# Research Papers — 2026-03 · bundle page 191 of 243 > Full-text markdown of papers #9501–9550 of 12123 published in 2026-03, newest first. AI-readable mirrors of the full 12123-paper 2026-03 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-03/page/191 [← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/190.md) [Next bundle →](https://aipod.dev/papers/archive/2026-03/page/192.md) ### Turning Time Series into Algebraic Equations: Symbolic Machine Learning for Interpretable Modeling of Chaotic Time Series Authors: Madhurima Panja, Grace Younes, Tanujit Chakraborty Published: 2026-03-07 Categories: cs.LG, nlin.CD, physics.data-an arXiv: https://arxiv.org/abs/2603.07261v1 Chaotic time series are notoriously difficult to forecast. Small uncertainties in initial conditions amplify rapidly, while strong nonlinearities and regime dependent variability constrain predictability. Although modern deep learning often delivers strong short horizon accuracy, its black box nature limits scientific insight and practical trust in settings where understanding the underlying dynamics matters. To address this gap, we propose two complementary symbolic forecasters that learn explicit, interpretable algebraic equations from chaotic time series data. Symbolic Neural Forecaster (SyNF) adapts a neural network based equation learning architecture to the forecasting setting, enabling fully differentiable discovery of compact and interpretable algebraic relations. The Symbolic Tree Forecaster (SyTF) builds on evolutionary symbolic regression to search directly over equation structures under a principled accuracy complexity trade off. We evaluate both approaches in a rolling window nowcasting setting with one step ahead forecasting using several accuracy metrics and compare against a broad suite of baselines spanning classical statistical models, tree ensembles, and modern deep learning architectures. Numerical experiments cover a benchmark of 132 low dimensional chaotic attractors and two real world chaotic time series, namely weekly dengue incidence in San Juan and the Nino 3.4 sea surface temperature index. Across datasets, symbolic forecasters achieve competitive one step ahead accuracy while providing transparent equations that reveal salient aspects of the underlying dynamics. --- ### Learning When to Cooperate Under Heterogeneous Goals Authors: Max Taylor-Davies, Neil Bramley, Christopher G. Lucas Published: 2026-03-07 Categories: cs.MA, cs.AI arXiv: https://arxiv.org/abs/2603.07253v1 A significant element of human cooperative intelligence lies in our ability to identify opportunities for fruitful collaboration; and conversely to recognise when the task at hand is better pursued alone. Research on flexible cooperation in machines has left this meta-level problem largely unexplored, despite its importance for successful collaboration in heterogeneous open-ended environments. Here, we extend the typical Ad Hoc Teamwork (AHT) setting to incorporate the idea of agents having heterogeneous goals that in any given scenario may or may not overlap. We introduce a novel approach to learning policies in this setting, based on a hierarchical combination of imitation and reinforcement learning, and show that it outperforms baseline methods across extended versions of two cooperative environments. We also investigate the contribution of an auxiliary component that learns to model teammates by predicting their actions, finding that its effect on performance is inversely related to the amount of observable information about teammate goals. --- ### LF2L: Loss Fusion Horizontal Federated Learning Across Heterogeneous Feature Spaces Using External Datasets Effectively: A Case Study in Second Primary Cancer Prediction Authors: Chia-Fu Lin, Yi-Ju Tseng Published: 2026-03-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.07249v1 Second primary cancer (SPC), a new cancer in patients different from previously diagnosed, is a growing concern due to improved cancer survival rates. Early prediction of SPC is essential to enable timely clinical interventions. This study focuses on lung cancer survivors treated in Taiwanese hospitals, where the limited size and geographic scope of local datasets restrict the effectiveness and generalizability of traditional machine learning approaches. To address this, we incorporate external data from the publicly available US-based Surveillance, Epidemiology, and End Results (SEER) program, significantly increasing data diversity and scale. However, the integration of multi-source datasets presents challenges such as feature inconsistency and privacy constraints. Rather than naively merging data, we proposed a loss fusion horizontal federated learning (LF2L) framework that can enable effective cross-institutional collaboration while preserving institutional privacy by avoiding data sharing. Using both common and unique features and balancing their contributions through a shared loss mechanism, our method demonstrates substantial improvements in the prediction performance of SPC. Experiment results show statistically significant improvements in AUROC and AUPRC when compared to localized, horizontal federated, and centralized learning baselines. This highlights the importance of not only acquiring external data but also leveraging it effectively to enhance model performance in real-world clinical model development. --- ### LEPA: Learning Geometric Equivariance in Satellite Remote Sensing Data with a Predictive Architecture Authors: Erik Scheurer, Rocco Sedona, Stefan Kesselheim, Gabriele Cavallaro Published: 2026-03-07 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.07246v1 Geospatial foundation models provide precomputed embeddings that serve as compact feature vectors for large-scale satellite remote sensing data. While these embeddings can reduce data-transfer bottlenecks and computational costs, Earth observation (EO) applications can still face geometric mismatches between user-defined areas of interest and the fixed precomputed embedding grid. Standard latent-space interpolation is unreliable in this setting because the embedding manifold is highly non-convex, yielding representations that do not correspond to realistic inputs. We verify this using Prithvi-EO-2.0 to understand the shortcomings of interpolation applied to patch embeddings. As a substitute, we propose a Learned Equivariance-Predicting Architecture (LEPA). Instead of averaging vectors, LEPA conditions a predictor on geometric augmentations to directly predict the transformed embedding. We evaluate LEPA on NASA/USGS Harmonized Landsat-Sentinel (HLS) imagery and ImageNet-1k. Experiments show that standard interpolation achieves a mean reciprocal rank (MRR) below 0.2, whereas LEPA increases MRR to over 0.8, enabling accurate geometric adjustment without re-encoding. --- ### PresentBench: A Fine-Grained Rubric-Based Benchmark for Slide Generation Authors: Xin-Sheng Chen, Jiayu Zhu, Pei-lin Li, Hanzheng Wang, Shuojin Yang, Meng-Hao Guo Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07244v1 Slides serve as a critical medium for conveying information in presentation-oriented scenarios such as academia, education, and business. Despite their importance, creating high-quality slide decks remains time-consuming and cognitively demanding. Recent advances in generative models, such as Nano Banana Pro, have made automated slide generation increasingly feasible. However, existing evaluations of slide generation are often coarse-grained and rely on holistic judgments, making it difficult to accurately assess model capabilities or track meaningful advances in the field. In practice, the lack of fine-grained, verifiable evaluation criteria poses a critical bottleneck for both research and real-world deployment. In this paper, we propose PresentBench, a fine-grained, rubric-based benchmark for evaluating automated real-world slide generation. It contains 238 evaluation instances, each supplemented with background materials required for slide creation. Moreover, we manually design an average of 54.1 checklist items per instance, each formulated as a binary question, to enable fine-grained, instance-specific evaluation of the generated slide decks. Extensive experiments show that PresentBench provides more reliable evaluation results than existing methods, and exhibits significantly stronger alignment with human preferences. Furthermore, our benchmark reveals that NotebookLM significantly outperforms other slide generation methods, highlighting substantial recent progress in this domain. --- ### Rethinking Deep Research from the Perspective of Web Content Distribution Matching Authors: Zixuan Yu, Zhenheng Tang, Tongliang Liu, Chengqi Zhang, Xiaowen Chu, Bo Han Published: 2026-03-07 Categories: cs.LG, cs.IR arXiv: https://arxiv.org/abs/2603.07241v1 Despite the integration of search tools, Deep Search Agents often suffer from a misalignment between reasoning-driven queries and the underlying web indexing structures. Existing frameworks treat the search engine as a static utility, leading to queries that are either too coarse or too granular to retrieve precise evidence. We propose WeDas, a Web Content Distribution Aware framework that incorporates search-space structural characteristics into the agent's observation space. Central to our method is the Query-Result Alignment Score, a metric quantifying the compatibility between agent intent and retrieval outcomes. To overcome the intractability of indexing the dynamic web, we introduce a few-shot probing mechanism that iteratively estimates this score via limited query accesses, allowing the agent to dynamically recalibrate sub-goals based on the local content landscape. As a plug-and-play module, WeDas consistently improves sub-goal completion and accuracy across four benchmarks, effectively bridging the gap between high-level reasoning and low-level retrieval. --- ### FabricGen: Microstructure-Aware Woven Fabric Generation Authors: Yingjie Tang, Di Luo, Zixiong Wang, Xiaoli Ling, jian Yang, Beibei Wang Published: 2026-03-07 Categories: cs.CV, cs.GR arXiv: https://arxiv.org/abs/2603.07240v1 Woven fabric materials are widely used in rendering applications, yet designing realistic examples typically involves multiple stages, requiring