# Research Papers — 2026-04 · bundle page 193 of 233 > Full-text markdown of papers #9601–9650 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/193 [← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/192.md) [Next bundle →](https://aipod.dev/papers/archive/2026-04/page/194.md) ### Part-Level 3D Gaussian Vehicle Generation with Joint and Hinge Axis Estimation Authors: Shiyao Qian, Yuan Ren, Dongfeng Bai, Bingbing Liu Published: 2026-04-06 Categories: cs.AI, cs.CV, cs.RO arXiv: https://arxiv.org/abs/2604.05070v1 Simulation is essential for autonomous driving, yet current frameworks often model vehicles as rigid assets and fail to capture part-level articulation. With perception algorithms increasingly leveraging dynamics such as wheel steering or door opening, realistic simulation requires animatable vehicle representations. Existing CAD-based pipelines are limited by library coverage and fixed templates, preventing faithful reconstruction of in-the-wild instances. We propose a generative framework that, from a single image or sparse multi-view input, synthesizes an animatable 3D Gaussian vehicle. Our method addresses two challenges: (i) large 3D asset generators are optimized for static quality but not articulation, leading to distortions at part boundaries when animated; and (ii) segmentation alone cannot provide the kinematic parameters required for motion. To overcome this, we introduce a part-edge refinement module that enforces exclusive Gaussian ownership and a kinematic reasoning head that predicts joint positions and hinge axes of movable parts. Together, these components enable faithful part-aware simulation, bridging the gap between static generation and animatable vehicle models. --- ### Towards Scaling Law Analysis For Spatiotemporal Weather Data Authors: Alexander Kiefer, Prasanna Balaprakash, Xiao Wang Published: 2026-04-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.05068v1 Compute-optimal scaling laws are relatively well studied for NLP and CV, where objectives are typically single-step and targets are comparatively homogeneous. Weather forecasting is harder to characterize in the same framework: autoregressive rollouts compound errors over long horizons, outputs couple many physical channels with disparate scales and predictability, and globally pooled test metrics can disagree sharply with per-channel, late-lead behavior implied by short-horizon training. We extend neural scaling analysis for autoregressive weather forecasting from single-step training loss to long rollouts and per-channel metrics. We quantify (1) how prediction error is distributed across channels and how its growth rate evolves with forecast horizon, (2) if power law scaling holds for test error, relative to rollout length when error is pooled globally, and (3) how that fit varies jointly with horizon and channel for parameter, data, and compute-based scaling axes. We find strong cross-channel and cross-horizon heterogeneity: pooled scaling can look favorable while many channels degrade at late leads. We discuss implications for weighted objectives, horizon-aware curricula, and resource allocation across outputs. --- ### AutoLALA: Automatic Loop Algebraic Locality Analysis for AI and HPC Kernels Authors: Yifan Zhu, Yekai Pan, Yanghui Wu, Chen Ding Published: 2026-04-06 Categories: cs.PL, cs.AI, cs.PF arXiv: https://arxiv.org/abs/2604.05066v1 Data movement is the primary bottleneck in modern computing systems. For loop-based programs common in high-performance computing (HPC) and AI workloads, including matrix multiplication, tensor contraction, stencil computation, and einsum operations, the cost of moving data through the memory hierarchy often exceeds the cost of arithmetic. This paper presents AutoLALA, an open-source tool that analyzes data locality in affine loop programs. The tool accepts programs written in a small domain-specific language (DSL), lowers them to polyhedral sets and maps, and produces closed-form symbolic formulas for reuse distance and data movement complexity. AutoLALA implements the fully symbolic locality analysis of Zhu et al. together with the data movement distance (DMD) framework of Smith et al. In particular, it computes reuse distance as the image of the access space under the access map, avoiding both stack simulation and Denning's recursive working-set formulation. We describe the DSL syntax and its formal semantics, the polyhedral lowering pipeline that constructs timestamp spaces and access maps via affine transformations, and the sequence of Barvinok counting operations used to derive symbolic reuse-interval and reuse-distance distributions. The system is implemented in Rust as a modular library spanning three crates, with safe bindings to the Barvinok library. We provide both a command-line interface and an interactive web playground with LaTeX rendering of the output formulas. The tool handles arbitrary affine loop nests, covering workloads such as tensor contractions, einsum expressions, stencil computations, and general polyhedral programs. --- ### Dynamic Linear Coregionalization for Realistic Synthetic Multivariate Time Series Authors: Annita Vapsi, Penghang Liu, Saheed Obitayo, Aakriti, Manoj Cherukumalli, Prathamesh Patil, Amit Varshney, Nicolas Marchesotti, Elizabeth Fons, Vamsi K. Potluru, Manuela Veloso Published: 2026-04-06 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.05064v2 Synthetic data is essential for training foundation models for time series (FMTS), but most generators assume static correlations, and are typically missing realistic inter-channel dependencies. We introduce DynLMC, a Dynamic Linear Model of Coregionalization, that incorporates time-varying, regime-switching correlations and cross-channel lag structures. Our approach produces synthetic multivariate time series with correlation dynamics that closely resemble real data. Fine-tuning three foundational models on DynLMC-generated data yields consistent zero-shot forecasting improvements across nine benchmarks. Our results demonstrate that modeling dynamic inter-channel correlations enhances FMTS transferability, highlighting the importance of data-centric