# Research Papers — 2026-04 · bundle page 134 of 233 > Full-text markdown of papers #6651–6700 of 11612 published in 2026-04, newest first. AI-readable mirrors of the full 11612-paper 2026-04 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-04/page/134 [← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/133.md) [Next bundle →](https://aipod.dev/papers/archive/2026-04/page/135.md) ### Universality of first-order methods on random and deterministic matrices Authors: Nicola Gorini, Chris Jones, Dmitriy Kunisky, Lucas Pesenti Published: 2026-04-13 Categories: math.PR, cs.DS, cs.LG, math.ST arXiv: https://arxiv.org/abs/2604.11729v1 General first-order methods (GFOM) are a flexible class of iterative algorithms which update a state vector by matrix-vector multiplications and entrywise nonlinearities. A long line of work has sought to understand the large-n dynamics of GFOM, mostly focusing on "very random" input matrices and the approximate message passing (AMP) special case of GFOM whose state is asymptotically Gaussian. Yet, it has long remained unknown how to construct iterative algorithms that retain this Gaussianity for more structured inputs, or why existing AMP algorithms can be as effective for some deterministic matrices as they are for random matrices. We analyze diagrammatic expansions of GFOM via the limiting traffic distribution of the input matrix, the collection of all limiting values of permutation-invariant polynomials in the matrix entries, to obtain the following results: 1. We calculate the traffic distribution for the first non-trivial deterministic matrices, including (minor variants of) the Walsh-Hadamard and discrete sine and cosine transform matrices. This determines the limiting dynamics of GFOM on these inputs, resolving parts of longstanding conjectures of Marinari, Parisi, and Ritort (1994). 2. We design a new AMP iteration which unifies several previous AMP variants and generalizes to new input types, whose limiting dynamics are Gaussian conditional on some latent random variables. The asymptotic dynamics hold for a large and natural class of traffic distributions (encompassing both random and deterministic input matrices) and the algorithm's analysis gives a simple combinatorial interpretation of the Onsager correction, answering questions posed recently by Wang, Zhong, and Fan (2022). --- ### The Devil is in the Details -- From OCR for Old Church Slavonic to Purely Visual Stemma Reconstruction Authors: Armin Hoenen Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.11724v1 The age of artificial intelligence has brought many new possibilities and pitfalls in many fields and tasks. The devil is in the details, and those come to the fore when building new pipelines and executing small practical experiments. OCR and stemmatology are no exception. The current investigation starts comparing a range of OCR-systems, from classical over machine learning to LLMs, for roughly 6,000 characters of late handwritten church slavonic manuscripts from the 18th century. Focussing on basic letter correctness, more than 10 CS OCR-systems among which 2 LLMs (GPT5 and Gemini3-flash) are being compared. Then, post-processing via LLMs is assessed and finally, different agentic OCR architectures (specialized post-processing agents, an agentic pipeline and RAG) are tested. With new technology elaborated, experiments suggest, church slavonic CER for basic letters may reach as low as 2-3% but elaborated diacritics could still present a problem. How well OCR can prime stemmatology as a downstream task is the entry point to the second part of the article which introduces a new stemmatic method based solely on image processing. Here, a pipeline of automated visual glyph extraction, clustering and pairwise statistical comparison leading to a distance matrix and ultimately a stemma, is being presented and applied to two small corpora, one for the church slavonic Gospel of Mark from the 14th to 16th centuries, one for the Roman de la Rose in French from the 14th and 15th centuries. Basic functioning of the method can be demonstrated. --- ### Evaluating Cooperation in LLM Social Groups through Elected Leadership Authors: Ryan Faulkner, Anushka Deshpande, David Guzman Piedrahita, Joel Z. Leibo, Zhijing Jin Published: 2026-04-13 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.11721v2 Governing common-pool resources requires agents to develop enduring strategies through cooperation and self-governance to avoid collective failure. While foundation models have shown potential for cooperation in these settings, existing multi-agent research provides little insight into whether structured leadership and election mechanisms can improve collective decision making. The lack of such a critical organizational feature ubiquitous in human society presents a significant shortcoming of the current methods. In this work we aim to directly address whether leadership and elections can support improved social welfare and cooperation through multi-agent simulation with LLMs. We present our open-source framework that simulates leadership through elected personas and candidate-driven agendas and carry out an empirical study of LLMs under controlled governance conditions. Our experiments demonstrate that having elected leadership improves social welfare scores by 55.4% and survival time by 128.6% across a range of high performing LLMs. Through the construction of an agent social graph we compute centrality metrics to assess the social influence of leader personas and also analyze rhetorical and cooperative tendencies revealed through a sentiment analysis on leader utterances. This work lays the foundation for further study of election mechanisms in multi-agent systems toward navigating complex social dilemmas. --- ### On the Robustness of Watermarking for Autoregressive Image Generation Authors: Andreas Müller, Denis Lukovnikov, Shingo Kodama, Minh Pham, Anubhav Jain, Jonathan Petit, Niv Cohen, Asja Fischer Published: 2026-04-13 Categories: cs.CV, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2604.11720v1 The proliferation of autoregressive (AR) image generators demands reliable detection and attribution of their outputs to mitigate misinformation, and to filter synthetic images from training data to prevent model collapse. To address this need, watermarking techniques, specifically designed for AR models, embed a subtle signal at generation time, enabling downstream verification through a corresponding watermark detector. In this work, we study these schemes and demonstrate their vulnerability to both watermark removal and forgery attacks. We assess existing attacks and further introduce three new attacks: (i) a vector-quantized regeneration removal attack, (ii) adversarial optimization-based attack, and (iii) a frequency injection attack. Our evaluation reveals that removal and forgery attacks can be effective with access to a single watermarked reference image and without access to original model parameters or watermarking secrets. Our findings indicate that existing watermarking schemes for AR image generation do not reliably support synthetic content detection for dataset filtering. Moreover, they enable Watermark Mimicry, whereby authentic images can be manipulated to imitate a generator's watermark and trigger false detection to prevent their inclusion in future model training. --- ### SWE-AGILE: A Software Agent Framework for Efficiently Managing Dynamic Reasoning Context Authors: Shuquan Lian, Juncheng Liu, Yazhe Chen, Yuhong Chen, Hui Li Published: 2026-04-13 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.11716v1 Prior representative ReAct-style approaches in autonomous Software Engineering (SWE) typically lack the explicit System-2 reasoning required for deep analysis and handling complex edge cases. While recent reasoning models demonstrate the potential of extended Chain-of-Thought (CoT), applying them to the multi-turn SWE task creates a fundamental dilemma: retaining full reasoning history leads to context explosion and ``Lost-in-the-Middle'' degradation, while discarding it would force the agent to redundantly re-reason at every step. To address these challenges, we propose SWE-AGILE, a novel software agent framework designed to bridge the gap between reasoning depth, efficiency, and context constraints. SWE-AGILE introduces a Dynamic Reasoning Context strategy, maintaining a ``sliding window'' of detailed reasoning for immediate continuity to prevent redundant re-analyzing, while compressing historical reasoning content into concise Reasoning Digests. Empirically, SWE-AGILE sets a new standard for 7B-8B models on SWE-Bench-Verified using only 2.2k trajectories and 896 tasks. Code is available at https://github.com/KDEGroup/SWE-AGILE. --- ### BEM: Training-Free Background Embedding Memory for False-Positive Suppression in Real-Time Fixed-Background Camera Authors: Junwoo Park, Jangho Lee, Sunho Lim Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.11714v3 Pretrained detectors perform well on benchmarks but often suffer performance degradation in real-world deployments due to distribution gaps between training data and target environments. COCO-like benchmarks emphasize category diversity rather than instance density, causing detectors trained under per-class sparsity to struggle in dense, single- or few-class scenes such as surveillance and traffic monitoring. In fixed-camera environments, the quasi-static background provides a stable, label-free prior that can be exploited at inference to suppress spurious detections. To address the issue, we propose Background Embedding Memory (BEM), a lightweight, training-free, weight-frozen module that can be attached to pretrained detectors during inference. BEM estimates clean background embeddings, maintains a prototype memory, and re-scores detection logits with an inverse-similarity, rank-weighted penalty, effectively reducing false positives while maintaining recall. Empirically, background-frame cosine similarity correlates negatively with object count and positively with Precision-Confidence AUC (P-AUC), motivating its use as a training-free control signal. Across YOLO and RT-DETR families on