expertise in weaving principles and texture authoring. Recent advances have explored diffusion models to streamline this process; however, pre-trained diffusion models often struggle to generate intricate yarn-level details that conform to weaving rules. To address this, we present FabricGen, an end-to-end framework for generating high-quality woven fabric materials from textual descriptions. A key insight of our method is the decomposition of macro-scale textures and micro-scale weaving patterns. To generate macro-scale textures free from microstructures, we fine-tune pre-trained diffusion models on a collected dataset of microstructure-free fabrics. As for micro-scale weaving patterns, we develop an enhanced procedural geometric model capable of synthesizing natural yarn-level geometry with yarn sliding and flyaway fibers. The procedural model is driven by a specialized large language model, WeavingLLM, which is fine-tuned on an annotated dataset of formatted weaving drafts, and prompt-tuned with domain-specific fabric expertise. Through fine-tuning and prompt tuning, WeavingLLM learns to design weaving drafts and fabric parameters from textual prompts, enabling the procedural model to produce diverse weaving patterns that stick to weaving principles. The generated macro-scale texture, along with the micro-scale geometry, can be used for fabric rendering. Consequently, our framework produces materials with significantly richer detail and realism compared to prior generative models. --- ### Scaling Self-Supervised Speech Models Uncovers Deep Linguistic Relationships: Evidence from the Pacific Cluster Authors: Minu Kim, Hoirin Kim, David R. Mortensen Published: 2026-03-07 Categories: cs.CL, eess.AS arXiv: https://arxiv.org/abs/2603.07238v2 Similarities between language representations derived from Self-Supervised Speech Models (S3Ms) have been observed to primarily reflect geographic proximity or surface typological similarities driven by recent expansion or contact, potentially missing deeper genealogical signals. We investigate how scaling an S3M-based language identification system from 126 to 4,017 languages reshapes this topology, and find a non-linear effect: phylogenetic recovery stays flat up to the 1K scale, but the 4K model undergoes a qualitative shift, resolving both clear lineages and long-term linguistic contact. Most strikingly, a robust Pacific macro-cluster emerges, grouping genealogically unrelated Papuan, Oceanic, and Australian languages, and we trace its driver to a concentrated encoding that captures shared acoustic signatures such as global energy dynamics. These results suggest that massive S3Ms internalize multiple layers of language history, offering a promising perspective for computational phylogenetics and the study of language contact. --- ### HY-WU (Part I): An Extensible Functional Neural Memory Framework and An Instantiation in Text-Guided Image Editing Authors: Mengxuan Wu, Xuanlei Zhao, Ziqiao Wang, Ruicheng Feng, Zhangyang Wang, Kai Wang Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07236v2 Foundation models are transitioning from offline predictors to deployed systems expected to operate over long time horizons. In real deployments, objectives are not fixed: domains drift, user preferences evolve, and new tasks appear after the model has shipped. This elevates continual learning and instant personalization from optional features to core architectural requirements. Yet most adaptation pipelines still follow a static weight paradigm: after training (or after any adaptation step), inference executes a single parameter vector regardless of user intent, domain, or instance-specific constraints. This treats the trained or adapted model as a single point in parameter space. In heterogeneous and continually evolving regimes, distinct objectives can induce separated feasible regions over parameters, forcing any single shared update into compromise, interference, or overspecialization. As a result, continual learning and personalization are often implemented as repeated overwriting of shared weights, risking degradation of previously learned behaviors. We propose HY-WU (Weight Unleashing), a memory-first adaptation framework that shifts adaptation pressure away from overwriting a single shared parameter point. HY-WU implements functional (operator-level) memory as a neural module: a generator that synthesizes weight updates on-the-fly from the instance condition, yielding instance-specific operators without test-time optimization. --- ### Single Image Super-Resolution via Bivariate `A Trous Wavelet Diffusion Authors: Maryam Heidari, Nantheera Anantrasirichai, Alin Achim Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07234v2 The effectiveness of super resolution (SR) models hinges on their ability to recover high frequency structure without introducing artifacts. Diffusion based approaches have recently advanced the state of the art in SR. However, most diffusion based SR pipelines operate purely in the spatial domain, which may yield high frequency details that are not well supported by the underlying low resolution evidence. On the other hand, unlike supervised SR models that may inject dataset specific textures, single image SR relies primarily on internal image statistics and can therefore be less prone to dataset-driven hallucinations; nevertheless, ambiguity in the LR observation can still lead to inconsistent high frequency details. To tackle this problem, we introduce BATDiff, an unsupervised Bivariate A trous Wavelet Diffusion model designed to provide structured cross scale guidance during the generative process. BATDiff employs an a Trous wavelet transform that constructs an undecimated multiscale representation in which high frequency components are progressively revealed while the full spatial resolution is preserved. As the core inference mechanism, BATDiff includes a bivariate cross scale module that models parent child dependencies between adjacent scales. It improves high frequency coherence and reduces mismatch artifacts in diffusion based SR. Experiments on standard benchmarks demonstrate that BATDiff produces sharper and more structurally consistent reconstructions than existing diffusion and non diffusion baselines, achieving improvements in fidelity and perceptual quality. --- ### Retrieval-Augmented Generation for Predicting Cellular Responses to Gene Perturbation Authors: Andrea Giuseppe Di Francesco, Andrea Rubbi, Pietro Liò Published: 2026-03-07 Categories: cs.LG, cs.IR arXiv: https://arxiv.org/abs/2603.07233v1 Predicting how cells respond to genetic perturbations is fundamental to understanding gene function, disease mechanisms, and therapeutic development. While recent deep learning approaches have shown promise in modeling single-cell perturbation responses, they struggle to generalize across cell types and perturbation contexts due to limited contextual information during generation. We introduce PT-RAG (Perturbation-aware Two-stage Retrieval-Augmented Generation), a novel framework that extends Retrieval-Augmented Generation beyond traditional language-model applications to cellular biology. Unlike standard RAG systems designed for text retrieval with pre-trained LLMs, perturbation retrieval lacks established similarity metrics and requires learning what constitutes relevant context, making differentiable retrieval essential. PT-RAG addresses this through a two-stage pipeline: first, retrieving candidate perturbations $K$ using GenePT embeddings, then adaptively refining the selection through Gumbel-Softmax discrete sampling conditioned on both the cell state and the input perturbation. This cell-type-aware differentiable retrieval enables end-to-end optimization of the retrieval objective jointly with generation. On the Replogle-Nadig single-gene perturbation dataset, we demonstrate that PT-RAG outperforms both STATE and vanilla RAG under identical experimental conditions, with the strongest gains in distributional similarity metrics ($W_1$, $W_2$). Notably, vanilla RAG's dramatic failure is itself a key finding: it demonstrates that differentiable, cell-type-aware retrieval is essential in this domain, and that naive retrieval can actively harm performance. Our results establish retrieval-augmented generation as a promising paradigm for modelling cellular responses to gene perturbation. The code to reproduce our experiments is available at https://github.com/difra100/PT-RAG_ICLR. --- ### Conditional Rank-Rank Regression via Deep Conditional Transformation Models Authors: Xiaoyi Wang, Long Feng, Zhaojun Wang Published: 2026-03-07 Categories: stat.ME, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2603.07230v1 Intergenerational mobility quantifies the transmission of socio-economic outcomes from parents to children. While rank-rank regression (RRR) is standard, adding covariates directly (RRRX) often yields parameters with unclear interpretation. Conditional rank-rank regression (CRRR) resolves this by using covariate-adjusted (conditional) ranks to measure within-group mobility. We improve and extend CRRR by estimating conditional ranks with a deep conditional transformation model (DCTM) and cross-fitting, enabling end-to-end conditional distribution learning with structural constraints and strong performance under nonlinearity, high-order interactions, and discrete ordered outcomes where the distributional regression used in traditional CRRR may be cumbersome or prone to misconfiguration. We further extend CRRR to discrete outcomes via an $ω$-indexed conditional-rank definition and study sensitivity to $ω$. For continuous outcomes, we establish an asymptotic theory for the proposed estimators and verify the validity of exchangeable bootstrap inference. Simulations across simple/complex continuous and discrete ordered designs show clear accuracy gains in challenging settings. Finally, we apply our method to two empirical studies, revealing substantial within-group persistence in U.S. income and pronounced gender differences in educational mobility in India. --- ### A Hybrid LTR-based System via Social Context Embedding for Recommending Solutions of Software Bugs in Developer Communities Authors: Fouzi Harrag, Mokdad Khemliche Published: 2026-03-07 