS (Glockner et al., 2025), we model each explanation as a rea- soning graph that links the false claim to the relevant study context, study findings, fallacy- supporting premises, and fallacy labels. This representation enables one-to-one alignment of human and LLM reasoning at the level of fallacy-specific sub-graphs. For non-human- aligned LLM paths, we validate grounding in the cited study, relevance to the claim, and suf- ficiency for the verdict. Using 84 false claims from MISSCIPLUS, we evaluate GPT-5, Claude Opus 4.7, and Qwen3-32B across prompt and evidence settings. Results show distinct perfor- mance dimensions: Qwen3-32B has the lowest verdict failure rate, GPT-5 the highest human alignment, and Claude Opus 4.7 weak verdict prediction but often valid reasoning in success- ful cases --- ### From Inertia to Objectivity: Improving Deep Research Agents with Noise Isolation Authors: Xiangxin Zhang, Zhanwei Zhang, Zhihang Fu, Binbin Lin, Wenxiao Wang Published: 2026-08-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.23045v2 Web search agents powered by Large Language Models (LLMs) show strong promise, but deep research tasks expose a recurring failure mode: once an agent has produced a query, plan, or intermediate conclusion, it becomes less objective when later judging the consequences of that same action. We term this phenomenon inertia bias. To make it measurable, we introduce the IBIS benchmark, which controls the search observations while varying whether the model is evaluating the outcome of its own prior action. We find that models are substantially worse when they "own" the preceding search step, showing that self-authored action history can systematically distort subsequent judgment. We further show that this bias propagates into two forms of system-level degradation: search noise at the worker level and contextual noise at the manager level. To address this problem, we propose NIS-Agent, which applies context isolation at the two decision points most vulnerable to inertia bias: webpage triage and final-answer validation. Across GAIA, WebWalkerQA, BrowseComp, and BrowseComp-zh, NIS-Agent achieves competitive performance while reducing token cost by 33% compared to our baseline. We further train an 8B model to be intrinsically more resistant to inertia bias; under the same NIS-Agent framework, it attains average performance comparable to GPT-4o on deep research benchmarks. Our code is publicly available at https://github.com/PangSMPang/NIS-Agent. --- ### AutoSaddler: Automatic Harness Optimization with Durable Updates from Agent Execution Traces Authors: Sungho Park, Wonjoong Kim, Rongyuan Tan, Jue Zhang, Wook-Shin Han, Pengfei Gao, Chanyoung Park, Yongqiang Yao, Rao Fu, Elsie Nallipogu, Qingwei Lin, Saravan Rajmohan et al. Published: 2026-08-24 Categories: cs.AI, cs.CL, cs.LG, cs.MA, cs.SE arXiv: https://arxiv.org/abs/2608.23041v1 LLM agents remain unreliable on long-horizon tasks, where small local failures can compound over extended interactions and lead to overall task failure. Although external harnesses can substantially improve robustness, harness design remains a manual and expensive process that requires searching over a large space of prompts, tool configurations, and control logic. We propose AutoSaddler, an automatic harness optimization framework that formulates harness improvement as an offline learning problem and iteratively updates the harness using failure signals from mini-batches. AutoSaddler combines failure-trace diagnosis, structured patch generation that treats the harness as code, and validation-based update selection. Experiments on GAIA2, SWE-Bench Pro, and Terminal-Bench 2.0 show that AutoSaddler substantially improves agent performance over the corresponding base harnesses, achieving gains of 9.0, 9.6, and 10.0 percentage points, respectively. Ablation studies further suggest that effective harness optimization benefits from three ingredients: deep debugging rather than shallow reflection, targeted modifications rather than unconstrained editing, and generalization-aware selection rather than trajectory-specific repair. Together, these results suggest that automatic harness optimization is a promising path toward more performant and reliable agent systems. --- ### The Multilingual FrameNet Corpus Authors: Beatrice Fiumanò, Nicolas Lazzari, Simone Paolo Ponzetto, Valentina Presutti Published: 2026-08-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.23037v1 This paper introduces the Multilingual FrameNet Corpus (mFNC), a novel resource that extends the English Berkeley FrameNet corpus by collecting and harmonizing existing language-specific corpora across nine additional languages: Brazilian Portuguese, Chinese, Dutch, French, German, Italian, Korean, Latvian and Swedish. By training models that rely on different architectures on the mFNC, we consistently outperform existing state-of-the-art Frame Semantic Parsers in both multilingual and