pretraining. --- ### R3PM-Net: Real-time, Robust, Real-world Point Matching Network Authors: Yasaman Kashefbahrami, Erkut Akdag, Panagiotis Meletis, Evgeniya Balmashnova, Dip Goswami, Egor Bondarau Published: 2026-04-06 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2604.05060v2 Accurate Point Cloud Registration (PCR) is an important task in 3D data processing, involving the estimation of a rigid transformation between two point clouds. While deep-learning methods have addressed key limitations of traditional non-learning approaches, such as sensitivity to noise, outliers, occlusion, and initialization, they are developed and evaluated on clean, dense, synthetic datasets (limiting their generalizability to real-world industrial scenarios). This paper introduces R3PM-Net, a lightweight, global-aware, object-level point matching network designed to bridge this gap by prioritizing both generalizability and real-time efficiency. To support this transition, two datasets, Sioux-Cranfield and Sioux-Scans, are proposed. They provide an evaluation ground for registering imperfect photogrammetric and event-camera scans to digital CAD models, and have been made publicly available. Extensive experiments demonstrate that R3PM-Net achieves competitive accuracy with unmatched speed. On ModelNet40, it reaches a perfect fitness score of $1$ and inlier RMSE of $0.029$ cm in only $0.007$s, approximately 7 times faster than the state-of-the-art method RegTR. This performance carries over to the Sioux-Cranfield dataset, maintaining a fitness of $1$ and inlier RMSE of $0.030$ cm with similarly low latency. Furthermore, on the highly challenging Sioux-Scans dataset, R3PM-Net successfully resolves edge cases in under 50 ms. These results confirm that R3PM-Net offers a robust, high-speed solution for critical industrial applications, where precision and real-time performance are indispensable. The code and datasets are available at https://github.com/YasiiKB/R3PM-Net. --- ### Blind-Spot Mass: A Good-Turing Framework for Quantifying Deployment Coverage Risk in Machine Learning Systems Authors: Biplab Pal, Santanu Bhattacharya, Madanjit Singh Published: 2026-04-06 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2604.05057v1 Blind-spot mass is a Good-Turing framework for quantifying deployment coverage risk in machine learning. In modern ML systems, operational state distributions are often heavy-tailed, implying that a long tail of valid but rare states is structurally under-supported in finite training and evaluation data. This creates a form of 'coverage blindness': models can appear accurate on standard test sets yet remain unreliable across large regions of the deployment state space. We propose blind-spot mass B_n(tau), a deployment metric estimating the total probability mass assigned to states whose empirical support falls below a threshold tau. B_n(tau) is computed using Good-Turing unseen-species estimation and yields a principled estimate of how much of the operational distribution lies in reliability-critical, under-supported regimes. We further derive a coverage-imposed accuracy ceiling, decomposing overall performance into supported and blind components and separating capacity limits from data limits. We validate the framework in wearable human activity recognition (HAR) using wrist-worn inertial data. We then replicate the same analysis in the MIMIC-IV hospital database with 275 admissions, where the blind-spot mass curve converges to the same 95% at tau = 5 across clinical state abstractions. This replication across structurally independent domains - differing in modality, feature space, label space, and application - shows that blind-spot mass is a general ML methodology for quantifying combinatorial coverage risk, not an application-specific artifact. Blind-spot decomposition identifies which activities or clinical regimes dominate risk, providing actionable guidance for industrial practitioners on targeted data collection, normalization/renormalization, and physics- or domain-informed constraints for safer deployment. --- ### This Treatment Works, Right? Evaluating LLM Sensitivity to Patient Question Framing in Medical QA Authors: Hye Sun Yun, Geetika Kapoor, Michael Mackert, Ramez Kouzy, Wei Xu, Junyi Jessy Li, Byron C. Wallace Published: 2026-04-06 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.05051v1 Patients are increasingly turning to large language models (LLMs) with medical questions that are complex and difficult to articulate clearly. However, LLMs are sensitive to prompt phrasings and can be influenced by the way questions are worded. Ideally, LLMs should respond consistently regardless of phrasing, particularly when grounded in the same underlying evidence. We investigate this through a systematic evaluation in a controlled retrieval-augmented generation (RAG) setting for medical question answering (QA), where expert-selected documents are used rather than retrieved automatically. We examine two dimensions of patient query variation: question framing (positive vs. negative) and language style (technical vs. plain language). We construct a dataset of 6,614 query pairs grounded in clinical trial abstracts and evaluate response consistency across eight LLMs. Our findings show that positively- and negatively-framed pairs are significantly more likely to produce contradictory conclusions than same-framing pairs. This framing effect is further amplified in multi-turn conversations, where sustained persuasion increases inconsistency. We find no significant interaction between framing and language style. Our results demonstrate that LLM responses in medical QA can be systematically influenced through query phrasing alone, even when grounded in the same evidence, highlighting the importance of phrasing robustness as an evaluation criterion for RAG-based systems in high-stakes settings. --- ### PCA-Driven Adaptive Sensor Triage for Edge AI Inference Authors: Ankit Hemant Lade, Sai Krishna Jasti, Nikhil Sinha, Indar Kumar, Akanksha Tiwari Published: 2026-04-06 Categories: cs.LG, cs.AI, eess.SY, stat.ML arXiv: https://arxiv.org/abs/2604.05045v1 Multi-channel sensor networks in industrial IoT often