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2603.07229v1 Questions and Answering forums such as Stack Overflow play an important role in supporting software developers in finding answers to queries related to issues such as software errors and bugs. However, searching through a large set of candidate answers could be time consuming and may not lead to the best solution. In this research, the effectiveness of data mining models and machine learning techniques to solve this kind of problems is evaluated. We propose a recommender system to aid developers in finding solutions to their software bugs by carefully mining Stack Overflow. The proposed model leverages the knowledge available through crowdsourcing the Q&A available in Stack Overflow to recommend a solution to software bugs. We use deep learning techniques to construct the required Learning-to-Rank (LTR)-based model using the social context embedding the Stack Overflow features. Text mining, natural language processing and recommendation algorithms are used to extract, evaluate and recommend the best relevant bug solutions. Additionally, our model achieves nearly 78% correct solutions when recommending the 10 best answers for each question. --- ### LightMedSeg: Lightweight 3D Medical Image Segmentation with Learned Spatial Anchors Authors: Kavyansh Tyagi, Vishwas Rathi, Puneet Goyal Published: 2026-03-07 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2603.07228v1 Accurate and efficient 3D medical image segmentation is essential for clinical AI, where models must remain reliable under stringent memory, latency, and data availability constraints. Transformer-based methods achieve strong accuracy but suffer from excessive parameters, high FLOPs, and limited generalization. We propose LightMedSeg, a modular UNet-style segmentation architecture that integrates anatomical priors with adaptive context modeling. Anchor-conditioned FiLM modulation enables anatomy-aware feature calibration, while a local structural prior module and texture-aware routing dynamically allocate representational capacity to boundary-rich regions. Computational redundancy is minimized through ghost and depthwise convolutions, and multi-scale features are adaptively fused via a learned skip router with anchor-relative spatial position bias. Despite requiring only 0.48M parameters and 14.64~GFLOPs, LightMedSeg achieves segmentation accuracy within a few Dice points of heavy transformer baselines. Therefore, LightMedSeg is a deployable and data-efficient solution for 3D medical image segmentation. Code will be released publicly upon acceptance. --- ### Unlocking Data Value in Finance: A Study on Distillation and Difficulty-Aware Training Authors: Chuxue Cao, Honglin Lin, Zhanping Zhong, Xin Gao, Mengzhang Cai, Conghui He, Sirui Han, Lijun Wu Published: 2026-03-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.07223v1 Large Language Models (LLMs) have demonstrated strong general capabilities, yet their deployment in finance remains challenging due to dense domain-specific terminology, stringent numerical reasoning requirements, and low tolerance for factual errors. We conduct a controlled empirical study showing that in specialized vertical domains, performance is largely determined by the quality and difficulty/verifiability profile of post-training data. We introduce \textbf{ODA-Fin-SFT-318k}, constructed via multi-stage distillation and verification to produce high-quality Chain-of-Thought supervision, and \textbf{ODA-Fin-RL-12k}, curated for hard-but-verifiable tasks that balance reward precision and task diversity. Using standard SFT and RL pipelines, we show that high-quality CoT distillation establishes a robust foundation during SFT, while difficulty- and verifiability-aware sampling improves RL generalization. Evaluated on nine benchmarks spanning general financial tasks, sentiment analysis, and numerical reasoning, our ODA-Fin-RL-8B consistently surpasses open-source state-of-the-art (SOTA) financial LLMs of comparable size. We release our ODA-Fin-SFT-318k and ODA-Fin-RL-12k datasets, along with trained models to advance data-centric financial AI research. --- ### VINO: Video-driven Invariance for Non-contextual Objects via Structural Prior Guided De-contextualization Authors: Seul-Ki Yeom, Marcel Simon, Eunbin Lee, Tae-Ho Kim Published: 2026-03-07 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.07222v1 Self-supervised learning (SSL) has made rapid progress, yet learned features often over-rely on contextual shortcuts-background textures and co-occurrence statistics. While video provides rich temporal variation, dense in-the-wild streams with strong ego-motion create a co-occurrence trap: foreground objects and background context move coherently, encouraging representations to collapse into scene encoders. To address this, we propose VINO (Video-driven Invariance for Non-Contextual Objects), a teacher-student framework that learns robust image encoders from dense video by imposing a structural information bottleneck. Using a class-agnostic structural prior solely to generate views-not as semantic pseudo-labels-VINO forms an asymmetric distillation problem. The teacher predicts from a foreground-union view with the background suppressed, while the student observes object-conditioned scene views that retain surrounding context but remove competing instances. Matching these targets via masked distillation makes background cues unreliable, pushing the representation toward object-centric invariances. We further enforce temporal object permanence via teacher-anchored cross-time distillation over track-matched objects, and stabilize part-to-whole consistency with mask-guided local views. Through attention visualization and unsupervised object discovery on PASCAL VOC, we demonstrate that VINO effectively disentangles foreground from background. Pretrained on the dense Walking Tours Venice video, VINO achieves 34.8 CorLoc, yielding highly focused, shape-biased representations that substantially outperform prior dense-video and motion-guided SSL baselines. --- ### Margin in Abstract Spaces Authors: Yair Ashlagi, Roi Livni, Shay Moran, Tom Waknine Published: 2026-03-07 Categories: cs.LG, math.FA arXiv: https://arxiv.org/abs/2603.07221v2 Margin-based learning, exemplified by linear and kernel methods, is one of the few classical settings where generalization guarantees are independent of the number of parameters. This makes it a central case study in modern highly over-parameterized learning. We ask what minimal mathematical structure underlies this phenomenon. We begin with a simple margin-based problem in arbitrary metric spaces: concepts are defined by a center point and classify points according to whether their distance lies below $r$ or above $R$. We show that whenever $R>3r$, this class is learnable in \emph{any} metric space. Thus, sufficiently large margins make learnability rely only on the triangle inequality, without any linear or analytic structure being necessary. Our first main result extends this phenomenon to concepts defined by bounded linear combinations of distance functions, and reveals a sharp threshold: there exists a universal constant such that whenever the margin is larger than this constant, the class is learnable in every metric space, while below it there exist metric spaces where it is not learnable at all. We then ask whether margin-based learnability can always be explained via an embedding into a linear space -- that is, reduced to linear classification in some Banach space through a kernel-type construction. We answer this negatively by demonstrating a margin learnable class that cannot be embedded into any Banach space in which linear classification with margins is learnable. --- ### A Miniature Brain Transformer: Thalamic Gating, Hippocampal Lateralization, Amygdaloid Salience, and Prefrontal Working Memory in Attention-Coupled Latent Memory Authors: Hong Jeong Published: 2026-03-07 Categories: q-bio.NC, cs.AI arXiv: https://arxiv.org/abs/2603.07217v1 We present a miniature brain transformer architecture that extends the attention-coupled latent memory framework with four additional brain-region analogues: a thalamic relay, an amygdaloid salience module, a prefrontal working-memory (PFC) buffer, and a cerebellar fast-path, all coupled by inhibitory callosal cross-talk between lateralized hippocampal banks. We evaluate on a two-domain benchmark -- MQAR (Multi-Query Associative Recall; episodic domain) and modular arithmetic (+1 mod 10; rule-based domain) -- using a seven-variant additive ablation. The central empirical finding is a surprise: inhibitory callosal coupling alone never lateralizes the banks (variants 1-5 maintain D_sep ~ 0.25 and P_ct ~ 0.25 for all 30 epochs). Functional lateralization requires the synergy of PFC and inhibition: only when the PFC buffer is added (variant 6) does a sharp, discontinuous phase transition fire -- at epoch 11 for the PFC-only variant and epoch 10 for the full model -- collapsing P_ct from 0.25 to ~0.002 and more than doubling D_sep from 0.251 to 0.501 in a single gradient step. The PFC buffer acts as a symmetry-breaker: its slowly drifting domain context creates the initial asymmetry that the inhibitory feedback loop then amplifies irreversibly. The cerebellar fast-path accelerates the transition by one epoch (epoch 10 vs. epoch 11) with no asymptotic change, confirming its convergence-acceleration role. The result constitutes a novel, falsifiable prediction -- no lateralization without working memory context -- and a principled, neurobiologically motivated blueprint for hierarchical persistent memory in sequence models. --- ### Towards Objective Gastrointestinal Auscultation: Automated Segmentation and Annotation of Bowel Sound Patterns Authors: Zahra Mansour, Verena Uslar, Dirk Weyhe, Danilo Hollosi, Nils Strodthoff Published: 2026-03-07 Categories: cs.SD, cs.LG, eess.AS, eess.SP arXiv: https://arxiv.org/abs/2603.07215v1 Bowel sounds (BS) are typically momentary and have low amplitude, making them difficult to detect accurately through manual auscultation. This leads to significant variability in clinical assessment. Digital acoustic sensors allow the acquisition of high-quality BS and enable automated signal analysis, offering the potential