cross-lingual settings, underscoring the importance of multilingual training data. The mFNC and our trained FSP models are openly available at https://github.com/beatrice-f/mFNC. --- ### MobilePA-Bench: Benchmarking Mobile Planner Agents on Complex Real-World Tasks Authors: Yi Zhu, Xiongwei Wu, Qiyi Wang, Tingyu Qu, Jiajun Liu, Sihan Cao, Long Chen, Weigao Sun, Feida Zhu, Yiran Zhong, Steven Hoi Published: 2026-08-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.23035v2 As on-device LLM agents evolve into personal copilots, the mobile operating system has become a key testbed for this paradigm, making rigorous capability evaluation essential. Yet existing benchmarks fall into two camps, each with a critical blind spot: GUI-centric benchmarks test surface-level screen manipulation while overlooking background tool use and long-horizon planning, whereas static function-calling benchmarks rely on offline API matching that is detached from real runtime constraints. To close this gap, we present \textbf{MobilePA-Bench}, an interactive, stateful, and tool-centric benchmark for evaluating the tool-calling and planning abilities of mobile planning agents. MobilePA-Bench runs on an executable sandbox that maintains live application databases and returns structured feedback, spanning $13$ functional domains and $212$ realistic mobile tools. Beyond basic tool use, it evaluates a central planning agent along three advanced dimensions: \emph{(1)~Sub-agent Collaboration}---decomposing a complex task and delegating specialized work to capable sub-agents; \emph{(2)~Memory Usage}---recalling stored memories, user profiles, and past preferences to resolve implicit requests; and \emph{(3)~Skill Usage}---invoking pre-packaged composite skills instead of planning every step from scratch. Extensive experiments show that current frontier LLMs remain unreliable in mobile settings: performance drops sharply under strict tool ordering, permission limits, and unexpected runtime errors. By pairing an interactive function-calling sandbox with evidence-based verification, MobilePA-Bench serves as both a practical diagnostic benchmark and an interactive foundation for agentic reinforcement learning---accelerating the development of dependable mobile agents. --- ### ST$^2$U: Stateful Test-Time Unlearning via Restricted Knowledge Boundary Control Authors: Xunlei Chen, Qinghui Gong, Ruini Xue, Yaodong Hu, Tian Lan, Wenhong Tian Published: 2026-08-24 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2608.23034v1 Controlling restricted knowledge in large language models is essential for model alignment and safe deployment. Test-time unlearning avoids costly retraining and parameter updates by intervening only during inference. However, existing activation-editing methods apply isolated pointwise corrections, overlooking how autoregressive generation continually reconstructs hidden states from the prompt, cache, and generated prefix. Consequently, later states may return to restricted knowledge regions after a locally successful correction, causing restricted knowledge re-entry. In this work, we propose Stateful Test-Time Unlearning via restricted knowledge boundary control (ST$^2$U), which formulates test-time unlearning as trajectory-wide boundary control. ST$^2$U first models restricted knowledge boundaries in low-dimensional invertible coordinates while leaving orthogonal non-target components unchanged. During inference, ST$^2$U monitors risk along the trajectory, applies minimal boundary corrections with contextual anchoring, and propagates historical correction states across tokens to mitigate knowledge re-entry. This trajectory-wide control enables more persistent forgetting while preserving non-target capabilities and limiting inference overhead. Across three benchmarks and three model families, ST$^2$U delivers the strongest overall balance, combining best or second-best retention with competitive forgetting and substantially less restricted-knowledge re-entry than test-time baselines (13.76%-19.84% versus 46.50%-59.10%). --- ### FedCC: Towards Addressing Label Distribution Skews in Distillation-Based Federated Learning Authors: Wenxuan Ye, Onur Ayan, Xueli An, Georg Carle Published: 2026-08-24 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.23031v2 Federated Learning (FL) enables distributed clients to collaboratively train models without sharing raw data, making it promising for leveraging massive devices in communication networks. In distillation-based FL, each client applies its local model on an unlabeled public dataset, and shares only prediction results with the server. While heterogeneous local data introduces label distribution skew, thus biasing client models toward majority classes and leading to potentially inaccurate predictions. The lack of ground-truth labels in the public dataset hampers the server's ability to calibrate predictions, which ultimately degrades overall performance. To address