exceed available bandwidth. We propose PCA-Triage, a streaming algorithm that converts incremental PCA loadings into proportional per-channel sampling rates under a bandwidth budget. PCA-Triage runs in O(wdk) time with zero trainable parameters (0.67 ms per decision). We evaluate on 7 benchmarks (8--82 channels) against 9 baselines. PCA-Triage is the best unsupervised method on 3 of 6 datasets at 50% bandwidth, winning 5 of 6 against every baseline with large effect sizes (r = 0.71--0.91). On TEP, it achieves F1 = 0.961 +/- 0.001 -- within 0.1% of full-data performance -- while maintaining F1 > 0.90 at 30% budget. Targeted extensions push F1 to 0.970. The algorithm is robust to packet loss and sensor noise (3.7--4.8% degradation under combined worst-case). --- ### Energy-Based Dynamical Models for Neurocomputation, Learning, and Optimization Authors: Arthur N. Montanari, Francesco Bullo, Dmitry Krotov, Adilson E. Motter Published: 2026-04-06 Categories: cs.LG, cond-mat.dis-nn, eess.SY, math.DS, q-bio.NC arXiv: https://arxiv.org/abs/2604.05042v1 Recent advances at the intersection of control theory, neuroscience, and machine learning have revealed novel mechanisms by which dynamical systems perform computation. These advances encompass a wide range of conceptual, mathematical, and computational ideas, with applications for model learning and training, memory retrieval, data-driven control, and optimization. This tutorial focuses on neuro-inspired approaches to computation that aim to improve scalability, robustness, and energy efficiency across such tasks, bridging the gap between artificial and biological systems. Particular emphasis is placed on energy-based dynamical models that encode information through gradient flows and energy landscapes. We begin by reviewing classical formulations, such as continuous-time Hopfield networks and Boltzmann machines, and then extend the framework to modern developments. These include dense associative memory models for high-capacity storage, oscillator-based networks for large-scale optimization, and proximal-descent dynamics for composite and constrained reconstruction. The tutorial demonstrates how control-theoretic principles can guide the design of next-generation neurocomputing systems, steering the discussion beyond conventional feedforward and backpropagation-based approaches to artificial intelligence. --- ### ID-Sim: An Identity-Focused Similarity Metric Authors: Julia Chae, Nicholas Kolkin, Jui-Hsien Wang, Richard Zhang, Sara Beery, Cusuh Ham Published: 2026-04-06 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.05039v1 Humans have remarkable selective sensitivity to identities -- easily distinguishing between highly similar identities, even across significantly different contexts such as diverse viewpoints or lighting. Vision models have struggled to match this capability, and progress toward identity-focused tasks such as personalized image generation is slowed by a lack of identity-focused evaluation metrics. To help facilitate progress, we propose ID-Sim, a feed-forward metric designed to faithfully reflect human selective sensitivity. To build ID-Sim, we curate a high-quality training set of images spanning diverse real-world domains, augmented with generative synthetic data that provides controlled, fine-grained identity and contextual variations. We evaluate our metric on a new unified evaluation benchmark for assessing consistency with human annotations across identity-focused recognition, retrieval, and generative tasks. --- ### Learning to Unscramble Feynman Loop Integrals with SAILIR Authors: David Shih Published: 2026-04-06 Categories: hep-ph, cs.LG, hep-th arXiv: https://arxiv.org/abs/2604.05034v1 Integration-by-parts (IBP) reduction of Feynman integrals to master integrals is a key computational bottleneck in precision calculations in high-energy physics. Traditional approaches based on the Laporta algorithm require solving large systems of equations, leading to memory consumption that grows rapidly with integral complexity. We present SAILIR (Self-supervised AI for Loop Integral Reduction), a new machine learning approach in which a transformer-based classifier guides the reduction of integrals one step at a time in a fully online fashion. The classifier is trained in an entirely self-supervised manner on synthetic data generated by a scramble/unscramble procedure: known reduction identities are applied in reverse to build expressions of increasing complexity, and the classifier learns to undo these steps. When combined with beam search and a highly parallelized, asynchronous, single-episode reduction strategy, SAILIR can reduce integrals of arbitrarily high weight with bounded memory. We benchmark SAILIR on the two-loop triangle-box topology, comparing against the state-of-the-art IBP reduction code Kira across 16 integrals of varying complexity. While SAILIR is slower in wall-clock time, its per-worker memory consumption remains approximately flat regardless of integral complexity, in contrast to Kira whose memory grows rapidly with complexity. For the most complex integrals considered here, SAILIR uses only 40\% of the memory of Kira while achieving comparable reduction times. This demonstrates a fundamentally new paradigm for IBP reduction in which the memory bottleneck of Laporta-based approaches could be entirely overcome, potentially opening the door to precision calculations that are currently intractable. --- ### Phase-Associative Memory: Sequence Modeling in Complex Hilbert Space Authors: Gowrav Vishwakarma, Christopher J. Agostino Published: 2026-04-06 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.05030v2 Experiments probing natural language processing by both humans and LLMs suggest that the meaning of a semantic expression is indeterminate prior to the act of interpretation rather than being specifiable simply as the sum of its parts (i.e. compositionality). This observer-dependent act dynamically actualizes meaning under genuine contextuality more consistent with quantum logical mechanisms than with classical Boolean approaches that assume separability, motivating an approach to language modeling that utilizes a Hilbert space formalism. In this work, we introduce Phase-Associative Memory (PAM) -- a complex-valued sequence model whose state S_t \in \mathbb{C}^{d \times d} accumulates outer products of complex token embeddings retrieved through the conjugate inner product $\mathrm{Re}\langle K \mid Q\rangle / \sqrt{d}$ -- and evaluate it against a structurally matched real-valued ablation. Both architectures train stably across a 5M--100M parameter sweep on WikiText-103 under identical conditions; PAM sits at higher absolute loss at every measured scale but improves more rapidly with parameter count, with power-law exponents of $-0.15$ vs.