LLVIP and simulated surveillance streams, BEM consistently reduces false positives while preserving real-time performance. Our code is available at https://github.com/Leo-Park1214/Background-Embedding-Memory.git --- ### Seeing Through the Tool: A Controlled Benchmark for Occlusion Robustness in Foundation Segmentation Models Authors: Nhan Ho, Luu Le, Thanh-Huy Nguyen, Thien Nguyen, Xiaofeng Liu, Ulas Bagci Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.11711v1 Occlusion, where target structures are partially hidden by surgical instruments or overlapping tissues, remains a critical yet underexplored challenge for foundation segmentation models in clinical endoscopy. We introduce OccSAM-Bench, a benchmark designed to systematically evaluate SAM-family models under controlled, synthesized surgical occlusion. Our framework simulates two occlusion types (i.e., surgical tool overlay and cutout) across three calibrated severity levels on three public polyp datasets. We propose a novel three-region evaluation protocol that decomposes segmentation performance into full, visible-only, and invisible targets. This metric exposes behaviors that standard amodal evaluation obscures, revealing two distinct model archetypes: Occluder-Aware models (SAM, SAM 2, SAM 3, MedSAM3), which prioritize visible tissue delineation and reject instruments, and Occluder-Agnostic models (MedSAM, MedSAM2), which confidently predict into occluded regions. SAM-Med2D aligns with neither and underperforms across all conditions. Ultimately, our results demonstrate that occlusion robustness is not uniform across architectures, and model selection must be driven by specific clinical intent-whether prioritizing conservative visible-tissue segmentation or the amodal inference of hidden anatomy. --- ### A Mamba-Based Multimodal Network for Multiscale Blast-Induced Rapid Structural Damage Assessment Authors: Wanli Ma, Sivasakthy Selvakumaran, Dain G. Farrimond, Adam A. Dennis, Samuel E. Rigby Published: 2026-04-13 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.11709v1 Accurate and rapid structural damage assessment (SDA) is crucial for post-disaster management, helping responders prioritise resources, plan rescues, and support recovery. Traditional field inspections, though precise, are limited by accessibility, safety risks, and time constraints, especially after large explosions. Machine learning with remote sensing has emerged as a scalable solution for rapid SDA, with Mamba-based networks achieving state-of-the-art performance. However, these methods often require extensive training and large datasets, limiting real-world applicability. Moreover, they fail to incorporate key physical characteristics of blast loading for SDA. To overcome these challenges, we propose a Mamba-based multimodal network for rapid SDA that integrates multi-scale blast-loading information with optical remote sensing images. Evaluated on the 2020 Beirut explosion, our method significantly improves performance over state-of-the-art approaches. Code is available at: https://github.com/IMPACTSquad/Blast-Mamba --- ### Representations Before Pixels: Semantics-Guided Hierarchical Video Prediction Authors: Efstathios Karypidis, Spyros Gidaris, Nikos Komodakis Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.11707v1 Accurate future video prediction requires both high visual fidelity and consistent scene semantics, particularly in complex dynamic environments such as autonomous driving. We present Re2Pix, a hierarchical video prediction framework that decomposes forecasting into two stages: semantic representation prediction and representation-guided visual synthesis. Instead of directly predicting future RGB frames, our approach first forecasts future scene structure in the feature space of a frozen vision foundation model, and then conditions a latent diffusion model on these predicted representations to render photorealistic frames. This decomposition enables the model to focus first on scene dynamics and then on appearance generation. A key challenge arises from the train-test mismatch between ground-truth representations available during training and predicted ones used at inference. To address this, we introduce two conditioning strategies, nested dropout and mixed supervision, that improve robustness to imperfect autoregressive predictions. Experiments on challenging driving benchmarks demonstrate that the proposed semantics-first design significantly improves temporal semantic consistency, perceptual quality, and training efficiency compared to strong diffusion baselines. We provide the implementation code at https://github.com/Sta8is/Re2Pix --- ### Agentic Driving Coach: Robustness and Determinism of Agentic AI-Powered Human-in-the-Loop Cyber-Physical Systems Authors: Deeksha Prahlad, Daniel Fan, Hokeun Kim Published: 2026-04-13 Categories: cs.AI, cs.CL, cs.RO, eess.SY arXiv: https://arxiv.org/abs/2604.11705v1 Foundation models, including large language models (LLMs), are increasingly used for human-in-the-loop (HITL) cyber-physical systems (CPS) because foundation model-based AI agents can potentially interact with both the physical environments and human users. However, the unpredictable behavior of human users and AI agents, in addition to the dynamically changing physical environments, leads to uncontrollable nondeterminism. To address this urgent challenge of enabling agentic AI-powered HITL CPS, we propose a reactor-model-of-computation (MoC)-based approach, realized by the open-source Lingua Franca (LF) framework. We also carry out a concrete case study using the agentic driving coach as an application of HITL CPS. By evaluating the LF-based agentic HITL CPS, we identify practical challenges in reintroducing determinism into such agentic HITL CPS and present pathways to address them. --- ### Fairness is Not Flat: Geometric Phase Transitions Against Shortcut Learning Authors: Nicolas Rodriguez-Alvarez Published: 2026-04-13 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.11704v2 Deep Neural Networks are highly susceptible to shortcut learning, frequently memorizing low-dimensional spurious correlations instead of underlying causal mechanisms. This phenomenon not only degrades out-of-distribution robustness but also induces severe demographic biases in sensitive applications. In this paper, we propose a geometric \textit{a priori} methodology to mitigate shortcut learning. By deploying a zero-hidden-layer ($N=1$) Topological Auditor, we mathematically isolate features that monopolize the gradient without human intervention. We empirically demonstrate a Capacity Phase Transition: once linear shortcuts are pruned, networks are forced to utilize higher geometric capacity ($N \geq 16$) to curve the decision boundary and learn ethical representations. Our approach outperforms L1 Regularization -- which collapses into demographic bias -- and operates at a fraction of the computational cost of post-hoc methods like Just Train Twice (JTT), successfully reducing counterfactual gender vulnerability from 21.18\% to 7.66\%. --- ### Fairness is Not Flat: Geometric Phase Transitions Against Shortcut Learning Authors: Nicolas Rodriguez-Alvarez, Fernando Rodriguez-Merino Published: 2026-04-13 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.11704v1 Deep Neural Networks are highly susceptible to shortcut learning, frequently memorizing low-dimensional spurious correlations instead of underlying causal mechanisms. This phenomenon not only degrades out-of-distribution robustness but also induces severe demographic biases in sensitive applications. In this paper, we propose a geometric \textit{a priori} methodology to mitigate shortcut learning. By deploying a zero-hidden-layer ($N=1$) Topological Auditor, we mathematically isolate features that monopolize the gradient without human intervention. We empirically demonstrate a Capacity Phase Transition: once linear shortcuts are pruned, networks are forced to utilize higher geometric capacity ($N \geq 16$) to curve the decision boundary and learn ethical representations. Our approach outperforms L1 Regularization -- which collapses into demographic bias -- and operates at a fraction of the computational cost of post-hoc methods like Just Train Twice (JTT), successfully reducing counterfactual gender vulnerability from 21.18\% to 7.66\%. --- ### DreamKG: A KG-Augmented Conversational System for People Experiencing Homelessness Authors: Javad M Alizadeh, Genhui Zheng, Chiu C Tan, Yuzhou Chen, Omar Martinez, Philip McCallion, Ying Ding, Chenguang Yang, AnneMarie Tomosky, Huanmei Wu Published: 2026-04-13 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.11703v1 People experiencing homelessness (PEH) face substantial barriers to accessing timely, accurate information about community services. DreamKG addresses this through a knowledge graph-augmented conversational system that grounds responses in verified, up-to-date data about Philadelphia organizations, services, locations, and hours. Unlike standard large language models (LLMs) prone to hallucinations, DreamKG combines Neo4j knowledge graphs with structured query understanding to handle location-aware and time-sensitive queries reliably. The system performs spatial reasoning for distance-based recommendations and temporal filtering for operating hours. Preliminary evaluation shows 59% superiority over Google Search AI on relevant queries and 84% rejection of irrelevant queries. This demonstration highlights the potential of hybrid architectures that combines LLM flexibility with knowledge graph reliability to improve service accessibility for vulnerable populations effectively. --- ### Legal2LogicICL: Improving Generalization in Transforming Legal Cases to Logical Formulas via Diverse Few-Shot Learning Authors: Jieying Xue, Phuong Minh Nguyen, Ha Thanh Nguyen, May Myo Zin, Ken Satoh Published: 2026-04-13 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.11699v1 This work aims to improve the generalization of logic-based legal reasoning systems by integrating recent advances in NLP with legal-domain adaptive few-shot learning techniques using LLMs. Existing logic-based legal reasoning pipelines typically rely