this, we propose FedCC, a simple and effective algorithm for mitigating client misclassification. Instead of being forced to classify and risking error propagation, clients are allowed to tag ambiguous samples as 'unknown'. This additional class, together with calibrated pseudo-labels on the public data, balances confidence in majority classes against uncertainty in under-represented ones. Extensive experiments demonstrate that FedCC significantly outperforms existing methods, especially under severe label skew. In the extreme scenario where each client holds samples from only one of ten classes, FedCC achieves 67.3% accuracy, while baselines collapse to near-random results. --- ### Artificial Empathy: Towards a Framework for Unsupervised Agency Detection and Policy Reconstruction Authors: Peter Kuhn, Chris Pang, Sonakshi Chauhan Published: 2026-08-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.23030v1 We study how an AI system can identify and model other agents in its environment from observation alone, which is a capability necessary for cooperative behaviour in the real world. This problem is less constrained than inverse reinforcement learning and remains largely unexplored. We propose a framework that uses a reinforcement learning agent, trained on an independent task as a prior about agentic dynamics, to perform agency detection and policy reconstruction. --- ### Meta-Moderator: Empowering Multi-Agent Debate with Meta-Cognition Authors: Wentao Hu, Zhuoyue Wan, Jinhao Shen, Chen Jason Zhang, Xiaoyong Wei, Qing Li Published: 2026-08-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.23029v1 Multi-agent debate can improve large language model reasoning by eliciting diverse hypotheses and critiques, yet its performance is often constrained by weak moderation. Common pipelines rely on fixed budgets, agreement-based stopping, or untrained judges, leading to redundant deliberation and unreliable evidence aggregation. We cast moderation as a meta-cognitive process, monitoring debate utility, controlling deliberation, and adjudicating a final answer, and introduce Meta-Moderator, a learnable framework that dynamically regulates debate and decides when to finalize an answer. Meta-Moderator is trained independently of the debaters via outcome-driven policy optimization, making debate regulation an explicit capability rather than an incidental effect of prompting. Across five benchmarks, Meta-Moderator outperforms widely used decision layers and transfers across tasks and system configurations. Further analyses show that it allocates debate more selectively and reduces mis-aggregation after informative hypotheses appear. --- ### PsychJail: Exploring Psychological Jailbreaks via Multi-Turn Persuasion of LLM Policies Authors: Zeyu Feng, Qingyu Wu, Yuzhe Luo, Hua Cheng Published: 2026-08-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.23028v1 Large language models (LLMs) are increasingly deployed in education, healthcare, policy advising, and other interactive settings, where users engage them as sustained social interlocutors rather than one-shot query engines. This shift makes jailbreaks a growing safety threat, yet most research emphasizes single-turn prompt optimization or iterative attack refinement, leaving psychologically grounded multi-turn vulnerabilities underexplored. We present PsychJail, a psychology-guided framework for red teaming aligned LLMs through theory-grounded, multi-turn persuasion. PsychJail maps established social-psychological persuasion techniques into a tactic-conditioned attack policy. It factorizes each attacker action into a Change-of-Meaning analysis, tactic selection, and victim-visible message, operationalizing the Persuasion Knowledge Model (PKM). The policy is refined with trajectory-level reinforcement learning using a PKM-gated reward that credits early jailbreak success only when every turn contains a well-formed Change-of-Meaning analysis. Across four aligned victim models, PsychJail achieves the highest average attack success rate (87.3%) and outperforms strong single-turn and multi-turn baselines on every model. We also measure susceptibility at the action that breaks each victim, revealing four distinct model-level fingerprints that identify which persuasion levers affect each model and how broadly. These fingerprints help explain cross-model transfer asymmetry. We interpret them as four candidate psychological profiles-rationalist, credibility-driven, narrative-monoculture, and broadly persuadable-while treating this interpretation as a conjecture requiring future validation. Our findings establish psychological jailbreaks as a distinct red-teaming frontier for increasingly interactive LLMs. --- ### Beyond Surface Cues: Disentangling Sociocultural Signals in Multilingual LLMs Authors: Yuanjun Feng, Tanzhou Liu, Stefan Feuerriegel, Yash Raj Shrestha Published: 2026-08-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.23026v1 Multilingual LLM