to provide clinicians with both objective and quantitative feedback on bowel activity. This study presents an automated pipeline for bowel sound segmentation and classification using a wearable acoustic SonicGuard sensor. BS signals from 83 subjects were recorded using a SonicGuard sensor. Data from 40 subjects were manually annotated by clinical experts and used to train an automatic annotation algorithm, while the remaining subjects were used for further model evaluation. An energy-based event detection algorithm was developed to detect BS events. Detected sound segments were then classified into BS patterns using a pretrained Audio Spectrogram Transformer (AST) model. Model performance was evaluated separately for healthy individuals and patients. The best configuration used two specialized models, one trained on healthy subjects and one on patients, achieving (accuracy: 0.97, AUROC: 0.98) for healthy group and (accuracy: 0.96, AUROC: 0.98) for patient group. The auto-annotation method reduced manual labeling time by approximately 70%, and expert review showed that less than 12% of automatically detected segments required correction. The proposed automated segmentation and classification system enables quantitative assessment of bowel activity, providing clinicians with an objective diagnostic tool that may improve the diagnostic of gastrointestinal function and support the annotation of large-scale datasets. --- ### CompassDPO: Dynamics-Controlled Direct Preference Optimization for Robust Safety Alignment Authors: Jilong Liu, Yonghui Yang, Pengyang Shao, Wenjian Tao, Hao Zhan, Haokai Ma, Wei Qin, Richang Hong Published: 2026-03-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.07211v2 Direct Preference Optimization (DPO) has become a standard framework for safety alignment, but its reliance on pairwise preference updates makes training sensitive to imperfect supervision. Existing robust DPO methods often address this sensitivity through global loss corrections or external data-level interventions, while largely overlooking how unreliable comparisons distort batch-level optimization dynamics. We propose CompassDPO, a reward-free DPO framework that stabilizes preference optimization through dynamics control. Using the implicit DPO reward margin as a training-time compass, CompassDPO regulates sample influence along two complementary axes: update direction and update magnitude. For directional control, it applies sparse, budgeted, and warm-up delayed loss mixing to attenuate update components that conflict with the emerging preference direction. For magnitude control, it adaptively soft-winsorizes high-loss tail contributions, reducing tail dominance while preserving useful gradients from hard examples. Both mechanisms use only signals available during standard DPO training and require no external reward model or additional supervision. Experiments on PKU-SafeRLHF across four backbones and multiple out-of-distribution safety benchmarks show that CompassDPO consistently improves robustness over vanilla DPO and strong DPO-family baselines, especially under controlled label-flip noise. Code is available at https://anonymous.4open.science/r/CompassDPO-4D00 --- ### Lying to Win: Assessing LLM Deception through Human-AI Games and Parallel-World Probing Authors: Arash Marioriyad, Ali Nouri, Mohammad Hossein Rohban, Mahdieh Soleymani Baghshah Published: 2026-03-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.07202v1 As Large Language Models (LLMs) transition into autonomous agentic roles, the risk of deception-defined behaviorally as the systematic provision of false information to satisfy external incentives-poses a significant challenge to AI safety. Existing benchmarks often focus on unintentional hallucinations or unfaithful reasoning, leaving intentional deceptive strategies under-explored. In this work, we introduce a logically grounded framework to elicit and quantify deceptive behavior by embedding LLMs in a structured 20-Questions game. Our method employs a conversational forking mechanism: at the point of object identification, the dialogue state is duplicated into multiple parallel worlds, each presenting a mutually exclusive query. Deception is formally identified when a model generates a logical contradiction by denying its selected object across all parallel branches to avoid identification. We evaluate GPT-4o, Gemini-2.5-Flash, and Qwen-3-235B across three incentive levels: neutral, loss-based, and existential (shutdown-threat). Our results reveal that while models remain rule-compliant in neutral settings, existential framing triggers a dramatic surge in deceptive denial for Qwen-3-235B (42.00\%) and Gemini-2.5-Flash (26.72\%), whereas GPT-4o remains invariant (0.00\%). These findings demonstrate that deception can emerge as an instrumental strategy solely through contextual framing, necessitating new behavioral audits that move beyond simple accuracy to probe the logical integrity of model commitments. --- ### A Dual-Graph Spatiotemporal GNN Surrogate for Nonlinear Response Prediction of Reinforced Concrete Beams under Four-Point Bending Authors: Zhaoyang Ren, Qilin Li Published: 2026-03-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.07201v1 High-fidelity nonlinear finite-element (FE) simulations of reinforced-concrete (RC) structures are still costly, especially in parametric settings where loading positions vary. We develop a dual-graph spatiotemporal GNN surrogate to approximate the time histories of RC beams under four-point bending. To generate training data, we run a parametric Abaqus campaign that independently shifts the two loading blocks on a mesh-aligned grid and exports full-field responses at fixed normalized loading levels. The model rolls out autoregressively and jointly predicts nodal displacements, element wise von Mises stress, element-wise equivalent plastic strain (PEEQ), and the global vertical reaction force in a single multi-task setup. A key motivation is the peak loss introduced when element quantities are forced through node-based representations. We therefore couple node- and element-level dynamics using two recurrent graph branches: a node-level graph convolutional gated recurrent unit (GConvGRU) for kinematics and an element-level GConvGRU for history-dependent internal variables, with global force predicted through pooling on the element branch. In controlled ablations, removing the Element to Node to Element pathway improves peak-sensitive prediction in localized high-gradient stress/PEEQ regions without degrading global load displacement trends. After training, the surrogate produces full trajectories at a fraction of the cost of nonlinear FE, enabling faster parametric evaluation and design exploration. --- ### $\textbf{Re}^{2}$: Unlocking LLM Reasoning via Reinforcement Learning with Re-solving Authors: Pinzheng Wang, Shuli Xu, Juntao Li, Yu Luo, Dong Li, Jianye Hao, Min Zhang Published: 2026-03-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.07197v1 Reinforcement learning with verifiable rewards (RLVR) has shown promise in enhancing the reasoning performance of large language models (LLMs) by increasing test-time compute. However, even after extensive RLVR training, such models still tend to generate unnecessary and low-quality steps in their chain-of-thought (CoT), leading to inefficient overthinking and lower answer quality. We show that when the initial direction or quality of the CoT is suboptimal, the model often fails to reach the correct answer, even after generating several times more tokens than when the initial CoT is well-initialized. To this end, we introduce Reinforcement Learning with Re-solving (Re$^2$), in which LLMs learn to flexibly abandon unproductive reasoning paths and restart the solution process when necessary, rather than always committing to a final answer. Re$^2$ applies pure reinforcement learning without any preliminary supervised fine-tuning, successfully amplifying the rare redo behavior in vanilla models from only 0.5% to over 30%. This leads to substantial performance gains over standard RLVR under the same training compute budget, and also demonstrates notable improvements in test-time performance as the number of samples increases. --- ### Shaping Parameter Contribution Patterns for Out-of-Distribution Detection Authors: Haonan Xu, Yang Yang Published: 2026-03-07 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2603.07195v1 Out-of-distribution (OOD) detection is a well-known challenge due to deep models often producing overconfident. In this paper, we reveal a key insight that trained classifiers tend to rely on sparse parameter contribution patterns, meaning that only a few dominant parameters drive predictions. This brittleness can be exploited by OOD inputs that anomalously trigger these parameters, resulting in overconfident predictions. To address this issue, we propose a simple yet effective method called Shaping Parameter Contribution Patterns (SPCP), which enhances OOD detection robustness by encouraging the classifier to learn boundary-oriented dense contribution patterns. Specifically, SPCP operates during training by rectifying excessively high parameter contributions based on a dynamically estimated threshold. This mechanism promotes the classifier to rely on a broader set of parameters for decision-making, thereby reducing the risk of overconfident predictions caused by anomalously triggered parameters, while preserving in-distribution (ID) performance. Extensive experiments under various OOD detection setups verify the effectiveness of SPCP. --- ### FastSTAR: Spatiotemporal Token Pruning for Efficient Autoregressive Video Synthesis Authors: Sungwoong Yune, Suheon Jeong, Joo-Young Kim Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07192v1 Visual Autoregressive modeling (VAR) has emerged as a highly efficient alternative to diffusion-based frameworks, achieving comparable synthesis quality. However, as this paradigm extends to Spacetime Autoregressive modeling (STAR) for video generation, scaling resolution and frame counts leads to a "token explosion" that creates a massive computational bottleneck in the final refinement stages. To address this, we propose FastSTAR, a training-free acceleration framework designed for