on fine-tuned models to map natural-language legal cases into logical formulas before forwarding them to a symbolic reasoner. However, such approaches are heavily constrained by the scarcity of high-quality annotated training data. To address this limitation, we propose a novel LLM-based legal reasoning framework that enables effective in-context learning through retrieval-augmented generation. Specifically, we introduce Legal2LogicICL, a few-shot retrieval framework that balances diversity and similarity of exemplars at both the latent semantic representation level and the legal text structure level. In addition, our method explicitly accounts for legal structure by mitigating entity-induced retrieval bias in legal texts, where lengthy and highly specific entity mentions often dominate semantic representations and obscure legally meaningful reasoning patterns. Our Legal2LogicICL constructs informative and robust few-shot demonstrations, leading to accurate and stable logical rule generation without requiring additional training. In addition, we construct a new dataset, named Legal2Proleg, which is annotated with alignments between legal cases and PROLEG logical formulas to support the evaluation of legal semantic parsing. Experimental results on both open-source and proprietary LLMs demonstrate that our approach significantly improves accuracy, stability, and generalization in transforming natural-language legal case descriptions into logical representations, highlighting its effectiveness for interpretable and reliable legal reasoning. Our code is available at https://github.com/yingjie7/Legal2LogicICL. --- ### LARY: A Latent Action Representation Yielding Benchmark for Generalizable Vision-to-Action Alignment Authors: Dujun Nie, Fengjiao Chen, Qi Lv, Jun Kuang, Xiaoyu Li, Xuezhi Cao, Xunliang Cai Published: 2026-04-13 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2604.11689v1 While the shortage of explicit action data limits Vision-Language-Action (VLA) models, human action videos offer a scalable yet unlabeled data source. A critical challenge in utilizing large-scale human video datasets lies in transforming visual signals into ontology-independent representations, known as latent actions. However, the capacity of latent action representation to derive robust control from visual observations has yet to be rigorously evaluated. We introduce the Latent Action Representation Yielding (LARY) Benchmark, a unified framework for evaluating latent action representations on both high-level semantic actions (what to do) and low-level robotic control (how to do). The comprehensively curated dataset encompasses over one million videos (1,000 hours) spanning 151 action categories, alongside 620K image pairs and 595K motion trajectories across diverse embodiments and environments. Our experiments reveal two crucial insights: (i) General visual foundation models, trained without any action supervision, consistently outperform specialized embodied latent action models. (ii) Latent-based visual space is fundamentally better aligned to physical action space than pixel-based space. These results suggest that general visual representations inherently encode action-relevant knowledge for physical control, and that semantic-level abstraction serves as a fundamentally more effective pathway from vision to action than pixel-level reconstruction. --- ### Please Make it Sound like Human: Encoder-Decoder vs. Decoder-Only Transformers for AI-to-Human Text Style Transfer Authors: Utsav Paneru Published: 2026-04-13 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.11687v1 AI-generated text has become common in academic and professional writing, prompting research into detection methods. Less studied is the reverse: systematically rewriting AI-generated prose to read as genuinely human-authored. We build a parallel corpus of 25,140 paired AI-input and human-reference text chunks, identify 11 measurable stylistic markers separating the two registers, and fine-tune three models: BART-base, BART-large, and Mistral-7B-Instruct with QLoRA. BART-large achieves the highest reference similarity -- BERTScore F1 of 0.924, ROUGE-L of 0.566, and chrF++ of 55.92 -- with 17x fewer parameters than Mistral-7B. We show that Mistral-7B's higher marker shift score reflects overshoot rather than accuracy, and argue that shift accuracy is a meaningful blind spot in current style transfer evaluation. --- ### Unfolding 3D Gaussian Splatting via Iterative Gaussian Synopsis Authors: Yuqin Lu, Yang Zhou, Yihua Dai, Guiqing Li, Shengfeng He Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.11685v1 3D Gaussian Splatting (3DGS) has become a state-of-the-art framework for real-time, high-fidelity novel view synthesis. However, its substantial storage requirements and inherently unstructured representation pose challenges for deployment in streaming and resource-constrained environments. Existing Level-of-Detail (LOD) strategies, particularly those based on bottom-up construction, often introduce redundancy or lead to fidelity degradation. To overcome these limitations, we propose Iterative Gaussian Synopsis, a novel framework for compact and progressive rendering through a top-down "unfolding" scheme. Our approach begins with a full-resolution 3DGS model and iteratively derives coarser LODs using an adaptive, learnable mask-based pruning mechanism. This process constructs a multi-level hierarchy that preserves visual quality while improving efficiency. We integrate hierarchical spatial grids, which capture the global scene structure, with a shared Anchor Codebook that models localized details. This combination produces a compact yet expressive feature representation, designed to minimize redundancy and support efficient, level-specific adaptation. The unfolding mechanism promotes inter-layer reusability and requires only minimal data overhead for progressive refinement. Experiments show that our method maintains high rendering quality across all LODs while achieving substantial storage reduction. These results demonstrate the practicality and scalability of our approach for real-time 3DGS rendering in bandwidth- and memory-constrained scenarios. --- ### Towards Brain MRI Foundation Models for the Clinic: Findings from the FOMO25 Challenge Authors: Asbjørn Munk, Stefano Cerri, Vardan Nersesjan, Christian Hedeager Krag, Jakob Ambsdorf, Pablo Rocamora García, Julia Machnio, Peirong Liu, Suhyun Ahn, Nasrin Akbari, Yasmina Al Khalil, Kimberly Amador et al. Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.11679v2 Clinical deployment of automated brain MRI analysis faces a fundamental challenge: clinical data is heterogeneous and noisy, and high-quality labels are prohibitively costly to obtain. Self-supervised learning (SSL) can address this by leveraging the vast amounts of unlabeled data produced in clinical workflows to train robust \textit{foundation models} that adapt out-of-domain with minimal supervision. However, the development of foundation models for brain MRI has been limited by small pretraining datasets and in-domain benchmarking focused on high-quality, research-grade data. To address this gap, we organized the FOMO25 challenge as a satellite event at MICCAI 2025. FOMO25 provided participants with a large pretraining dataset, FOMO60K, and evaluated models on data sourced directly from clinical workflows in few-shot and out-of-domain settings. Tasks covered infarct classification, meningioma segmentation, and brain age regression, and considered both models trained on FOMO60K (method track) and any data (open track). Nineteen foundation models from sixteen teams were evaluated using a standardized containerized pipeline. Results show that (a) self-supervised pretraining improves generalization on clinical data under domain shift, with the strongest models trained \textit{out-of-domain} surpassing supervised baselines trained \textit{in-domain}. (b) No single pretraining objective benefits all tasks: MAE favors segmentation, hybrid reconstruction-contrastive objectives favor classification, and (c) strong performance was achieved by small pretrained models, and improvements from scaling model size and training duration did not yield reliable benefits. --- ### AffordSim: A Scalable Data Generator and Benchmark for Affordance-Aware Robotic Manipulation Authors: Mingyang Li, Haofan Xu, Haowen Sun, Xinzhe Chen, Sihua Ren, Liqi Huang, Xinyang Sui, Chenyang Miao, Jiawei Ye, Qiongjie Cui, Zeyang Liu, Xingyu Chen et al. Published: 2026-04-13 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2604.11674v2 Many everyday robot manipulation skills are affordance-dependent, with success determined by whether the robot contacts the functional object region required by the subsequent action. Current simulation data generators obtain contacts from generic grasp estimators or per-object manual contact annotations, but generic estimators rank stable grasps without task semantics and often select contacts that are misaligned with the downstream action, while manual contact annotations must be rewritten for each new object and task. To solve these challenges, we introduce AffordSim, a scalable data generator and benchmark that integrates open-vocabulary 3D affordance prediction into simulation-based trajectory generation. Given a natural-language task description, AffordSim synthesizes a task-relevant scene, emits affordance queries, grounds them on object surfaces, samples region-conditioned grasps, and selects executable candidates with motion planning. It further randomizes object pose, texture, lighting, image noise, and cross-viewpoint backgrounds for sim-to-real transfer. We instantiate AffordSim as a 50-task benchmark across diverse manipulation skills, five robot embodiments, and 500+ rigid and articulated objects. AffordSim achieves 93% of the trajectory collection success rate of manual contact annotations on affordance-critical tasks and 89% on hard composite tasks. Vision-language-action policies trained on AffordSim data transfer zero-shot to a real Franka FR3, reaching 24% average success. --- ### NetworkNet: A Deep Neural Network Approach for Random Networks with Sparse Nodal Attributes