\ $-0.12$ in loss and $-0.65$ vs.\ $-0.49$ in perplexity that narrow the gap between the two architectures monotonically. Further investigation of complex-valued sequence modeling at larger scales could reveal that the loss plateau characteristic of real-valued state-of-the-art language models (e.g. transformers) is reachable with PAM-style architectures with an order of magnitude fewer parameters than the current frontier ($\sim$1T), implying that similar capabilities are achievable at sizes runnable on consumer-grade hardware. --- ### PaperOrchestra: A Multi-Agent Framework for Automated AI Research Paper Writing Authors: Yiwen Song, Yale Song, Tomas Pfister, Jinsung Yoon Published: 2026-04-06 Categories: cs.AI, cs.LG, cs.MA arXiv: https://arxiv.org/abs/2604.05018v1 Synthesizing unstructured research materials into manuscripts is an essential yet under-explored challenge in AI-driven scientific discovery. Existing autonomous writers are rigidly coupled to specific experimental pipelines, and produce superficial literature reviews. We introduce PaperOrchestra, a multi-agent framework for automated AI research paper writing. It flexibly transforms unconstrained pre-writing materials into submission-ready LaTeX manuscripts, including comprehensive literature synthesis and generated visuals, such as plots and conceptual diagrams. To evaluate performance, we present PaperWritingBench, the first standardized benchmark of reverse-engineered raw materials from 200 top-tier AI conference papers, alongside a comprehensive suite of automated evaluators. In side-by-side human evaluations, PaperOrchestra significantly outperforms autonomous baselines, achieving an absolute win rate margin of 50%-68% in literature review quality, and 14%-38% in overall manuscript quality. --- ### Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding Authors: Chaoyou Fu, Haozhi Yuan, Yuhao Dong, Yi-Fan Zhang, Yunhang Shen, Xiaoxing Hu, Xueying Li, Jinsen Su, Chengwu Long, Xiaoyao Xie, Yongkang Xie, Xiawu Zheng et al. Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05015v1 With the rapid advancement of video understanding, existing benchmarks are becoming increasingly saturated, exposing a critical discrepancy between inflated leaderboard scores and real-world model capabilities. To address this widening gap, we introduce Video-MME-v2, a comprehensive benchmark designed to rigorously evaluate the robustness and faithfulness of video understanding. To systematically evaluate model capabilities, we design a \textbf{progressive tri-level hierarchy} that incrementally increases the complexity of video comprehension, ranging from multi-point visual information aggregation, to temporal dynamics modeling, and ultimately to complex multimodal reasoning. Besides, in contrast to conventional per-question accuracy, we propose a \textbf{group-based non-linear evaluation} strategy that enforces both consistency across related queries and coherence in multi-step reasoning. It penalizes fragmented or guess-based correctness and assigns credit only to answers supported by valid reasoning. To guarantee data quality, Video-MME-v2 is constructed through a rigorously controlled human annotation pipeline, involving 12 annotators and 50 independent reviewers. Backed by \textbf{3,300 human-hours} and up to \textbf{5 rounds} of quality assurance, Video-MME-v2 aims to serve as one of the most authoritative video benchmarks. Extensive experiments reveal a substantial gap between current best model Gemini-3-Pro and human experts, and uncover a clear hierarchical bottleneck where errors in visual information aggregation and temporal modeling propagate to limit high-level reasoning. We further find that thinking-based reasoning is highly dependent on textual cues, improving performance with subtitles but sometimes degrading it in purely visual settings. By exposing these limitations, Video-MME-v2 establishes a demanding new testbed for the development of next-generation video MLLMs. --- ### StarVLA: A Lego-like Codebase for Vision-Language-Action Model Developing Authors: StarVLA Community Published: 2026-04-06 Categories: cs.RO, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2604.05014v1 Building generalist embodied agents requires integrating perception, language understanding, and action, which are core capabilities addressed by Vision-Language-Action (VLA) approaches based on multimodal foundation models, including recent advances in vision-language models and world models. Despite rapid progress, VLA methods remain fragmented across incompatible architectures, codebases, and evaluation protocols, hindering principled comparison and reproducibility. We present StarVLA, an open-source codebase for VLA research. StarVLA addresses these challenges in three aspects. First, it provides a modular backbone--action-head architecture that supports both VLM backbones (e.g., Qwen-VL) and world-model backbones (e.g., Cosmos) alongside representative action-decoding paradigms, all under a shared abstraction in which backbone and action head can each be swapped independently. Second, it provides reusable training strategies, including cross-embodiment learning and multimodal co-training, that apply consistently across supported paradigms. Third, it integrates major benchmarks, including LIBERO, SimplerEnv, RoboTwin~2.0, RoboCasa-GR1, and BEHAVIOR-1K, through a unified evaluation interface that supports both simulation and real-robot deployment. StarVLA also ships simple, fully reproducible single-benchmark training recipes