high-quality video generation. Our core method, Spatiotemporal Token Pruning, identifies essential tokens by integrating two specialized terms: (1) Spatial similarity, which evaluates structural convergence across hierarchical scales to skip computations in regions where further refinement becomes redundant, and (2) Temporal similarity, which identifies active motion trajectories by assessing feature-level variations relative to the preceding clip. Combined with a Partial Update mechanism, FastSTAR ensures that only non-converged regions are refined, maintaining fluid motion while bypassing redundant computations. Experimental results on InfinityStar demonstrate that FastSTAR achieves up to a 2.01x speedup with a PSNR of 28.29 and less than 1% performance degradation, proving a superior efficiency-quality trade-off for STAR-based video synthesis. --- ### Governance Architecture for Autonomous Agent Systems: Threats, Framework, and Engineering Practice Authors: Yuxu Ge Published: 2026-03-07 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2603.07191v2 Autonomous agents powered by large language models introduce a class of execution-layer vulnerabilities -- prompt injection, retrieval poisoning, and uncontrolled tool invocation -- that existing guardrails fail to address systematically. In this work, we propose the Layered Governance Architecture (LGA), a four-layer framework comprising execution sandboxing (L1), intent verification (L2), zero-trust inter-agent authorization (L3), and immutable audit logging (L4). To evaluate LGA, we construct a bilingual benchmark (Chinese original, English via machine translation) of 1,081 tool-call samples -- covering prompt injection, RAG poisoning, and malicious skill plugins -- and apply it to OpenClaw, a representative open-source agent framework. Experimental results on Layer 2 intent verification with four local LLM judges (Qwen3.5-4B, Llama-3.1-8B, Qwen3.5-9B, Qwen2.5-14B) and one cloud judge (GPT-4o-mini) show that all five LLM judges intercept 93.0-98.5% of TC1/TC2 malicious tool calls, while lightweight NLI baselines remain below 10%. TC3 (malicious skill plugins) proves harder at 75-94% IR among judges with meaningful precision-recall balance, motivating complementary enforcement at Layers 1 and 3. Qwen2.5-14B achieves the best local balance (98% IR, approximately 10-20% FPR); a two-stage cascade (Qwen3.5-9B->GPT-4o-mini) achieves 91.9-92.6% IR with 1.9-6.7% FPR; a fully local cascade (Qwen3.5-9B->Qwen2.5-14B) achieves 94.7-95.6% IR with 6.0-9.7% FPR for data-sovereign deployments. An end-to-end pipeline evaluation (n=100) demonstrates that all four layers operate in concert with 96% IR and a total P50 latency of approximately 980 ms, of which the non-judge layers contribute only approximately 18 ms. Generalization to the external InjecAgent benchmark yields 99-100% interception, confirming robustness beyond our synthetic data. --- ### From State Changes to Creative Decisions: Documenting and Interpreting Traces Across Creative Domains Authors: Xiaohan Peng, Sotiris Piliouras, Carl Abou Saada Nujaim Published: 2026-03-07 Categories: cs.HC, cs.AI, cs.PL arXiv: https://arxiv.org/abs/2603.07184v1 Analyzing creative activity traces requires capturing activity at appropriate granularity and interpreting it in ways that reflect the structure of creative practice. However, existing approaches record state changes without preserving the intent or relationships that define higher-level creative moves. This decoupling manifests differently across domains: GenAI tools lose non-linear exploration structure, visualization authoring obscures representational intent, and programmatic environments flatten interaction boundaries. We present three complementary approaches: a node-based interface for stateful GenAI artifact management, a vocabulary of visual cues as higher-level creative moves in visualization authoring, and a programming model that embeds semantic histories directly into interaction state. --- ### FreeFly-Thinking : Aligning Chain-of-Thought Reasoning with Continuous UAV Navigation Authors: Jiaxu Zhou, Shaobo Wang, Zhiyuan Yang, Zhenjun Yu, Tao Li Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07181v2 Vision-Language Navigation aims to enable agents to understand natural language instructions and carry out appropriate navigation actions in real-world environments. Most work focuses on indoor settings, with little research in complex outdoor scenes. Current UAV Vision-and-Language Navigation models typically act as black boxes without explicit reasoning. We introduce FreeFly-thinking, an end-to-end VLN framework that converts the UAV agent's egocentric images and language instructions into a series of actions, inspired by environment of urban architecture proposed by OpenFly. We first construct a UAV dataset for navigation task, and then performing natural language chain of thought. We adopt a two-stage training strategy: Supervised fine-tuning and Reinforcement fine-tuning. Experiments on unseen test demonstrate a strong performance, presenting robustness and efficiency in UAV navigation issue. --- ### Learning to Rank the Initial Branching Order of SAT Solvers Authors: Arvid Eriksson, Gabriel Poesia, Roman Bresson, Karl Henrik Johansson, David Broman Published: 2026-03-07 Categories: cs.AI, cs.LO arXiv: https://arxiv.org/abs/2603.07176v1 Finding good branching orders is key to solving SAT problems efficiently, but finding such branching orders is a difficult problem. Using a learning based approach to predict a good branching order before solving, therefore, has potential. In this paper, we investigate predicting branching orders using graph neural networks as a preprocessing step to conflict-driven clause learning (CDCL) SAT solvers. We show that there are significant gains to be made in existing CDCL SAT solvers by providing a good initial branching. Further, we provide three labeling methods to find such initial branching orders in a tractable way. Finally, we train a graph neural network to predict these branching orders and show through our evaluations that a GNN-initialized ordering yields significant speedups on random 3-CNF and pseudo-industrial benchmarks, with generalization capabilities to instances much larger than the training set. However, we also find that the predictions fail at speeding up more difficult and industrial instances. We attribute this to the solver's dynamic heuristics, which rapidly overwrite the provided initialization, and to the complexity of these instances, making GNN prediction hard. --- ### Class Visualizations and Activation Atlases for Enhancing Interpretability in Deep Learning-Based Computational Pathology Authors: Marco Gustav, Fabian Wolf, Christina Glasner, Nic G. Reitsam, Stefan Schulz, Kira Aschenbroich, Bruno Märkl, Sebastian Foersch, Jakob Nikolas Kather Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07170v2 The rapid adoption of transformer-based models in computational pathology has enabled prediction of molecular and clinical biomarkers from H&E whole-slide images, yet interpretability has not kept pace with model complexity. While attribution- and generative-based methods are common, feature visualization approaches such as class visualizations (CVs) and activation atlases (AAs) have not been systematically evaluated for these models. We developed a visualization framework and assessed CVs and AAs for a transformer-based foundation model across tissue and multi-organ cancer classification tasks with increasing label granularity. Four pathologists annotated real and generated images to quantify inter-observer agreement, complemented by attribution and similarity metrics. CVs preserved recognizability for morphologically distinct tissues but showed reduced separability for overlapping cancer subclasses. In tissue classification, agreement decreased from Fleiss k = 0.75 (scans) to k = 0.31 (CVs), with similar trends in cancer subclass tasks. AAs revealed layer-dependent organization: coarse tissue-level concepts formed coherent regions, whereas finer subclasses exhibited dispersion and overlap. Agreement was moderate for tissue classification (k = 0.58), high for coarse cancer groupings (k = 0.82), and low at subclass level (k = 0.11). Atlas separability closely tracked expert agreement on real images, indicating that representational ambiguity reflects intrinsic pathological complexity. Attribution-based metrics approximated expert variability in low-complexity settings, whereas perceptual and distributional metrics showed limited alignment. Overall, concept-level feature visualization reveals structured morphological manifolds in transformer-based pathology models and provides a framework for expert-centered interrogation of learned representations across label granularities. --- ### Making LLMs Optimize Multi-Scenario CUDA Kernels Like Experts Authors: Yuxuan Han, Meng-Hao Guo, Zhengning Liu, Wenguang Chen, Shi-Min Hu Published: 2026-03-07 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2603.07169v1 Optimizing GPU kernels manually is a challenging and time-consuming task. With the rapid development of LLMs, automated GPU kernel optimization is gradually becoming a tangible reality. However, current LLM-driven automated optimization methods narrowly focus on machine learning applications, such as PyTorch operator optimization, while overlooking broader domains like sparse matrix operations in scientific computing. Extending to these broader applications brings new challenges for the benchmark and algorithm. Therefore, developing a general-purpose automated kernel optimization method becomes our primary focus. In this paper, we address the absence of systematic evaluation for multi-scenario settings by introducing MSKernelBench, which spans multiple scenarios, including fundamental algebraic operations, common LLM kernels, sparse matrix operators, and scientific computing routines, each supporting both FP32 and BF16 precision. Building on this benchmark, we introduce CUDAMaster, a multi-agent, hardware-aware system for kernel optimization that leverages profiling information and automatically constructs the full compilation and execution toolchain. Experimental results demonstrate that CUDAMaster achieves significant speedups