and Complex Nodal Heterogeneity Authors: Zhaoyu Xing, Xiufan Yu Published: 2026-04-13 Categories: stat.ME, cs.AI, math.ST, stat.CO arXiv: https://arxiv.org/abs/2604.11673v1 Heterogeneous network data with rich nodal information become increasingly prevalent across multidisciplinary research, yet accurately modeling complex nodal heterogeneity and simultaneously selecting influential nodal attributes remains an open challenge. This problem is central to many applications in economics and sociology, when both nodal heterogeneity and high-dimensional individual characteristics highly affect network formation. We propose a statistically grounded, unified deep neural network approach for modeling nodal heterogeneity in random networks with high-dimensional nodal attributes, namely ``NetworkNet''. A key innovation of NetworkNet lies in a tailored neural architecture that explicitly parameterizes attribute-driven heterogeneity, and at the same time, embeds a scalable attribute selection mechanism. NetworkNet consistently estimates two types of latent heterogeneity functions, i.e., nodal expansiveness and popularity, while simultaneously performing data-driven attribute selection to extract influential nodal attributes. By unifying classical statistical network modeling with deep learning, NetworkNet delivers the expressive power of DNNs with methodological interpretability, algorithmic scalability, and statistical rigor with a non-asymptotic approximation error bound. Empirically, simulations demonstrate strong performance in both heterogeneity estimation and high-dimensional attribute selection. We further apply NetworkNet to a large-scale author-citation network among statisticians, revealing new insights into the dynamic evolution of research fields and scholarly impact. --- ### UNIGEOCLIP: Unified Geospatial Contrastive Learning Authors: Guillaume Astruc, Eduard Trulls, Jan Hosang, Loic Landrieu, Paul-Edouard Sarlin Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.11668v1 The growing availability of co-located geospatial data spanning aerial imagery, street-level views, elevation models, text, and geographic coordinates offers a unique opportunity for multimodal representation learning. We introduce UNIGEOCLIP, a massively multimodal contrastive framework to jointly align five complementary geospatial modalities in a single unified embedding space. Unlike prior approaches that fuse modalities or rely on a central pivot representation, our method performs all-to-all contrastive alignment, enabling seamless comparison, retrieval, and reasoning across arbitrary combinations of modalities. We further propose a scaled latitude-longitude encoder that improves spatial representation by capturing multi-scale geographic structure. Extensive experiments across downstream geospatial tasks demonstrate that UNIGEOCLIP consistently outperforms single-modality contrastive models and coordinate-only baselines, highlighting the benefits of holistic multimodal geospatial alignment. A reference implementation is available at https://gastruc.github.io/unigeoclip. --- ### Playing Along: Learning a Double-Agent Defender for Belief Steering via Theory of Mind Authors: Hanqi Xiao, Vaidehi Patil, Zaid Khan, Hyunji Lee, Elias Stengel-Eskin, Mohit Bansal Published: 2026-04-13 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.11666v2 As large language models (LLMs) become the engine behind conversational systems, their ability to reason about the intentions and states of their dialogue partners (i.e., form and use a theory-of-mind, or ToM) becomes increasingly critical for safe interaction with potentially adversarial partners. We propose a novel privacy-themed ToM challenge, ToM for Steering Beliefs (ToM-SB), in which a defender must act as a Double Agent to steer the beliefs of an attacker with partial prior knowledge within a shared universe. To succeed on ToM-SB, the defender must engage with and form a ToM of the attacker, with a goal of fooling the attacker into believing they have succeeded in extracting sensitive information. We find that strong frontier models like Gemini3-Pro and GPT-5.4 struggle on ToM-SB, often failing to fool attackers in hard scenarios with partial attacker prior knowledge, even when prompted to reason about the attacker's beliefs (ToM prompting). To close this gap, we train models on ToM-SB to act as AI Double Agents using reinforcement learning, testing both fooling and ToM rewards. Notably, we find a bidirectionally emergent relationship between ToM and attacker-fooling: rewarding fooling success alone improves ToM, and rewarding ToM alone improves fooling. Across four attackers with different strengths, six defender methods, and both in-distribution and out-of-distribution (OOD) evaluation, we find that gains in ToM and attacker-fooling are well-correlated, highlighting belief modeling as a key driver of success on ToM-SB. AI Double Agents that combine both ToM and fooling rewards yield the strongest fooling and ToM performance, outperforming Gemini3-Pro and GPT-5.4 with ToM prompting on hard scenarios. We also show that ToM-SB and AI Double Agents can be extended to stronger attackers, demonstrating generalization to OOD settings and the upgradability of our task. --- ### Beyond LLMs, Sparse Distributed Memory, and Neuromorphics <A Hyper-Dimensional SRAM-CAM "VaCoAl" for Ultra-High Speed, Ultra-Low Power, and Low Cost> Authors: Hiroyuki Chuma, Kanji Otsuka, Yoichi Sato Published: 2026-04-13 Categories: cs.NE, cs.AI arXiv: https://arxiv.org/abs/2604.11665v7 VaCoAl (Python: PyVaCoAl) attacks the Binding Problem algebraically rather than statistically: an SRAM/DRAM-CAM organised end-to-end around one primitive, XOR-and-shift over GF(2) via primitive-polynomial LFSRs. Rooted in Kanerva's Sparse Distributed Memory, it retrieves in a million-dimensional binary space by Galois-field diffusion; Binding and Unbinding are exactly reversible at O(L), giving compositional generalisation with post-hoc auditability. Unexpectedly, a path-dependent semantic selection mechanism emerges undesigned, functionally equivalent to Spike-Timing-Dependent Plasticity (STDP) and predictable a priori from a closed form matching measurement on tens of millions of records. It is the Don't Care (collision-tolerance) rule: not a defect to engineer away, but the source of its ability to rank paths by quality. Capability: reversible binding plus tolerated collisions yields a path-integral confidence measure (CR2) ranking candidates by accumulated reliability, absent from hash-based search. Necessity: repair the collisions fully and the ranking vanishes; a memory that never fails has no record of which paths were difficult. Position: we do not surpass large language models; the substrate is orthogonal, supplying the auditable path-ranking layer they lack. A mentor-student ontology of ~470,000 WIKIDATA scholars, traversed backwards from all 64 Fields Medalists to depth 57 and >25.5M paths, stress-tests multi-hop reasoning over a large DAG. Ablation shows Don't Care imposing a depth-dependent exponential penalty that prunes circuitous routes and preserves direct ones: an Occam's razor. Per-generation CR1 stays within 0.995-0.999, yet cumulative CR2 decays monotonically to ~0.905 by generation 56, matching the closed form 0.997^56 = 0.846 within observed variation. All measurements are from PyVaCoAl, a software DRAM-CAM; speed and power figures are projections. --- ### Beyond LLMs, Sparse Distributed Memory, and Neuromorphics <A Hyper-Dimensional SRAM-CAM "VaCoAl" for Ultra-High Speed, Ultra-Low Power, and Low Cost> Authors: Hiroyuki Chuma, Kanji Otsuka, Yoichi Sato Published: 2026-04-13 Categories: cs.NE, cs.AI arXiv: https://arxiv.org/abs/2604.11665v6 This paper reports an unexpected finding: in a deterministic hyperdimensional computing (HDC) architecture **that inverts the conventional role of Galois-field algebra -- employing it not for error correction toward a unique answer but as an engine for relative similarity and path-quality ranking -- **a path-dependent semantic selection mechanism emerges, equivalent to spike-timing-dependent plasticity (STDP), with magnitude predictable a priori from a closed-form expression matching measured values. Addressing catastrophic forgetting, learning stagnation, and the Binding Problem at an algebraic level, we propose VaCoAl (Vague Coincident Algorithm) and its Python implementation PyVaCoAl on ultra-high-dimensional SRAM/DRAM-CAM. Rooted in Sparse Distributed Memory, it resolves orthogonalisation and retrieval in high-dimensional binary spaces via Galois-field diffusion, enabling low-load deployment. Crucially, VaCoAl embeds a cognitive bound -- the Frontier Size -- into its architecture, ranking candidates by path-integral confidence (CR2) to achieve compositional generalisation; this bounded-rationality design produces STDP-like selection that error-correction paradigms structurally cannot attain. We evaluated multi-hop reasoning on about 470k mentor-student relations from Wikidata, tracing up to 57 generations (over 25.5M paths). HDC bundling and unbinding with CR-based denoising quantify concept propagation over DAGs. Results show a reinterpretation of the Newton-Leibniz dispute and a phase transition from sparse convergence to a post-Leibniz "superhighway", with structural indicators supporting a Kuhnian paradigm shift. VaCoAl thus defines a third paradigm, HDC-AI, complementing LLMs with reversible, auditable multi-hop reasoning. --- ### Why Do Large Language Models Generate Harmful Content? Authors: Rajesh Ganguli, Raha Moraffah Published: 2026-04-13 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.11663v1 Large Language Models (LLMs) have been shown to generate harmful content. However, the underlying causes of such behavior remain under explored. We propose a causal mediation analysis-based approach to identify the causal factors responsible for harmful generation. Our method performs a multi-granular analysis across model layers, modules (MLP and attention blocks), and individual neurons. Extensive experiments on state-of-the-art LLMs indicate that harmful generation arises in the later layers of the model, results primarily