that, despite minimal data engineering, already match or surpass prior methods on multiple benchmarks with both VLM and world-model backbones. To our best knowledge, StarVLA is one of the most comprehensive open-source VLA frameworks available, and we expect it to lower the barrier for reproducing existing methods and prototyping new ones. StarVLA is being actively maintained and expanded; we will update this report as the project evolves. The code and documentation are available at https://github.com/starVLA/starVLA. --- ### Scaling Coding Agents via Atomic Skills Authors: Yue Liu Published: 2026-04-06 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2604.05013v2 Current LLM coding agents are predominantly trained on composite benchmarks (e.g., bug fixing), which often leads to task-specific overfitting and limited generalization. To address this, we propose a novel scaling paradigm that shifts the focus from task-level optimization to atomic skill mastery. We first formalize five fundamental atomic skills, code localization, code editing, unit-test generation, issue reproduction, and code review, that serve as the basis vectors for complex software engineering tasks. Compared with composite coding tasks, these atomic skills are more generalizable and composable. Then, we scale coding agents by performing joint RL over atomic skills. In this manner, atomic skills are consistently improved without negative interference or trade-offs between them. Notably, we observe that improvements in these atomic skills generalize well to other unseen composite coding tasks, such as bug-fixing, code refactoring, machine learning engineering, and code security. The observation motivates a new scaling paradigm for coding agents by training with atomic skills. Extensive experiments demonstrate the effectiveness of our proposed paradigm. Notably, our joint RL improves average performance by 18.7% on 5 atomic skills and 5 composite tasks. --- ### Comparative Characterization of KV Cache Management Strategies for LLM Inference Authors: Oteo Mamo, Olga Kogiou, Hyunjin Yi, Weikuan Yu Published: 2026-04-06 Categories: cs.AR, cs.AI arXiv: https://arxiv.org/abs/2604.05012v2 Efficient inference with Large Language Models (LLMs) increasingly relies on Key-Value (KV) caches to store previously computed key and value vectors at each layer. These caches are essential to minimize redundant computation during autoregressive token generation, lowering computational complexity from quadratic to linear. However, the growth of KV caches has posed significant system-level challenges, particularly as model sizes increase, context lengths grow, and concurrent requests compete for limited memory resources. Even though several recent frameworks for KV cache management have emerged, their comparative trade-offs in memory consumption and inference performance have not been fully understood, especially under varying request sizes and model configurations. In this work, we conduct an empirical study of three state-of-the-art KV cache management frameworks: vLLM, InfiniGen, and H2O. These frameworks employ techniques such as tensor offloading, token eviction heuristics, and speculative scheduling to balance memory usage and performance. We evaluate their performance in terms of a range of metrics such as latency, throughput, and memory usage across a spectrum of key parameters including request rates, model sizes, and sparsity levels. Our results pinpoint the conditions for each framework to perform the best, revealing the most suitable selection and configuration of KV cache strategies under memory and performance constraints. --- ### Comparative Characterization of KV Cache Management Strategies for LLM Inference Authors: Oteo Mamo, Olga Kogiou, Hyunjin Yi, Weikuan Yu Published: 2026-04-06 Categories: cs.AR, cs.AI arXiv: https://arxiv.org/abs/2604.05012v1 Efficient inference with Large Language Models (LLMs) increasingly relies on Key-Value (KV) caches to store previously computed key and value vectors at each layer. These caches are essential to minimize redundant computation during autoregressive token generation, lowering computational complexity from quadratic to linear. However, the growth of KV caches has posed significant system-level challenges, particularly as model sizes increase, context lengths grow, and concurrent requests compete for limited memory resources. Even though several recent frameworks for KV cache management have emerged, their comparative trade-offs in memory consumption and inference performance have not been fully understood, especially under varying request sizes and model configurations. In this work, we conduct an empirical study of three state-of-the-art KV cache management frameworks: vLLM, InfiniGen, and H2O. These frameworks employ techniques such as tensor offloading, token eviction heuristics, and speculative scheduling to balance memory usage and performance. We evaluate their performance in terms of a range of metrics such as latency, throughput, and memory usage across a spectrum of key parameters including request rates, model sizes, and sparsity levels. Our results pinpoint the conditions for each framework to perform the best, revealing the most suitable selection and configuration of KV cache strategies under memory and performance constraints. --- ### YMIR: A new Benchmark Dataset and Model for Arabic Yemeni Music Genre Classification Using Convolutional Neural Networks Authors: Moeen AL-Makhlafi, Abdulrahman A. AlKannad, Eiad Almekhlafi, Nawaf Q. Othman Ahmed Mohammed, Saher Qaid Published: 2026-04-06 Categories: cs.SD, cs.AI arXiv: https://arxiv.org/abs/2604.05011v1 Automatic music genre classification is a major task in music information retrieval; however, most current benchmarks and models have been developed primarily for Western music, leaving culturally specific traditions underrepresented. In this paper, we introduce the Yemeni Music Information Retrieval (YMIR) dataset, which contains 1,475 carefully selected audio clips covering five traditional Yemeni genres: Sanaani, Hadhrami, Lahji, Tihami, and Adeni. The dataset was labeled by five Yemeni music experts following a clear