outputs can vary across sociocultural contexts. However, evidence of cultural grounding can be misleading: identity labels may be inferred from explicit or indirect textual cues, while names and wording can reveal the source language. Treating all these signals as evidence of cultural grounding may obscure potential biases. We present a human-validated, multi-agent audit that separates three questions: whether outputs reproduce social biases, whether identity groups are represented differently, and whether outputs reflect cross-cultural patterns. The study analyzes 89,253 outputs from 12 LLMs in English, French, and Chinese, spanning 18 occupations and three task conditions. We find that bias representation varies systematically across languages and tasks. Removing direct identity cues sharply reduces identity-label prediction in English and Chinese, but has a much smaller effect in French. Across all language-genre settings, the cultural context associated with the source language receives the highest average relevance score, with moderate agreement between automated and human ratings. However, the ability to identify the source language drops substantially after translation and again after masking names. Without these controls, multilingual audits may mistake surface cues for cultural understanding, leading to misleading conclusions about cross-cultural variation and bias. Our audit offers a practical framework for separating such shortcuts from more meaningful cross-cultural patterns. --- ### When the Edit Changes the Patient: Measuring Identity Preservation in Counterfactual Retinal Images Authors: Andrea Posada, Wenke Karbole, Bach Ngoc Doan, Alexander Weers, Solmaz Abdolrahimzadeh, Maria Patsiamanidi, Kahkashan Haider, Vaishali Khare, Daniel Rueckert, Andrew Lotery, Sobha Sivaprasad, Martin J. Menten Published: 2026-08-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.23024v1 Counterfactual medical image generation aims to modify an existing image to reflect a hypothetical scenario in which certain characteristics of the imaged subject are altered, while keeping their identity fixed. Most existing works repurpose established image editing methods, which do not directly supervise identity preservation. Instead, they assume that identity is implicitly preserved by anchoring generation to the source image. This assumption is rarely tested and may fail in domains where biometric cues are subtle, such as retinal optical coherence tomography (OCT). In this work, we explicitly measure identity preservation for three groups of text-conditioned editing methods - source-anchored, structured-prompt, and paired-training - using referee classifiers, embedding alignment scores, and a blind reader study. We find that all methods produce high-quality OCT images with comparable editing success, yet their identity preservation differs markedly. Source-anchored editing frequently alters the depicted subject, while paired-training preserves it best. We argue that future work on medical counterfactual generation must explicitly measure and report identity preservation alongside image realism and editing success. --- ### Most of the LLM Routing Gap Is Task Type Authors: Janghoon Lee Published: 2026-08-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.23023v2 An LLM router picks which model should answer each query. The appeal is that models fail on different questions. Whatever single model is best overall still gets some wrong, and another model in the pool gets many of those right. Getting that choice right every time is the ceiling, and a router is an attempt to approach it. However, recent work reports that routers do not get close. Across 21 routing methods on five benchmarks, sharply different designs land within a fraction of a point of each other, and all of them stay far below that ceiling. Learned routers often fail to beat simply always calling the strongest model. We ask what those missed questions have in common. We set fourteen models to answer all 294 questions, with 7 task types across 3 languages: Korean, English and Hindi. We ran the whole matrix twice, changing nothing, but 5.37% of the 4,116 model-question pairs came out scored differently anyway. Run-to-run movement like that is normal, and we argue that a small win does not show that routing did anything, ours or anyone else's. Counting an answer correct only when the model got it right in both runs, 29 questions on this matrix can be improved with routing. Every correct-answer count here is on that rule. Task type accounts for most of them: assigning each task type one model in advance, chosen once and never updated, improves 21 of the 29. Splitting each task type by language improves 2 more and leaves 6 of 294 unoptimized. That handful is what a learned router would have been built for, and it is smaller than the run-to-run movement above, which is a share of pairs rather than of questions. The static table we adopted answers 262 