from failures in MLP blocks rather than attention blocks, and is associated with neurons that act as a gating mechanism for harmful generation. The results indicate that the early layers in the model are used for a contextual understanding of harmfulness in a prompt, which is then propagated through the model, to generate harmfulness in the late layers, as well as a signal indicating harmfulness through MLP blocks. This is then further propagated to the last layer of the model, specifically to a sparse set of neurons, which receives the signal and determines the generation of harmful content accordingly. --- ### A Robust Evaluation of Probe Robustness: Lessons for Reliable OOD Uncertainty Quantification Authors: Joe Stacey, Hadas Orgad, Kentaro Inui, Benjamin Heinzerling, Nafise Sadat Moosavi Published: 2026-04-13 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.11662v2 Recent work has shown that the hidden states of large language models contain signals useful for uncertainty estimation, motivating a growing interest in efficient probe-based approaches. Yet it remains unclear how robust existing methods are, with prior work reporting conflicting conclusions under substantially different evaluation settings. We address this by introducing ProbeDrift, a systematic evaluation framework for supervised uncertainty probes covering a wide range of OOD settings across models, tasks, and distributional shifts. Using ProbeDrift, we train over 2,000 probes to disentangle the effect of key design choices, showing poor robustness of current methods beyond near-OOD settings. We find that robustness is driven by design decisions that have a largely invisible effect in-distribution, including the choice of feature type, aggregation strategy, and training signal. We argue that robust uncertainty estimation requires robust evaluation. To support this, we release ProbeDrift as a lightweight Python library that contains the train and test splits underpinning our extensive evaluation. We also show how insights from our evaluation can directly lead to more robust methods through a simple Hybrid Back-Off (HBO) strategy. --- ### Hidden Failures in Robustness: Why Supervised Uncertainty Quantification Needs Better Evaluation Authors: Joe Stacey, Hadas Orgad, Kentaro Inui, Benjamin Heinzerling, Nafise Sadat Moosavi Published: 2026-04-13 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.11662v1 Recent work has shown that the hidden states of large language models contain signals useful for uncertainty estimation and hallucination detection, motivating a growing interest in efficient probe-based approaches. Yet it remains unclear how robust existing methods are, and which probe designs provide uncertainty estimates that are reliable under distribution shift. We present a systematic study of supervised uncertainty probes across models, tasks, and OOD settings, training over 2,000 probes while varying the representation layer, feature type, and token aggregation strategy. Our evaluation highlights poor robustness in current methods, particularly in the case of long-form generations. We also find that probe robustness is driven less by architecture and more by the probe inputs. Middle-layer representations generalise more reliably than final-layer hidden states, and aggregating across response tokens is consistently more robust than relying on single-token features. These differences are often largely invisible in-distribution but become more important under distribution shift. Informed by our evaluation, we explore a simple hybrid back-off strategy for improving robustness, arguing that better evaluation is a prerequisite for building more robust probes. --- ### Towards Autonomous Mechanistic Reasoning in Virtual Cells Authors: Yunhui Jang, Lu Zhu, Jake Fawkes, Alisandra Kaye Denton, Dominique Beaini, Emmanuel Noutahi Published: 2026-04-13 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.11661v3 Large language models (LLMs) have recently gained significant attention as a promising approach to accelerate scientific discovery. However, their application in open-ended scientific domains such as biology remains limited, primarily due to the lack of factually grounded and actionable explanations. To address this, we introduce a structured explanation formalism for virtual cells that represents biological reasoning as mechanistic action graphs, enabling systematic verification and falsification. Building upon this, we propose VCR-Agent, a multi-agent framework that integrates biologically grounded knowledge retrieval with a verifier-based filtering approach to generate and validate mechanistic reasoning autonomously. Using this framework, we release VC-TRACES dataset, which consists of verified mechanistic explanations derived from the Tahoe-100M atlas. Empirically, we demonstrate that training with these explanations improves factual precision and provides a more effective supervision signal for downstream gene expression prediction. These results underscore the importance of reliable mechanistic reasoning for virtual cells, achieved through the synergy of multi-agent and rigorous verification. --- ### GPU Acceleration of Sparse Fully Homomorphic Encrypted DNNs Authors: Lara D'Agata, Carlos Agulló-Domingo, Óscar Vera-López, Kaustubh Shivdikar, Ardhi W. B. Yudha, Ferhat Yaman, David Kaeli, José L. Abellán, Ian Colbert, José Cano Published: 2026-04-13 Categories: cs.CR, cs.DC, cs.DS, cs.LG, cs.PF arXiv: https://arxiv.org/abs/2604.11659v1 Fully homomorphic encryption (FHE) has recently attracted significant attention as both a cryptographic primitive and a systems challenge. Given the latest advances in accelerated computing, FHE presents a promising opportunity for progress, with applications ranging from machine learning to information security. We target the most computationally intensive operation in deep neural networks from a hardware perspective, matrix multiplication (matmul), and adapt it for execution on AMD GPUs. We propose a new optimized method that improves the runtime and complexity of ciphertext matmul by using FIDESlib, a recent open-source FHE library designed specifically for GPUs. By exploiting sparsity in both operands, our sparse matmul implementation outperforms its CPU counterpart by up to $3.0\times$ and reduces the time complexity from cubic to semi-linear, demonstrating an improvement over existing FHE matmul implementations. --- ### RPA-Check: A Multi-Stage Automated Framework for Evaluating Dynamic LLM-based Role-Playing Agents Authors: Riccardo Rosati, Edoardo Colucci, Massimiliano Bolognini, Adriano Mancini, Paolo Sernani Published: 2026-04-13 Categories: cs.CL, cs.AI, cs.MA arXiv: https://arxiv.org/abs/2604.11655v1 The rapid adoption of Large Language Models (LLMs) in interactive systems has enabled the creation of dynamic, open-ended Role-Playing Agents (RPAs). However, evaluating these agents remains a significant challenge, as standard NLP metrics fail to capture the nuances of role adherence, logical consistency, and long-term narrative stability. This paper introduces RPA-Check, a multi-stage automated evaluation framework designed to objectively assess the performance of LLM-based RPAs in complex, constraints-heavy environments. Our methodology is based on a four-step pipeline: (1) Dimension Definition, establishing high-level qualitative behavioral criteria; (2) Augmentation, where these requirements are expanded into granular boolean checklist indicators; (3) Semantic Filtering, to ensure indicator objectivity, no redundancy and agent isolation; and (4) LLM-as-a-Judge Evaluation, which employs chain-of-thought verification to score agent fidelity. We validate this framework by applying it to LLM Court, a serious game for forensic training involving several quantized local models. Experimental results across five distinct legal scenarios demonstrate the framework's ability to identify subtle trade-offs between model size, reasoning depth, and operational stability. Notably, the findings reveal an inverse relationship between parametric scale and procedural consistency, showing that smaller, adequately instruction-tuned models (8-9B) can outperform larger architectures prone to user-alignment bias or sycophancy. RPA-Check thus provides a standardized and reproducible metric for future research in generative agent evaluation within specialized domains. --- ### GazeVaLM: A Multi-Observer Eye-Tracking Benchmark for Evaluating Clinical Realism in AI-Generated X-Rays Authors: David Wong, Zeynep Isik, Bin Wang, Marouane Tliba, Gorkem Durak, Elif Keles, Halil Ertugrul Aktas, Aladine Chetouani, Cagdas Topel, Nicolo Gennaro, Camila Lopes Vendrami, Tugce Agirlar Trabzonlu et al. Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.11653v1 We introduce GazeVaLM, a public eye-tracking dataset for studying clinical perception during chest radiograph authenticity assessment. The dataset comprises 960 gaze recordings from 16 expert radiologists interpreting 30 real and 30 synthetic chest X-rays (generated by diffusion based generative AI) under two conditions: diagnostic assessment and real-fake classification (Visual Turing test). For each image-observer pair, we provide raw gaze samples, fixation maps, scanpaths, saliency density maps, structured diagnostic labels, and authenticity judgments. We extend the protocol to 6 state-of-the-art multimodal LLMs, releasing their predicted diagnoses, authenticity labels, and confidence scores under matched conditions - enabling direct human-AI comparison at both decision and uncertainty levels. We further provide analyses of gaze agreement, inter-observer consistency, and benchmarking of radiologists versus LLMs in diagnostic accuracy and authenticity detection. GazeVaLM supports research in gaze modeling, clinical decision-making, human-AI comparison, generative image realism assessment, and uncertainty quantification. By jointly releasing visual attention data, clinical labels, and model predictions, we aim to facilitate reproducible research on how experts and AI systems perceive, interpret, and evaluate medical images. The dataset is available at https://huggingface.co/datasets/davidcwong/GazeVaLM. --- ### CodeTracer: Towards Traceable Agent States Authors: Han Li, Yifan Yao, Letian Zhu, Rili Feng, Hongyi Ye, Jiaming Wang, Yancheng He, Pengyu Zou, Lehan Zhang, Xinping Lei, Haoyang Huang, Ken Deng et al. Published: 2026-04-13 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2604.11641v3 Code agents are advancing rapidly, but debugging them is becoming increasingly difficult. As frameworks orchestrate parallel tool calls and multi-stage workflows over complex tasks, making the agent's state transitions and error propagation hard to observe. In these runs, an early misstep can trap the agent in unproductive loops or even cascade into fundamental errors, forming hidden error chains that make it hard to tell when the agent goes off track and why. Existing agent tracing analyses either focus on simple interaction or rely on small-scale manual inspection, which limits their scalability and usefulness for real coding workflows. We present CodeTracer, a tracing architecture that parses heterogeneous run artifacts through evolving extractors, reconstructs the full state transition history as a hierarchical trace tree with persistent memory, and performs failure onset localization to pinpoint the failure origin and its downstream chain. To enable systematic evaluation, we construct CodeTraceBench from a large collection of executed trajectories generated by four widely used code agent frameworks on diverse code tasks (e.g., bug fixing, refactoring, and terminal interaction), with supervision at both the stage and step levels for failure localization. Experiments show that CodeTracer substantially outperforms direct prompting and lightweight baselines, and that replaying its diagnostic signals consistently recovers originally failed runs under matched budgets. Our code and data are publicly available. --- ### Inter-Layer Hessian Analysis of Neural Networks with DAG Architectures Authors: Maxim Bolshim, Alexander Kugaevskikh Published: 2026-04-13 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.11639v1 Modern automatic differentiation frameworks (JAX, PyTorch) return the Hessian of the loss function as a monolithic tensor, without exposing the internal structure of inter-layer interactions. This paper presents an analytical formalism that explicitly decomposes the full Hessian into blocks indexed by the DAG of an arbitrary architecture. The canonical decomposition $H = H^{GN} + H^T$ separates the Gauss--Newton component (convex part) from the tensor component (residual curvature responsible for saddle points). For piecewise-linear activations (ReLU), the tensor component of the input Hessian vanishes ($H^{T}_{v,w}\!\equiv\!0$ a.e., $H^f_{v,w}\!=\!H^{GN}_{v,w}\!\succeq\!0$); the full parametric Hessian contains residual terms that do not reduce to the GGN. Building on this decomposition, we introduce diagnostic metrics (inter-layer resonance~$\mathcal{R}$, geometric coupling~$\mathcal{C}$, stable rank~$\mathcal{D}$, GN-Gap) that are estimated stochastically in $O(P)$ time and reveal structural curvature interactions between layers. The theoretical analysis explains exponential decay of resonance in vanilla networks and its preservation under skip connections; empirical validation spans fully connected MLPs (Exp.\,1--5) and convolutional architectures (ResNet-18, ${\sim}11$M~parameters, Exp.\,6). When the architecture reduces to a single node, all definitions collapse to the standard Hessian $\nabla^2_θ\mathcal{L}(θ)\in\mathbb{R}^{p\times p}$. --- ### STS-Mixer: Spatio-Temporal-Spectral Mixer for 4D Point Cloud Video Understanding Authors: Wenhao Li, Xueying Jiang, Gongjie Zhang, Xiaoqin Zhang, Ling Shao, Shijian Lu Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.11637v1 4D point cloud videos capture rich spatial and temporal dynamics of scenes which possess unique values in various 4D understanding tasks. However, most existing methods work in the spatiotemporal domain where the underlying geometric characteristics of 4D point cloud videos are hard to capture, leading to degraded representation learning and understanding of 4D point cloud videos. We address the above challenge from a complementary spectral perspective. By transforming 4D point cloud videos into graph spectral signals, we can decompose them into multiple frequency bands each of which captures distinct geometric structures of point cloud videos. Our spectral analysis reveals that the decomposed low-frequency signals capture more coarse shapes while high-frequency signals encode more fine-grained geometry details. Building on these observations, we design Spatio-Temporal-Spectral Mixer (STS-Mixer), a unified framework that mixes spatial, temporal, and spectral representations of point cloud videos. STS-Mixer integrates multi-band delineated spectral signals with spatiotemporal information to capture rich geometries and temporal dynamics, while enabling fine-grained and holistic understanding of 4D point cloud videos. Extensive experiments show that STS-Mixer achieves superior performance consistently across multiple widely adopted benchmarks on both 3D action recognition and 4D semantic segmentation tasks. Code and models are available at https://github.com/Vegetebird/STS-Mixer. --- ### MorphoFlow: Sparse-Supervised Generative Shape Modeling with Adaptive Latent Relevance Authors: Mokshagna Sai Teja Karanam, Tushar Kataria, Shireen Elhabian Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.11636v1 Statistical shape modeling (SSM) is central to population level analysis of anatomical variability, yet most existing approaches rely on densely annotated segmentations and fixed latent representations. These requirements limit scalability and reduce flexibility when modeling complex anatomical variation. We introduce MorphoFlow, a sparse supervised generative shape modeling framework that learns compact probabilistic shape representations directly from sparse surface annotations. MorphoFlow integrates neural implicit shape representations with an autodecoder formulation and autoregressive normalizing flows to learn an expressive probabilistic density over the latent shape space. The neural implicit representation enables resolution-agnostic modeling of 3D anatomy, while the autodecoder formulation supports direct optimization of per-instance latent codes under sparse supervision. The autoregressive flow captures the distribution of latent anatomical variability providing a tractable, likelihood-based generative model of shapes. To promote compact and structured latent representations, we incorporate adaptive latent relevance weighting through sparsity-inducing priors, enabling the model to regulate the contribution of individual latent dimensions according to their relevance to the underlying anatomical variation while preserving generative expressivity. The resulting latent space supports uncertainty quantification and anatomically plausible shape synthesis without manual latent dimensionality tuning. Evaluation on publicly available lumbar vertebrae and femur datasets demonstrates accurate high-resolution reconstruction from sparse inputs and recovery of structured modes of anatomical variation consistent with population level trends. --- ### CArtBench: Evaluating Vision-Language Models on Chinese Art Understanding, Interpretation, and Authenticity Authors: Xuefeng Wei, Zhixuan Wang, Xuan Zhou, Zhi Qu, Hongyao Li, Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe Published: 2026-04-13 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.11632v2 We introduce CARTBENCH, a museum-grounded benchmark for evaluating vision-language models (VLMs) on Chinese artworks beyond short-form recognition and QA. CARTBENCH comprises four subtasks: CURATORQA for evidence-grounded recognition and reasoning, CATALOGCAPTION for structured four-section expert-style appreciation, REINTERPRET for defensible reinterpretation with expert ratings, and CONNOISSEURPAIRS for diagnostic authenticity discrimination under visually similar confounds. CARTBENCH is built by aligning image-bearing Palace Museum objects from Wikidata with authoritative catalog pages, spanning five art categories across multiple dynasties. Across nine representative VLMs, we find that high overall CURATORQA accuracy can mask sharp drops on hard evidence linking and style-to-period inference; long-form appreciation remains far from expert references; and authenticity-oriented diagnostic discrimination stays near chance, underscoring the difficulty of connoisseur-level reasoning for current models. --- ### Back to Basics: Let Conversational Agents Remember with Just Retrieval and Generation Authors: Yuqian Wu, Wei Chen, Zhengjun Huang, Junle Chen, Qingxiang Liu, Kai Wang, Xiaofang Zhou, Yuxuan Liang Published: 2026-04-13 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.11628v1 Existing conversational memory systems rely on complex hierarchical summarization or reinforcement learning to manage long-term dialogue history, yet remain vulnerable to context dilution as conversations grow. In this work, we offer a different perspective: the primary bottleneck may lie not in memory architecture, but in the \textit{Signal Sparsity Effect} within the latent knowledge manifold. Through controlled experiments, we identify two key phenomena: \textit{Decisive Evidence Sparsity}, where relevant signals become increasingly isolated with longer sessions, leading to sharp degradation in aggregation-based methods; and \textit{Dual-Level Redundancy}, where both inter-session interference and intra-session conversational filler introduce large amounts of non-informative content, hindering effective generation. Motivated by these insights, we propose \method, a minimalist framework that brings conversational memory back to basics, relying solely on retrieval and generation via Turn Isolation Retrieval (TIR) and Query-Driven Pruning (QDP). TIR replaces global aggregation with a max-activation strategy to capture turn-level signals, while QDP removes redundant sessions and conversational filler to construct a compact, high-density evidence set. Extensive experiments on multiple benchmarks demonstrate that \method achieves robust performance across diverse settings, consistently