and structured protocol, resulting in strong inter-annotator agreement (Fleiss kappa = 0.85). We also propose the Yemeni Music Classification Model (YMCM), a convolutional neural network (CNN)-based system designed to classify music genres from time-frequency features. Using a consistent preprocessing pipeline, we perform a systematic comparison across six experimental groups and five different architectures, resulting in a total of 30 experiments. Specifically, we evaluate several feature representations, including Mel-spectrograms, Chroma, FilterBank, and MFCCs with 13, 20, and 40 coefficients, and benchmark YMCM against standard models (AlexNet, VGG16, MobileNet, and a baseline CNN) under the same experimental conditions. The experimental findings reveal that YMCM is the most effective, achieving the highest accuracy of 98.8% with Mel-spectrogram features. The results also provide practical insights into the relationship between feature representation and model capacity. The findings establish YMIR as a useful benchmark and YMCM as a strong baseline for classifying Yemeni music genres. --- ### Generative Path-Law Jump-Diffusion: Sequential MMD-Gradient Flows and Generalisation Bounds in Marcus-Signature RKHS Authors: Daniel Bloch Published: 2026-04-06 Categories: stat.ML, cs.LG, q-fin.MF, q-fin.ST arXiv: https://arxiv.org/abs/2604.05008v1 This paper introduces a novel generative framework for synthesising forward-looking, càdlàg stochastic trajectories that are sequentially consistent with time-evolving path-law proxies, thereby incorporating anticipated structural breaks, regime shifts, and non-autonomous dynamics. By framing path synthesis as a sequential matching problem on restricted Skorokhod manifolds, we develop the \textit{Anticipatory Neural Jump-Diffusion} (ANJD) flow, a generative mechanism that effectively inverts the time-extended Marcus-sense signature. Central to this approach is the Anticipatory Variance-Normalised Signature Geometry (AVNSG), a time-evolving precision operator that performs dynamic spectral whitening on the signature manifold to ensure contractivity during volatile regime shifts and discrete aleatoric shocks. We provide a rigorous theoretical analysis demonstrating that the joint generative flow constitutes an infinitesimal steepest descent direction for the Maximum Mean Discrepancy functional relative to a moving target proxy. Furthermore, we establish statistical generalisation bounds within the restricted path-space and analyse the Rademacher complexity of the whitened signature functionals to characterise the expressive power of the model under heavy-tailed innovations. The framework is implemented via a scalable numerical scheme involving Nyström-compressed score-matching and an anticipatory hybrid Euler-Maruyama-Marcus integration scheme. Our results demonstrate that the proposed method captures the non-commutative moments and high-order stochastic texture of complex, discontinuous path-laws with high computational efficiency. --- ### Generalizable Audio-Visual Navigation via Binaural Difference Attention and Action Transition Prediction Authors: Jia Li, Yinfeng Yu Published: 2026-04-06 Categories: cs.SD, cs.AI, eess.AS arXiv: https://arxiv.org/abs/2604.05007v1 In Audio-Visual Navigation (AVN), agents must locate sound sources in unseen 3D environments using visual and auditory cues. However, existing methods often struggle with generalization in unseen scenarios, as they tend to overfit to semantic sound features and specific training environments. To address these challenges, we propose the \textbf{Binaural Difference Attention with Action Transition Prediction (BDATP)} framework, which jointly optimizes perception and policy. Specifically, the \textbf{Binaural Difference Attention (BDA)} module explicitly models interaural differences to enhance spatial orientation, reducing reliance on semantic categories. Simultaneously, the \textbf{Action Transition Prediction (ATP)} task introduces an auxiliary action prediction objective as a regularization term, mitigating environment-specific overfitting. Extensive experiments on the Replica and Matterport3D datasets demonstrate that BDATP can be seamlessly integrated into various mainstream baselines, yielding consistent and significant performance gains. Notably, our framework achieves state-of-the-art Success Rates across most settings, with a remarkable absolute improvement of up to 21.6 percentage points in Replica dataset for unheard sounds. These results underscore BDATP's superior generalization capability and its robustness across diverse navigation architectures. --- ### EduIllustrate: Towards Scalable Automated Generation Of Multimodal Educational Content Authors: Shuzhen Bi, Mingzi Zhang, Zhuoxuan Li, Xiaolong Wang, Keqian Li, Aimin Zhou Published: 2026-04-06 Categories: cs.CY, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.05005v2 Large language models are increasingly used as educational assistants, yet evaluation of their educational capabilities remains concentrated on question-answering and tutoring tasks. A critical gap exists for multimedia instructional content generation -- the ability to produce coherent, diagram-rich explanations that combine geometrically accurate visuals with step-by-step reasoning. We present EduIllustrate, a benchmark for evaluating LLMs on interleaved text-diagram explanation generation for K-12 STEM problems. The benchmark comprises 230 problems spanning five subjects and three grade levels, a standardized generation protocol with sequential anchoring to enforce cross-diagram visual consistency, and an 8-dimension evaluation rubric grounded in multimedia learning theory covering both text and visual quality. Evaluation of ten LLMs reveals a wide performance spread: Gemini 3.0 Pro Preview leads at 87.8\%, while Kimi-K2.5 achieves the best cost-efficiency (80.8\% at \\$0.12/problem). Workflow ablation confirms sequential anchoring improves Visual Consistency by 13\% at 94\% lower cost. Human evaluation with 20 expert raters validates LLM-as-judge reliability for objective dimensions ($ρ\geq 0.83$) while revealing limitations on subjective visual assessment. --- ### Vanast: Virtual Try-On with Human Image Animation via Synthetic Triplet Supervision Authors: Hyunsoo Cha, Wonjung Woo, Byungjun Kim, Hanbyul Joo Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04934v2 We present Vanast, a unified framework that generates garment-transferred human animation videos directly from a single human image, garment images, and a pose guidance video. Conventional two-stage pipelines treat image-based virtual try-on and pose-driven animation as separate processes, which often results in identity drift, garment distortion, and front-back inconsistency. Our model addresses these issues by performing the entire process in a single unified step to achieve coherent synthesis. To enable this setting, we construct large-scale triplet supervision. Our data generation pipeline includes generating identity-preserving human images in alternative outfits that differ from garment catalog images, capturing full upper and lower garment triplets to overcome the single-garment-posed video pair limitation, and assembling diverse in-the-wild triplets without requiring garment catalog images. We further introduce a Dual Module architecture for video diffusion transformers to stabilize training, preserve pretrained generative quality, and improve garment accuracy, pose adherence, and identity preservation while supporting zero-shot garment interpolation. Together, these contributions allow Vanast to produce high-fidelity, identity-consistent animation across a wide range of garment types. --- ### PointTPA: Dynamic Network Parameter Adaptation for 3D Scene Understanding Authors: Siyuan Liu, Chaoqun Zheng, Xin Zhou, Tianrui Feng, Dingkang Liang, Xiang Bai Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04933v1 Scene-level point cloud understanding remains challenging due to diverse geometries, imbalanced category distributions, and highly varied spatial layouts. Existing methods improve object-level performance but rely on static network parameters during inference, limiting their adaptability to dynamic scene data. We propose PointTPA, a Test-time Parameter Adaptation framework that generates input-aware network parameters for scene-level point clouds. PointTPA adopts a Serialization-based Neighborhood Grouping (SNG) to form locally coherent patches and a Dynamic Parameter Projector (DPP) to produce patch-wise adaptive weights, enabling the backbone to adjust its behavior according to scene-specific variations while maintaining a low parameter overhead. Integrated into the PTv3 structure, PointTPA demonstrates strong parameter efficiency by introducing two lightweight modules of less than 2% of the backbone's parameters. Despite this minimal parameter overhead, PointTPA achieves 78.4% mIoU on ScanNet validation, surpassing existing parameter-efficient fine-tuning (PEFT) methods across multiple benchmarks, highlighting the efficacy of our test-time dynamic network parameter adaptation mechanism in enhancing 3D scene understanding. The code is available at https://github.com/H-EmbodVis/PointTPA. --- ### Beyond the Final Actor: Modeling the Dual Roles of Creator and Editor for Fine-Grained LLM-Generated Text Detection Authors: Yang Li, Qiang Sheng, Zhengjia Wang, Yehan Yang, Danding Wang, Juan Cao Published: 2026-04-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.04932v3 The misuse of large language models (LLMs) requires precise detection of synthetic text. Existing works mainly follow binary or ternary classification settings, which can only distinguish pure human/LLM text or collaborative text at best. This remains insufficient for the nuanced regulation, as the LLM-polished human text and humanized LLM text often trigger different policy consequences. In this paper, we explore fine-grained LLM-generated text detection under a rigorous four-class setting. To handle such complexities, we propose RACE (Rhetorical Analysis for Creator-Editor Modeling), a fine-grained detection method that characterizes the distinct signatures of creator and editor. Specifically, RACE utilizes Rhetorical Structure Theory (RST) to construct a logic graph for the creator's foundation while extracting Elementary Discourse Unit (EDU)-level features for the editor's style. Experiments show that RACE outperforms 12 baselines in identifying fine-grained types with low false alarms, offering a policy-aligned solution for LLM regulation. --- ### LoMa: Local Feature Matching Revisited Authors: David Nordström, Johan Edstedt, Georg Bökman, Jonathan Astermark, Anders Heyden, Viktor Larsson, Mårten Wadenbäck, Michael Felsberg, Fredrik Kahl Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04931v2 Local feature matching has long been a fundamental component of 3D vision systems such as Structure-from-Motion (SfM), yet progress has lagged behind the rapid advances of modern data-driven approaches. The newer approaches, such as feed-forward reconstruction models, have benefited extensively from scaling dataset sizes, whereas local feature matching models are still only trained on a few mid-sized datasets. In this paper, we revisit local feature matching from a data-driven perspective. In our approach, which we call LoMa, we combine large and diverse data mixtures, modern training recipes, scaled model capacity, and scaled compute, resulting in remarkable gains in performance. Since current standard benchmarks mainly rely on collecting sparse views from successful 3D reconstructions, the evaluation of progress in feature matching has been limited to relatively easy image pairs. To address the resulting saturation of benchmarks, we collect 1000 highly challenging image pairs from internet data into a new dataset called HardMatch. Ground truth correspondences for HardMatch are obtained via manual annotation by the authors. In our extensive benchmarking suite, we find that LoMa makes outstanding progress across the board, outperforming the state-of-the-art method ALIKED+LightGlue by +18.6 mAA on HardMatch, +29.5 mAA on WxBS, +21.4 (1m, 10$^\circ$) on InLoc, +24.2 AUC on RUBIK, and +12.4 mAA on IMC 2022. We release our code and models publicly at https://github.com/davnords/LoMa. --- ### Early Stopping for Large