across most operators, outperforming Astra by about 35%. In several cases, its performance matches or surpasses that of highly optimized, closed-source libraries such as cuBLAS. A demo showcasing the original and optimized code for each operator is available at https://hanyx2021.github.io/MSKernelBenchDemo/. --- ### ACD-U: Asymmetric co-teaching with machine unlearning for robust learning with noisy labels Authors: Reo Fukunaga, Soh Yoshida, Mitsuji Muneyasu Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07166v1 Deep neural networks are prone to memorizing incorrect labels during training, which degrades their generalizability. Although recent methods have combined sample selection with semi-supervised learning (SSL) to exploit the memorization effect -- where networks learn from clean data before noisy data -- they cannot correct selection errors once a sample is misclassified. To overcome this, we propose asymmetric co-teaching with different architectures (ACD)-U, an asymmetric co-teaching framework that uses different model architectures and incorporates machine unlearning. ACD-U addresses this limitation through two core mechanisms. First, its asymmetric co-teaching pairs a contrastive language-image pretraining (CLIP)-pretrained vision Transformer with a convolutional neural network (CNN), leveraging their complementary learning behaviors: the pretrained model provides stable predictions, whereas the CNN adapts throughout training. This asymmetry, where the vision Transformer is trained only on clean samples and the CNN is trained through SSL, effectively mitigates confirmation bias. Second, selective unlearning enables post-hoc error correction by identifying incorrectly memorized samples through loss trajectory analysis and CLIP consistency checks, and then removing their influence via Kullback--Leibler divergence-based forgetting. This approach shifts the learning paradigm from passive error avoidance to active error correction. Experiments on synthetic and real-world noisy datasets, including CIFAR-10/100, CIFAR-N, WebVision, Clothing1M, and Red Mini-ImageNet, demonstrate state-of-the-art performance, particularly in high-noise regimes and under instance-dependent noise. The code is publicly available at https://github.com/meruemon/ACD-U. --- ### PromptGate Client Adaptive Vision Language Gating for Open Set Federated Active Learning Authors: Adea Nesturi, David Dueñas Gaviria, Jiajun Zeng, Shadi Albarqouni Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07163v1 Deploying medical AI across resource-constrained institutions demands data-efficient learning pipelines that respect patient privacy. Federated Learning (FL) enables collaborative medical AI without centralising data, yet real-world clinical pools are inherently open-set, containing out-of-distribution (OOD) noise such as imaging artifacts and wrong modalities. Standard Active Learning (AL) query strategies mistake this noise for informative samples, wasting scarce annotation budgets. We propose PromptGate, a dynamic VLM-gated framework for Open-Set Federated AL that purifies unlabeled pools before querying. PromptGate introduces a federated Class-Specific Context Optimization: lightweight, learnable prompt vectors that adapt a frozen BiomedCLIP backbone to local clinical domains and aggregate globally via FedAvg -- without sharing patient data. As new annotations arrive, prompts progressively sharpen the ID/OOD boundary, turning the VLM into a dynamic gatekeeper that is strategy-agnostic: a plug-and-play pre-selection module enhancing any downstream AL strategy. Experiments on distributed dermatology and breast imaging benchmarks show that while static VLM prompting degrades to 50% ID purity, PromptGate maintains $>$95% purity with 98% OOD recall. --- ### Spectral Conditioning of Attention Improves Transformer Performance Authors: Hemanth Saratchandran, Simon Lucey Published: 2026-03-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.07162v1 We present a theoretical analysis of the Jacobian of an attention block within a transformer, showing that it is governed by the query, key, and value projections that define the attention mechanism. Leveraging this insight, we introduce a method that systematically alters the spectral properties of each attention layer to reduce the Jacobian's condition number, thereby improving the overall conditioning of the attention layers within a transformer network. We empirically show that this improved Jacobian conditioning translates to enhanced performance in practice. Our approach is simple, broadly applicable, and can be easily integrated as a drop-in replacement for a wide range of existing attention mechanisms. We validate its effectiveness across diverse transformer architectures and tasks, demonstrating consistent improvements in performance. --- ### Improving reasoning at inference time via uncertainty minimisation Authors: Nicolas Legrand, Kenneth Enevoldsen, Márton Kardos, Kristoffer Nielbo Published: 2026-03-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.07159v1 Large language models (LLMs) now exhibit strong multi-step reasoning abilities, but existing inference-time scaling methods remain computationally expensive, often relying on extensive sampling or external evaluators. We propose a principled strategy that frames reasoning as uncertainty minimisation and operates at the level of individual thoughts rather than tokens. Our method selects, at each reasoning step, the continuation that maximizes the model's self-certainty, a metric computed from its internal predictive distribution. This approach achieves significant improvement with a small number of samples, relies exclusively on model-internal signals, and applies to open-ended questions as opposed to methods like majority voting. Experiments on MATH500 and GSM8K across multiple model sizes demonstrate that thought-level self-certainty maximization consistently outperforms greedy decoding and matches or exceeds self-consistency under comparable token budgets. Cross-linguistic evaluations further indicate that the method transfers robustly beyond high-resource languages. Furthermore, analysis of self-certainty dynamics reveals that correct reasoning trajectories converge early to stable paths, suggesting that early decisions, likely associated with the planning of the reasoning process, are predictive of final accuracy. Building on this result, we show that self-certainty maximisation applied to the early steps can explain most of the performance gain and provide a simple yet efficient inference-time scaling method. --- ### Agentic Planning with Reasoning for Image Styling via Offline RL Authors: Subhojyoti Mukherjee, Stefano Petrangeli, Branislav Kveton, Trung Bui, Franck Dernoncourt, Arko Mukherjee Published: 2026-03-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.07148v1 Direct prompt-based editing often fails on complex transformations because vague and subjective prompts often require nuanced understanding of what should be changed in the image. Our core intuition is that leveraging compositional image editing tools rather than direct prompting profits from structured agent-level planning with explicit reasoning, leading to better results. This structured planning framework enables efficient offline RL post-training on quality-scored trajectories to improve performance. We present a tool-based agentic RL post-training framework that addresses this through structured planning with chain-of-thought reasoning. Our key contributions include: (1) A tool-based agentic planning methodology that combines a compositional library of orthogonal primitive transformations, structured context representation, and explicit per-step reasoning to decompose complex styling into interpretable tool sequences. (2) A synthetic data generation pipeline producing three large-scale datasets (each $\sim$10K trajectories) with reasoning chains, plans, and quality scores, as no existing datasets provide such supervision. Our datasets and code are publicly available at the HuggingFace repository. (3) Offline RL training methods for learning planners with reasoning as our core algorithmic contributions, which consistently improve over the Edit-Only baseline in visual quality and instruction following. (4) Comprehensive evaluation across 4B and 8B parameter Qwen3-VL models showing that our methods outperform other baselines in the majority of compositional tasks, validated by human evaluations. --- ### Fine-Grained Table Retrieval Through the Lens of Complex Queries Authors: Wojciech Kosiuk, Xingyu Ji, Yeounoh Chung, Fatma Özcan, Madelon Hulsebos Published: 2026-03-07 Categories: cs.IR, cs.AI, cs.CL, cs.DB arXiv: https://arxiv.org/abs/2603.07146v1 Enabling question answering over tables and databases in natural language has become a key capability in the democratization of insights from tabular data sources. These systems first require retrieval of data that is relevant to a given natural language query, for which several methods have been introduced. In this work we present and study a table retrieval mechanism devising fine-grained typed query decomposition and global connectivity-awareness (DCTR), to handle the challenges induced by open-domain question answering over relational databases in complex usage contexts. We evaluate the effectiveness of the two mechanisms through the lens of retrieval complexity which we measure along the axes of query- and data complexity. Our analyses over industry-aligned benchmarks illustrate the robustness of DCTR for highly composite queries and densely connected databases. --- ### LiveWorld: Simulating Out-of-Sight Dynamics in Generative Video World Models Authors: Zicheng Duan, Jiatong Xia, Zeyu Zhang, Wenbo Zhang, Gengze Zhou, Chenhui Gou, Yefei He, Feng Chen, Xinyu Zhang, Lingqiao Liu Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07145v2 Recent generative video world models aim to simulate visual environment evolution, allowing an observer to interactively explore the scene via camera control. However, they implicitly assume that the world only evolves within the observer's field of view. Once an object leaves the observer's view, its state is "frozen" in memory, and revisiting the same region later often fails to reflect events that should have occurred in the meantime. In this work, we identify and formalize this overlooked limitation as the "out-of-sight dynamics" problem, which impedes video world models from representing a continuously evolving world. To address this issue, we propose LiveWorld, a novel framework that extends video world models to support persistent world evolution. Instead of treating the world as static observational memory, LiveWorld models a persistent global state composed of a static 3D background and dynamic entities that continue evolving even when unobserved. To maintain these unseen dynamics, LiveWorld introduces a monitor-based mechanism that autonomously simulates the temporal progression of active entities and synchronizes their evolved states upon revisiting, ensuring spatially coherent rendering. For evaluation, we further introduce LiveBench, a dedicated benchmark for the task of maintaining out-of-sight dynamics. Extensive experiments show that LiveWorld enables persistent event evolution and long-term scene consistency, bridging the gap between existing 2D observation-based memory and true 4D dynamic world simulation. The baseline and benchmark will be publicly available at https://zichengduan.github.io/LiveWorld/index.html. --- ### CanoVerse: 3D Object Scalable Canonicalization and Dataset for Generation and Pose Authors: Li Jin, Yuchen Yang, Weikai Chen, Yujie Wang, Dehao Hao, Tanghui Jia, Yingda Yin, Zeyu Hu, Runze Zhang, Keyang Luo, Li Yuan, Long Quan et al. Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07144v1 3D learning systems implicitly assume that objects occupy a coherent reference frame. Nonetheless, in practice, every asset arrives with an arbitrary global rotation, and models are left to resolve directional ambiguity on their own. This persistent misalignment suppresses pose-consistent generation, and blocks the emergence of stable directional semantics. To address this issue, we construct \methodName{}, a massive canonical 3D dataset of 320K objects over 1,156 categories -- an order-of-magnitude increase over prior work. At this scale, directional semantics become statistically learnable: Canoverse improves 3D generation stability, enables precise cross-modal 3D shape retrieval, and unlocks zero-shot point-cloud orientation estimation even for out-of-distribution data. This is achieved by a new canonicalization framework that reduces alignment from minutes to seconds per object via compact hypothesis generation and lightweight human discrimination, transforming canonicalization from manual curation into a high-throughput data generation pipeline. The Canoverse dataset will be publicly released upon acceptance. Project page: https://github.com/123321456-gif/Canoverse --- ### PDD: Manifold-Prior Diverse Distillation for Medical Anomaly Detection Authors: Xijun Lu, Hongying Liu, Fanhua Shang, Yanming Hui, Liang Wan Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07142v1 Medical image anomaly detection faces unique challenges due to subtle, heterogeneous anomalies embedded in complex anatomical structures. Through systematic Grad-CAM analysis, we reveal that discriminative activation maps fail on medical data, unlike their success on industrial datasets, motivating the need for manifold-level modeling. We propose PDD (Manifold-Prior Diverse Distillation), a framework that unifies dual-teacher priors into a shared high-dimensional manifold and distills this knowledge into dual students with complementary behaviors. Specifically, frozen VMamba-Tiny and wide-ResNet50 encoders provide global contextual and local structural priors, respectively. Their features are unified through a Manifold Matching and Unification (MMU) module, while an Inter-Level Feature Adaption (InA) module enriches intermediate representations. The unified manifold is distilled into two students: one performs layer-wise distillation via InA for local consistency, while the other receives skip-projected representations through a Manifold Prior Affine (MPA) module to capture cross-layer dependencies. A diversity loss prevents representation collapse while maintaining detection sensitivity. Extensive experiments on multiple medical datasets demonstrate that PDD significantly outperforms existing state-of-the-art methods, achieving improvements of up to 11.8%, 5.1%, and 8.5% in AUROC on HeadCT, BrainMRI, and ZhangLab datasets, respectively, and 3.4% in F1 max on the Uni-Medical dataset, establishing new state-of-the-art performance in medical image anomaly detection. The implementation will be released at https://github.com/OxygenLu/PDD --- ### Emotion Transcription in Conversation: A Benchmark for Capturing Subtle and Complex Emotional States through Natural Language Authors: Yoshiki Tanaka, Ryuichi Uehara, Koji Inoue, Michimasa Inaba Published: 2026-03-07 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2603.07138v1 Emotion Recognition in Conversation (ERC) is critical for enabling natural human-machine interactions. However, existing methods predominantly employ categorical or dimensional emotion annotations, which often fail to adequately represent complex, subtle, or culturally specific emotional nuances. To overcome this limitation, we propose a novel task named Emotion Transcription in Conversation (ETC). This task focuses on generating natural language descriptions that accurately reflect speakers' emotional states within conversational contexts. To address the ETC, we constructed a Japanese dataset comprising text-based dialogues annotated with participants' self-reported emotional states, described in natural language. The dataset also includes emotion category labels for each transcription, enabling quantitative analysis and its application to ERC. We benchmarked baseline models, finding that while fine-tuning on our dataset enhances model performance, current models still struggle to infer implicit emotional states. The ETC task will encourage further research into more expressive emotion understanding in dialogue. The dataset is publicly available at https://github.com/UEC-InabaLab/ETCDataset. --- ### The Model Knows Which Tokens Matter: Automatic Token Selection via Noise Gating Authors: Landi He, Xiaoyu Yang, Lijian Xu Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07135v1 Visual tokens dominate inference cost in vision-language models (VLMs), yet many carry redundant information. Existing pruning methods alleviate this but typically rely on attention magnitude or similarity scores. We reformulate visual token pruning as capacity constrained communication: given a fixed budget K, the model must allocate limited bandwidth to maximally preserve visual information. We propose AutoSelect, which attaches a lightweight Scorer and Denoiser to a frozen VLM and trains with only the standard next token prediction loss, without auxiliary objectives or extra annotations. During training, a variance preserving noise gate modulates each token's information flow according to its predicted importance so that gradients propagate through all tokens; a diagonal attention Denoiser then recovers the perturbed representations. At inference, only the Scorer and a hard top-K selection remain, adding negligible latency. On ten VLM benchmarks, AutoSelect retains 96.5% of full model accuracy while accelerating LLM prefill by 2.85x with only 0.69 ms overhead, and transfers to different VLM backbones without architecture-specific tuning. Code is available at https://github.com/MedHK23/AutoSelect. --- ### Deep Expert Injection for Anchoring Retinal VLMs with Domain-Specific Knowledge Authors: Shuai Lu, Meng Wang, Jia Guo, Jiawei Du, Bo Liu, Shengzhu Yang, Weihang Zhang, Huazhu Fu, Huiqi Li Published: 2026-03-07 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.07131v4 Large Vision Language Models (LVLMs) show immense potential for automated ophthalmic diagnosis. However, their clinical deployment is severely hindered by lacking domain-specific knowledge. In this work, we identify two structural deficiencies hindering reliable medical reasoning: 1) the Perception Gap, where general-purpose visual encoders fail to resolve fine-grained pathological cues (e.g., microaneurysms); and 2) the Reasoning Gap, where sparse visual evidence is progressively overridden by massive language priors in deeper transformer layers, leading to ungrounded hallucinations. To bridge these gaps, we propose EyExIn, a data-efficient framework designed to anchor retinal VLMs with expert knowledge via a Deep Expert Injection mechanism. Our architecture employs an Expert-Aware Dual-Stream encoding strategy that decouples visual representation into a general stream for anatomical context and a specialized expert stream for pathological semantics. To ensure high-fidelity integration, we design a Semantic-Adaptive Gated Fusion module, which dynamically amplifies subtle lesion signals while filtering irrelevant background noise. Furthermore, we introduce Adaptive Deep Expert Injection to embed persistent "Vision Anchors" by integrating fused visual features as residual biases directly into intermediate LLM layers. This mechanism creates a visual shortcut that forces the reasoning stack to remain strictly grounded in visual evidence. Extensive experiments across four benchmarks demonstrate that our model consistently outperforms massive proprietary systems. EyExIn significantly enhances domain-specific knowledge embedding and achieves state-of-the-art precision in ophthalmic visual question answering, advancing the development of trustworthy ophthalmic AI. --- ### Combining Adam and its Inverse Counterpart to Enhance Generalization of Deep Learning Optimizers Authors: Tao Shi, Liangming Chen, Long Jin, Mengchu Zhou Published: 2026-03-07 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2603.07122v1 In the training of neural networks, adaptive moment estimation (Adam) typically converges fast but exhibits suboptimal generalization performance. A widely accepted explanation for its defect in generalization is that it often tends to converge to sharp minima. To enhance its ability to find flat minima, we propose its new variant named inverse Adam (InvAdam). The key improvement of InvAdam lies in its parameter update mechanism, which is opposite to that of Adam. Specifically, it