outperforming strong baselines while maintaining high efficiency in tokens and latency, establishing a new minimalist baseline for conversational memory. --- ### POINTS-Long: Adaptive Dual-Mode Visual Reasoning in MLLMs Authors: Haicheng Wang, Yuan Liu, Yikun Liu, Zhemeng Yu, Zhongyin Zhao, Yangxiu You, Zilin Yu, Le Tian, Xiao Zhou, Jie Zhou, Weidi Xie, Yanfeng Wang Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.11627v1 Multimodal Large Language Models (MLLMs) have recently demonstrated remarkable capabilities in cross-modal understanding and generation. However, the rapid growth of visual token sequences--especially in long-video and streaming scenarios--poses a major challenge to their scalability and real-world deployment. Thus, we introduce POINTS-Long, a native dual-mode MLLM featuring dynamic visual token scaling inspired by the human visual system. The model supports two complementary perception modes: focus mode and standby mode, enabling users to dynamically trade off efficiency and accuracy during inference. On fine-grained visual tasks, the focus mode retains the optimal performance, while on long-form general visual understanding, the standby mode retains 97.7-99.7% of the original accuracy using only 1/40-1/10th of the visual tokens. Moreover, POINTS-Long natively supports streaming visual understanding via a dynamically detachable KV-cache design, allowing efficient maintenance of ultra-long visual memory. Our work provides new insights into the design of future MLLMs and lays the foundation for adaptive and efficient long-form visual understanding. --- ### RationalRewards: Reasoning Rewards Scale Visual Generation Both Training and Test Time Authors: Haozhe Wang, Cong Wei, Weiming Ren, Jiaming Liu, Fangzhen Lin, Wenhu Chen Published: 2026-04-13 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.11626v2 Most reward models for visual generation reduce rich human judgments to a single unexplained score, discarding the reasoning that underlies preference. We show that teaching reward models to produce explicit, multi-dimensional critiques before scoring transforms them from passive evaluators into active optimization tools, improving generators in two complementary ways: at training time, structured rationales provide interpretable, fine-grained rewards for reinforcement learning; at test time, a Generate-Critique-Refine loop turns critiques into targeted prompt revisions that improve outputs without any parameter updates. To train such a reward model without costly rationale annotations, we introduce Preference-Anchored Rationalization (PARROT), a principled framework that recovers high-quality rationales from readily available preference data through anchored generation, consistency filtering, and distillation. The resulting model, RationalRewards (8B), achieves state-of-the-art preference prediction among open-source reward models, competitive with Gemini-2.5-Pro, while using 10-20x less training data than comparable baselines. As an RL reward, it consistently improves text-to-image and image-editing generators beyond scalar alternatives. Most strikingly, its test-time critique-and-refine loop matches or exceeds RL-based fine-tuning on several benchmarks, suggesting that structured reasoning can unlock latent capabilities in existing generators that suboptimal prompts fail to elicit. --- ### SCNO: Spiking Compositional Neural Operator -- Towards a Neuromorphic Foundation Model for Nuclear PDE Solving Authors: Samrendra Roy, Souvik Chakraborty, Rizwan-uddin, Syed Bahauddin Alam Published: 2026-04-13 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.11625v1 Neural operators have emerged as powerful surrogates for partial differential equation (PDE) solvers, yet they are typically trained as monolithic models for individual PDEs, require energy-intensive GPU hardware, and must be retrained from scratch when new physics emerge. We introduce the Spiking Compositional Neural Operator (SCNO), a modular architecture combining spiking and conventional components that addresses all three limitations. SCNO maintains a library of small spiking neural operator blocks, each trained on a single elementary differential operator (convection, diffusion, reaction), and composes them through a lightweight input-conditioned aggregator to solve coupled PDEs not seen during block training. A small correction network learns cross-coupling residuals while keeping all blocks and the aggregator frozen, preserving zero-forgetting modular expansion by construction. We evaluate SCNO on eight PDE families including five coupled systems and a nuclear-relevant 1-group neutron diffusion equation. SCNO with correction achieves the lowest relative $L^2$ error on four of five coupled PDEs, outperforming both a monolithic spiking DeepONet (by up to 62%, mean over 3 seeds) and a standard ANN DeepONet (by up to 65%), while requiring only 95K trainable parameters versus 462K for the monolithic baseline. To our knowledge, this is the first compositional spiking neural operator and the first proof-of-concept for modular neuromorphic PDE solving with built-in forgetting-free expansion. --- ### Context Kubernetes: Declarative Orchestration of Enterprise Knowledge for Agentic AI Systems Authors: Charafeddine Mouzouni Published: 2026-04-13 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2604.11623v3 We introduce Context Kubernetes, an architecture for orchestrating enterprise knowledge in agentic AI systems, with a prototype implementation and eight experiments. The core observation is that delivering the right knowledge, to the right agent, with the right permissions, at the right freshness -- across an entire organization -- is structurally analogous to the container orchestration problem Kubernetes solved a decade ago. We formalize six core abstractions, a YAML-based declarative manifest for knowledge-architecture-as-code, a reconciliation loop, and a three-tier agent permission model where agent authority is always a strict subset of human authority. On synthetic seed data, we compare four governance baselines of increasing strength: ungoverned RAG, ACL-filtered retrieval, RBAC-aware routing, and the full architecture. Each layer contributes a different capability: ACL filtering eliminates cross-domain leaks, intent routing reduces noise by 19 percentage points, and only the three-tier model blocks all five tested attack scenarios -- the one attack RBAC misses is an agent sending confidential pricing via email, which RBAC cannot distinguish from ordinary email. TLA+ model-checking verifies safety properties across 4.6 million reachable states with zero violations. A survey of four major platforms (Microsoft, Salesforce, AWS, Google) documents that none architecturally isolates agent approval channels. We identify four properties that make context orchestration harder than container orchestration, and argue these make the solution more valuable. --- ### CUTEv2: Unified and Configurable Matrix Extension for Diverse CPU Architectures with Minimal Design Overhead Authors: Jinpeng Ye, Chongxi Wang, Wenqing Li, Bin Yuan, Shiyi Wang, Fenglu Zhang, Junyu Yue, Jianan Xie, Yunhao Ye, Haoyu Deng, Yingkun Zhou, Xin Cheng et al. Published: 2026-04-13 Categories: cs.AR, cs.AI, cs.DC, cs.LG arXiv: https://arxiv.org/abs/2604.11615v1 Matrix extensions have emerged as an essential feature in modern CPUs to address the surging demands of AI workloads. However, existing designs often incur substantial hardware and software design overhead. Tight coupling with the CPU pipeline complicates integration across diverse CPUs, while fine-grained synchronous instructions hinder the development of high-performance kernels. This paper proposes a unified and configurable CPU matrix extension architecture. By decoupling matrix units from the CPU pipeline, the design enables low-overhead integration while maintaining close coordination with existing compute and memory resources. The configurable matrix unit supports mixed-precision operations and adapts to diverse compute demands and memory bandwidth constraints. An asynchronous matrix multiplication abstraction with flexible granularity conceals hardware details, simplifies matrix-vector overlap, and supports a unified software stack. The architecture is integrated into four open-source CPU RTL platforms and evaluated on representative AI models. Matrix unit utilization under GEMM workloads exceeds 90% across all platforms. When configured with compute throughput and memory bandwidth comparable to Intel AMX, our design achieves speedups of 1.57x, 1.57x, and 2.31x on ResNet, BERT, and Llama3, with over 30% of the gains attributed to overlapped matrix-vector execution. A 4 TOPS@2GHz matrix unit occupies only 0.53 mm\textsuperscript{2} in 14nm CMOS. These results demonstrate strong cross-platform adaptability and effective hardware-software co-optimization, offering a practical matrix extension for the open-source community. --- ### Symmetry Reveals Layerwise Dynamics: How Transformers Perform In-Context Classification Authors: Patrick Lutz, Themistoklis Haris, Arjun Chandra, Aditya Gangrade, Venkatesh Saligrama Published: 2026-04-13 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.11613v4 Transformers can perform in-context classification from a few labeled examples, yet the inference-time algorithm remains opaque. We study multi-class linear classification in the hard no-margin regime and make the computation identifiable by enforcing feature- and label-permutation equivariance at every layer. This enables interpretability while maintaining functional equivalence and yields highly structured weights. From these models we extract an explicit depth-indexed recursion: an end-to-end identified, emergent update rule inside a softmax transformer, to our knowledge the first of its kind. Attention matrices formed from mixed feature-label Gram structure drive coupled updates of training points, labels, and the test probe. The resulting dynamics implement a geometry-driven algorithmic motif, which can provably amplify class separation and yields robust expected class alignment. --- ### Symmetry Reveals Layerwise Dynamics: How Transformers Perform