of 294 questions at \$3.33 per run, against the best single model's 245 at \$7.69. All of this is fitted and scored on the same 294 questions with no holdout. --- ### Neural Boltzmann Equations Authors: Jonas Spinner, Jack D. Shergold Published: 2026-08-24 Categories: hep-ph, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2608.23022v2 The dynamics of particles in the early universe are described by Boltzmann equations, which involve high-dimensional phase-space integrals. Classical approaches use quadrature integration and evolve the system on a fixed momentum grid, which scales poorly to complicated systems and parameter scans, severely limiting the complexity of processes that can be studied. We introduce Neural Boltzmann Equations (NBEs), which combine three coupled concepts to overcome these limitations. First, particle properties are encoded in physics-inspired neural distribution functions, with parameters that can be predicted using neural networks, enabling efficient parameter scans. Second, phase-space integrals are evaluated with Monte Carlo, using importance sampling tools from collider physics. Third, we use the natural gradient method to evolve the system. After demonstrating the individual benefits of NBEs, we use the framework to perform the most numerically precise calculation to date of the effective number of relativistic neutrino degrees of freedom, $N_\mathrm{eff}$, in the Standard Model. --- ### Unlearning Is Not Just Erasing: Temporal Decoupling via Generation Inequality Authors: Xunlei Chen, Qirui Ye, Yuang Li, Yi Gong, Zhaokun Wang, Wenyi Li, Shiyao Guo, Jinyu Guo Published: 2026-08-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.23020v1 Large language models (LLMs) require effective unlearning to address privacy regulations and safety concerns. However, achieving precise forgetting without compromising general utility remains challenging. Existing sequence- and token-level methods penalize target outputs without modeling their context-dependent retrieval paths, which can disrupt linguistic structure or suppress benign knowledge. We present ADU, a fine-grained, training-based framework that shifts unlearning from token erasure to contextual attention-pathway decoupling. Exploiting the functional distinction between local and global attention heads, ADU identifies preplan positions that retrieve persistent sensitive anchors and fixes their candidate paths under the original model. It then trains attention-projection adapters to suppress attention mass along these paths while preserving local-attention structure and retain-set language modeling. Post-training activation exchange tests whether the modified attention-output module transmits the learned forgetting effect. ADU achieves the strongest aggregate performance among evaluated baselines on the TOFU and WMDP benchmarks, including a Forget Quality of (0.93) on TOFU. It preserves 87--98% of model utility (92.9% on average versus 81.9% for baselines) while reducing side effects in benign contexts. --- ### SplitLite: Low-Rank Residual Compression for Split Learning Authors: Tao Li, Yulin Tang, Qi Guo, Xianhao Chen Published: 2026-08-24 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.23018v2 Federated fine-tuning of on-device large language models (LLMs) faces a significant computing burden. To overcome this limitation, split learning (SL) has emerged as a promising solution, which offloads the primary training workload to a powerful server. However, SL requires exchanging high-dimensional activations and gradients between clients and the server, resulting in prohibitive communication costs. To overcome this challenge, we propose SplitLite, a communication-efficient split federated LoRA fine-tuning method that exploits the low effective rank structure of consecutive-epoch activation and gradient residuals. Our key finding is that, when LoRA uses rank $r$ updates in parameter space, the activation and gradient residuals of the same data sample between adjacent epochs also exhibit effective rank-$2r$ and rank-$4r$ structures, respectively. By revealing this property, SplitLite transmits only quantized truncated singular value decomposition (SVD) residual factors, thereby significantly reducing both activation uplink and gradient downlink traffic. Extensive experiments on the GLUE benchmark across a series of advanced on-device LLMs demonstrate that our method reduces activation uplink communication costs by up to 93.5\% and total communication costs by up to 83.7\%, without performance degradation. --- ### AnaDiffusion: Anatomically CompositionalLatent Diffusion for Controllable 3D Brain MRI Generation Authors: Huiwen Han, Lulin Liu, Bangya Liu, Yuanhao Cai, Nuo Chen, Xiaoqing Wang, Ziqian Xie, Chenyu You, Shuiwang Ji, Degui Zhi, Zhiwen Fan Published: 2026-08-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.23014v2 3D brain MRI generation has made significant advances in medical imaging, simulation, and