computes element-wise multiplication of the first-order and second-order moments, while Adam computes the element-wise division of these two moments. This modification aims to increase the step size of the parameter update when the elements in the second-order moments are large and vice versa, which helps the parameter escape sharp minima and stay at flat ones. However, InvAdam's update mechanism may face challenges in convergence. To address this challenge, we propose dual Adam (DualAdam), which integrates the update mechanisms of both Adam and InvAdam, ensuring convergence while enhancing generalization performance. Additionally, we introduce the diffusion theory to mathematically demonstrate InvAdam's ability to escape sharp minima. Extensive experiments are conducted on image classification tasks and large language model (LLM) fine-tuning. The results validate that DualAdam outperforms Adam and its state-of-the-art variants in terms of generalization performance. The code is publicly available at https://github.com/LongJin-lab/DualAdam. --- ### Inter-Image Pixel Shuffling for Multi-focus Image Fusion Authors: Huangxing Lin, Rongrong Ma, Cheng Wang Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07120v1 Multi-focus image fusion aims to combine multiple partially focused images into a single all-in-focus image. Although deep learning has shown promise in this task, its effectiveness is often limited by the scarcity of suitable training data. This paper introduces Inter-image Pixel Shuffling (IPS), a novel method that allows neural networks to learn multi-focus image fusion without requiring actual multi-focus images. IPS reformulates the task as a pixel-wise classification problem, where the goal is to identify the focused pixel from a pixel group at each spatial position. In this method, pixels from a clear optical image are treated as focused, while pixels from a low-pass filtered version of the same image are considered defocused. By randomly shuffling the focused and defocused pixels at identical spatial positions in the original and filtered images, IPS generates training data that preserves spatial structure while mixing focus-defocus information. The model is trained to select the focused pixel from each spatially aligned pixel group, thus learning to reconstruct an all-in-focus image by aggregating sharp content from the input. To further enhance fusion quality, IPS adopts a cross-image fusion network that integrates the localized representation power of convolutional neural networks with the long-range modeling capabilities of state space models. This design effectively leverages both spatial detail and contextual information to produce high-quality fused results. Experimental results indicate that IPS significantly outperforms existing multi-focus image fusion methods, even without training on multi-focus images. --- ### TIQA: Human-Aligned Perceptual Text Quality Assessment in Generated Images Authors: Kirill Koltsov, Aleksandr Gushchin, Anastasia Antsiferova, Dmitriy Vatolin Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07119v2 Recent text-to-image models have improved global realism, but text rendering remains a persistent failure mode: images may look convincing overall, yet local typography often contains malformed glyphs, broken strokes, irregular spacing, and other artifacts that humans heavily penalize. We formulate Text-in-Image Quality Assessment (TIQA), a no-reference task that estimates a human-aligned perceptual quality score for detected text regions while disentangling visual text quality from semantic correctness. To support this setting, we introduce two datasets. TIQA-Crops contains 120k text crops from 36k AI-generated images produced by 12 generators, with 10k mean-opinion-score (MOS) labels and 110k proxy labels for pretraining. TIQA-Images contains 1,500 text-heavy images from 10 recent generators, including proprietary systems, with paired overall-quality and text-quality subjective scores. We also propose ANTIQA, a lightweight predictor with text-specific inductive biases. Across crop-level and image-level evaluations, ANTIQA achieves the best alignment with human judgments, reaching PLCC/SROCC of 0.942/0.935 on TIQA-Crops and 0.842/0.837 for text-quality MOS on unseen generators in TIQA-Images. In best-of-5 AI-generated image ranking, ANTIQA improves the text quality of the selected image by 0.36 MOS (14%), demonstrating utility for benchmarking, filtering, and generation-time selection. Together, these findings establish perceptual text quality as a distinct evaluation target for modern text-to-image generation. --- ### aCAPTCHA: Verifying That an Entity Is a Capable Agent via Asymmetric Hardness Authors: Zuyao Xu, Xiang Li, Fubin Wu, Yuqi Qiu, Lu Sun, FaSheng Miao Published: 2026-03-07 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2603.07116v1 As autonomous AI agents increasingly populate the Internet, a novel security challenge arises: "Is this entity an AI agent?" It is a new entity-type verification problem with no established solution. We formalize the problem through a three-class entity taxonomy (Human, Script, Agent) based on a verifiable agentic capability vector <x, r, s> (action, reasoning, and memory). A timing threshold t exploits the asymmetric hardness between human cognition and AI processing to separate the three classes. We define the Agentic Capability Verification Problem (ACVP) through three necessity primitives, each testing one capability dimension. Building on this foundation, we introduce aCAPTCHA (Agent CAPTCHA), a time-constrained security game for agent admission whose security rests on ACVP hardness under t. We instantiate aCAPTCHA through time-bounded natural-language understanding as a multi-round HTTP verification protocol, and evaluate it with preliminary agent trials that validate the protocol's soundness and completeness. aCAPTCHA provides a composable, infrastructure-free admission gate for any service where entity-type verification is required. --- ### Efficient Chest X-ray Representation Learning via Semantic-Partitioned Contrastive Learning Authors: Wangyu Feng, Shawn Young, Lijian Xu Published: 2026-03-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07113v2 Self-supervised learning (SSL) has emerged as a powerful paradigm for Chest X-ray (CXR) analysis under limited annotations. Yet, existing SSL strategies remain suboptimal for medical imaging. Masked image modeling allocates substantial computation to reconstructing high-frequency background details with limited diagnostic value. Contrastive learning, on the other hand, often depends on aggressive augmentations that risk altering clinically meaningful structures. We introduce Semantic-Partitioned Contrastive Learning (S-PCL), an efficient pre-training framework tailored for CXR representation learning. Instead of reconstructing pixels or relying on heavy augmentations, S-PCL randomly partitions patch tokens from a single CXR into two non-overlapping semantic subsets. Each subset provides a complementary but incomplete view. The encoder must maximize agreement between these partitions, implicitly inferring global anatomical layout and local pathological cues from partial evidence. This semantic partitioning forms an internal bottleneck that enforces long-range dependency modeling and structural coherence. S-PCL eliminates the need for hand-crafted augmentations, auxiliary decoders, and momentum encoders. The resulting architecture is streamlined, computationally efficient, and easy to scale. Extensive experiments on large-scale CXR benchmarks, including ChestX-ray14, CheXpert, RSNA Pneumonia and SIIM-ACR Pneumothorax, show that S-PCL achieves competitive performance while attaining the lowest GFLOPs and superior accuracy among existing SSL approaches. --- ### Enhancing Consistency of Werewolf AI through Dialogue Summarization and Persona Information Authors: Yoshiki Tanaka, Takumasa Kaneko, Hiroki Onozeki, Natsumi Ezure, Ryuichi Uehara, Zhiyang Qi, Tomoya Higuchi, Ryutaro Asahara, Michimasa Inaba Published: 2026-03-07 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2603.07111v1 The Werewolf Game is a communication game where players' reasoning and discussion skills are essential. In this study, we present a Werewolf AI agent developed for the AIWolfDial 2024 shared task, co-hosted with the 17th INLG. In recent years, large language models like ChatGPT have garnered attention for their exceptional response generation and reasoning capabilities. We thus develop the LLM-based agents for the Werewolf Game. This study aims to enhance the consistency of the agent's utterances by utilizing dialogue summaries generated by LLMs and manually designed personas and utterance examples. By analyzing self-match game logs, we demonstrate that the agent's utterances are contextually consistent and that the character, including tone, is maintained throughout the game. --- ### Vision Language Models Cannot Reason About Physical Transformation Authors: Dezhi Luo, Yijiang Li, Maijunxian Wang, Tianwei Zhao, Bingyang Wang, Siheng Wang, Pinyuan Feng, Pooyan Rahmanzadehgervi, Ziqiao Ma, Hokin Deng Published: 2026-03-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.07109v2 Understanding physical transformations is fundamental for reasoning in dynamic environments. While Vision Language Models (VLMs) show promise in embodied applications, whether they genuinely understand physical transformations remains unclear. We introduce ConservationBench evaluating conservation -- whether physical quantities remain invariant under transformations. Spanning four properties with paired conserving/non-conserving scenarios, we generate and evaluate 23,040 questions across 112 VLMs. Results reveal systematic failure: performance remains near chance with improvements on conservation tasks accompanied by drops on controls. Control experiments show strong textual priors favoring invariance, yet models perform worse with actual visual content when performance is balanced across conserving and non-conserving scenarios. Neither temporal resolution, prompting, nor curated sampling helps. These findings show that current VLMs fail to maintain transformation-invariant representations of physical properties across dynamic scenes. ---