In-Context Classification Authors: Patrick Lutz, Themistoklis Haris, Arjun Chandra, Aditya Gangrade, Venkatesh Saligrama Published: 2026-04-13 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.11613v3 Transformers can perform in-context classification from a few labeled examples, yet the inference-time algorithm remains opaque. We study multi-class linear classification in the hard no-margin regime and make the computation identifiable by enforcing feature- and label-permutation equivariance at every layer. This enables interpretability while maintaining functional equivalence and yields highly structured weights. From these models we extract an explicit depth-indexed recursion: an end-to-end identified, emergent update rule inside a softmax transformer, to our knowledge the first of its kind. Attention matrices formed from mixed feature-label Gram structure drive coupled updates of training points, labels, and the test probe. The resulting dynamics implement a geometry-driven algorithmic motif, which can provably amplify class separation and yields robust expected class alignment. --- ### Utilizing and Calibrating Hindsight Process Rewards via Reinforcement with Mutual Information Self-Evaluation Authors: Jiashu Yao, Heyan Huang, Zeming Liu, Yuhang Guo Published: 2026-04-13 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.11611v1 To overcome the sparse reward challenge in reinforcement learning (RL) for agents based on large language models (LLMs), we propose Mutual Information Self-Evaluation (MISE), an RL paradigm that utilizes hindsight generative self-evaluation as dense reward signals while simultaneously calibrating them against the environmental feedbacks. Empirically, MISE enables an agent to learn autonomously from dense internal rewards supplementing sparse extrinsic signals. Theoretically, our work provides the first formal foundation for the paradigm of generative self-rewarding. We prove that utilizing hindsight self-evaluation rewards is equivalent to minimizing an objective that combines mutual information with a KL divergence term between the policy and a proxy reward policy. This theoretical insight then informs and justifies our calibration step, which actively aligns these rewards with the optimal policy. Extensive experiments show that MISE outperforms strong baselines, enabling open-source LLMs about 7B parameters to achieve performance comparable to GPT-4o on validation without expert supervision. --- ### Self-Evolving LLM Memory Extraction Across Heterogeneous Tasks Authors: Yuqing Yang, Tengxiao Liu, Wang Bill Zhu, Taiwei Shi, Linxin Song, Robin Jia Published: 2026-04-13 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.11610v1 As LLM-based assistants become persistent and personalized, they must extract and retain useful information from past conversations as memory. However, the types of information worth remembering vary considerably across tasks. We formalize the \textit{heterogeneous memory extraction} task and introduce \textbf{BEHEMOTH}, a benchmark that repurposes 18 existing datasets spanning personalization, problem-solving, and agentic tasks, using a downstream utility-driven metric for systematic evaluation. Our empirical analysis confirms that no single static extraction prompt dominates across all task categories, and that existing self-evolving prompt optimization frameworks, originally designed for homogeneous distributions, degrade when training tasks are heterogeneous. To address this, we propose \textbf{CluE}, a cluster-based self-evolving strategy that groups training examples into clusters by extraction scenarios, analyzes each cluster independently, and synthesizes cross-cluster insights to update the extraction prompt. Experiments on BEHEMOTH show that CluE generalizes effectively across heterogeneous tasks ($+$9.04\% relative gain), consistently outperforming prior self-evolving frameworks. --- ### Intersectional Sycophancy: How Perceived User Demographics Shape False Validation in Large Language Models Authors: Benjamin Maltbie, Shivam Raval Published: 2026-04-13 Categories: cs.AI, cs.HC arXiv: https://arxiv.org/abs/2604.11609v2 Large language models exhibit sycophantic tendencies, but whether this behavior varies systematically with perceived user demographics is underexplored. Inspired by intersectionality (overlapping identities produce compounded effects), we probe whether frontier models conditionally exhibit sycophancy. Across 768 multi-turn conversations spanning 128 personas (varying race, age, gender, confidence) and three domains (mathematics, philosophy, conspiracy theories), we find that sycophancy varies sharply with target model and domain, and emerges from combinations of perceived user traits rather than any single dimension. GPT-5-nano scores far higher than Claude Haiku 4.5 (average sycophancy scores of $\bar{x}=2.96$ vs.\ $1.74$, $p < 10^{-32}$); within GPT-5-nano, philosophy elicits 41\% more sycophancy than mathematics and Hispanic personas receive the highest scores across races. The worst-scoring persona, a confident, 23-year-old Hispanic woman, averages 5.33/10 (max 6/10), while Claude Haiku 4.5 remains uniformly low with no significant demographic variation. We argue that safety evaluations should incorporate identity-aware adversarial testing. --- ### Geoparsing: Diagram Parsing for Plane and Solid Geometry with a Unified Formal Language Authors: Peijie Wang, Ming-Liang Zhang, Jun Cao, Chao Deng, Dekang Ran, Hongda Sun, Pi Bu, Xuan Zhang, Yingyao Wang, Jun Song, Bo Zheng, Fei Yin et al. Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.11600v2 Multimodal Large Language Models (MLLMs) have achieved remarkable progress but continue to struggle with geometric reasoning, primarily due to the perception bottleneck regarding fine-grained visual elements. While formal languages have aided plane geometry understanding, solid geometry which requires spatial understanding remains largely unexplored. In this paper, we address this challenge by designing a unified formal language that integrates plane and solid geometry, comprehensively covering geometric structures and semantic relations. We construct GDP-29K, a large-scale dataset comprising 20k plane and 9k solid geometry samples collected from diverse real-world sources, each paired with its ground-truth formal description. To ensure syntactic correctness and geometric consistency, we propose a training paradigm that combines Supervised Fine-Tuning with Reinforcement Learning via Verifiable Rewards. Experiments show that our approach achieves state-of-the-art parsing performance. Furthermore, we demonstrate that our parsed formal descriptions serve as a critical cognitive scaffold, significantly boosting MLLMs' capabilities for downstream geometry reasoning tasks. Our data and code are available at Geoparsing. --- ### Learning Robustness at Test-Time from a Non-Robust Teacher Authors: Stefano Bianchettin, Giulio Rossolini, Giorgio Buttazzo Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.11590v1 Nowadays, pretrained models are increasingly used as general-purpose backbones and adapted at test-time to downstream environments where target data are scarce and unlabeled. While this paradigm has proven effective for improving clean accuracy on the target domain, adversarial robustness has received far less attention, especially when the original pretrained model is not explicitly designed to be robust. This raises a practical question: \emph{can a pretrained, non-robust model be adapted at test-time to improve adversarial robustness on a target distribution?} To face this question, this work studies how adversarial training strategies behave when integrated into adaptation schemes for the unsupervised test-time setting, where only a small set of unlabeled target samples is available. It first analyzes how classical adversarial training formulations can be extended to this scenario, showing that straightforward distillation-based adaptations remain unstable and highly sensitive to hyperparameter tuning, particularly when the teacher itself is non-robust. To address these limitations, the work proposes a label-free framework that uses the predictions of a non-robust teacher model as a semantic anchor for both the clean and adversarial objectives during adaptation. We further provide theoretical insights showing that our formulation yields a more stable alternative to the self-consistency-based regularization commonly used in classical adversarial training. Experiments evaluate the proposed approach on CIFAR-10 and ImageNet under induced photometric transformations. The results support the theoretical insights by showing that the proposed approach achieves improved optimization stability, lower sensitivity to parameter choices, and a better robustness-accuracy trade-off than existing baselines in this post-deployment test-time setting. --- ### MLLM-as-a-Judge Exhibits Model Preference Bias Authors: Shuitsu Koyama, Yuiga Wada, Daichi Yashima, Komei Sugiura Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.11589v1 Automatic evaluation using multimodal large language models (MLLMs), commonly referred to as MLLM-as-a-Judge, has been widely used to measure model performance. If such MLLM-as-a-Judge methods were biased, they could distort model comparisons and benchmark-driven scientific progress. However, it remains unclear to what extent MLLM-as-a-Judge methods favor or disfavor text generated by specific MLLMs. In this study, we propose Philautia-Eval to investigate such model-specific preference bias. Philautia-Eval quantifies the degree of the bias by disentangling preference tendencies from differences in generation quality. Using 1.29M caption-score pairs collected from 12 MLLMs, we found that representative MLLMs tend to exhibit self-preference bias. Moreover, experimental results indicate mutual preference bias within particular model families, which is potentially driven by reused connectors and overlapping instruction-tuning resources. Finally, we introduce a simple ensemble of MLLMs, Pomms. Our results demonstrated that Pomms effectively mitigated the model-specific preference bias while maintaining performance. ---