controllable anatomical analysis. However, existing generative models typically synthesize 3D volumes monolithically, often overlooking regional anatomical structures and limiting local controllability. To address these limitations, we introduce AnaDiffusion, an anatomically compositional latent diffusion framework that factorizes the generation process into distinct, anatomically meaningful regions, followed by part-to-whole assembly and global refinement. Our approach first trains part diffusion models to capture local structural priors. We then inject an assembled anatomical composite of the parts into the whole-brain latent representation and continue denoising. This mechanism enables the model to resolve global context while preserving the injected anatomy. As a result, AnaDiffusion produces both explicit part assets and a globally coherent volume, thereby enabling controllable part editing without requiring subject-specific dense segmentation maps at inference time while maintaining consistent part-to-whole brain structure. On the subject-disjoint ADNI test split, AnaDiffusion achieves the lowest FID across the whole brain, left and right hemispheres, cerebellar-brainstem complex, and seam regions. It also achieves the best cerebellar and second-best ventricular and brainstem absolute Cohen's d values among the evaluated methods. In localized editing experiments, paired MS-SSIM demonstrates high target transfer and off-target preservation, supporting controllable part replacement with minimal unintended anatomical alterations. --- ### Misanthrope: A Privacy-Preserving Keypoint Detector Authors: Francesco Vultaggio, Predrag Djindjic, Markus Gerke, Sebastian Tschiatschek, Phillipp Fanta-Jende Published: 2026-08-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.23012v1 Image matching is a core component of applications such as Simultaneous Localization and Mapping (SLAM), Visual Localization, and Structure from Motion (SfM). However, the local image features central to this task are vulnerable to inversion attacks, which enable adversaries to reconstruct privacy-sensitive scene content from local features. These attacks pose a particular threat in distributed computing scenarios where the pre-computed features leave edge devices to be processed by remote servers. In this work, we introduce Misanthrope, a novel privacy-preserving keypoint detector trained through self-distillation to avoid detecting keypoints on people---a predominant source of privacy-sensitive content in most localization scenarios---thus mitigating inversion attacks at the source rather than through post-hoc obfuscation. We demonstrate how inverted images from traditional feature detection pipelines can be used to detect and re-identify people in the scene, while Misanthrope is able to mitigate these attacks. Furthermore, Misanthrope maintains image matching performance on par with the state of the art and even surpasses it in challenging settings where people act as distractors, such as phototourism and in-the-wild odometry. On the Image Matching Challenge 2021 Phototourism test set, Misanthrope is the top-performing sparse feature extractor in 7 out of 9 scenes. We make our model and its evaluation script available here: https://github.com/fratopa/misanthrope --- ### Coarse Indexing, Fine Evidence: Decoupling Temporal Granularity in Long-Video RAG Authors: Zhe Jin, Zhimin Lin, Bin Zheng, Junhua Fang, Huihua Yang Published: 2026-08-24 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.23011v1 Graph-based retrieval-augmented generation (RAG) provides a scalable paradigm for long-video understanding, but existing systems typically inherit a fixed temporal granularity from video segmentation when constructing their retrieval index. We argue that this design unnecessarily couples indexing granularity with evidence granularity: coarse representations can often suffice for locating relevant temporal regions, while fine-grained evidence remains important for downstream reasoning. We propose \textbf{Density-Aware Graph Construction (DAGC)}, a training-free approach that decouples a query-independent coarse retrieval index from the original fine-grained evidence space. DAGC constructs a compact, density-adaptive graph index by merging visually redundant neighboring chunks, while preserving mappings to the original temporal units. Retrieved coarse regions are subsequently expanded back to the original chunk granularity for fine-grained evidence refinement and answer generation. Experiments on MLVU, VideoMME, and LongVideoBench show that DAGC retains only about 40--50\% of the original graph nodes and achieves $1.3$--$1.7\times$ end-to-end wall-clock acceleration while preserving approximately 99\% of the original QA performance. The gains transfer across different LVLM backbones and video RAG pipelines, suggesting that long-video RAG need not maintain the same temporal granularity for indexing and evidence reasoning. ---