Reasoning Models via Confidence Dynamics Authors: Parsa Hosseini, Sumit Nawathe, Mahdi Salmani, Meisam Razaviyayn, Soheil Feizi Published: 2026-04-06 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.04930v2 Large reasoning models rely on long chain-of-thought generation to solve complex problems, but extended reasoning often incurs substantial computational cost and can even degrade performance due to overthinking. A key challenge is determining when the model should stop reasoning and produce the final answer. In this work, we study the confidence of intermediate answers during reasoning and observe two characteristic behaviors: correct reasoning trajectories often reach high-confidence answers early, while incorrect rollouts tend to produce long, unproductive reasoning traces and exhibit less reliable confidence dynamics. Motivated by these observations, we propose CoDE-Stop (Confidence Dynamics Early Stop), an early stopping method that leverages the dynamics of intermediate answer confidence to decide when to terminate reasoning, requiring no additional training and easily integrating into existing models. We evaluate CoDE-Stop on diverse reasoning and science benchmarks across multiple models. Compared to prior early stopping methods, it achieves a more favorable accuracy-compute tradeoff and reduces total token usage by 25-50% compared to standard full-length reasoning. In addition, we provide analyses of confidence dynamics during reasoning, offering insights into how confidence changes in both correct and incorrect trajectories. --- ### Early Stopping for Large Reasoning Models via Confidence Dynamics Authors: Parsa Hosseini, Sumit Nawathe, Mahdi Salmani, Meisam Razaviyayn, Soheil Feizi Published: 2026-04-06 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.04930v1 Large reasoning models rely on long chain-of-thought generation to solve complex problems, but extended reasoning often incurs substantial computational cost and can even degrade performance due to overthinking. A key challenge is determining when the model should stop reasoning and produce the final answer. In this work, we study the confidence of intermediate answers during reasoning and observe two characteristic behaviors: correct reasoning trajectories often reach high-confidence answers early, while incorrect rollouts tend to produce long, unproductive reasoning traces and exhibit less reliable confidence dynamics. Motivated by these observations, we propose CoDE-Stop (Confidence Dynamics Early Stop), an early stopping method that leverages the dynamics of intermediate answer confidence to decide when to terminate reasoning, requiring no additional training and easily integrating into existing models. We evaluate CoDE-Stop on diverse reasoning and science benchmarks across multiple models. Compared to prior early stopping methods, it achieves a more favorable accuracy-compute tradeoff and reduces total token usage by 25-50% compared to standard full-length reasoning. In addition, we provide analyses of confidence dynamics during reasoning, offering insights into how confidence changes in both correct and incorrect trajectories. --- ### Rethinking Model Efficiency: Multi-Agent Inference with Large Models Authors: Sixun Dong, Juhua Hu, Steven Li, Wei Wen, Qi Qian Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04929v1 Most vision-language models (VLMs) apply a large language model (LLM) as the decoder, where the response tokens are generated sequentially through autoregression. Therefore, the number of output tokens can be the bottleneck of the end-to-end latency. However, different models may require vastly different numbers of output tokens to achieve comparable performance. In this work, we conduct a comprehensive analysis of the latency across different components of VLMs on simulated data. The experiment shows that a large model with fewer output tokens can be more efficient than a small model with a long output sequence. The empirical study on diverse real-world benchmarks confirms the observation that a large model can achieve better or comparable performance as a small model with significantly fewer output tokens. To leverage the efficiency of large models, we propose a multi-agent inference framework that keeps large models with short responses but transfers the key reasoning tokens from the small model when necessary. The comparison on benchmark tasks demonstrates that by reusing the reasoning tokens from small models, it can help approach the performance of a large model with its own reasoning, which confirms the effectiveness of our proposal. --- ### SimpleProc: Fully Procedural Synthetic Data from Simple Rules for Multi-View Stereo Authors: Zeyu Ma, Alexander Raistrick, Jia Deng Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04925v3 Generating procedural synthetic data for multi-view stereo (MVS) usually requires writing complex rules to match the realism of curated datasets. We demonstrate that we can generate effective training data using SimpleProc: a new, fully procedural generator driven by a very small set of rules based on Non-Uniform Rational Basis Splines (NURBS), as well as simple displacement and texture patterns. At a modest scale of 8,000 images, our approach achieves superior results compared to manually curated images at the same scale sourced from games and real-world objects. When scaled to 352,000 images, our approach achieves similar, and in some benchmarks, even better results than the current state-of-the-art trained on over 692,000 manually curated images. The source code and the data are available at https://github.com/princeton-vl/SimpleProc. --- ### SimpleProc: Fully Procedural Synthetic Data from Simple Rules for Multi-View Stereo Authors: Zeyu Ma, Alexander Raistrick, Jia Deng Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04925v2 In this paper, we explore the design space of procedural rules for multi-view stereo (MVS). We demonstrate that we can generate effective training data using SimpleProc: a new, fully procedural generator driven by a very small set of rules using Non-Uniform Rational Basis Splines (NURBS), as well as basic displacement and texture patterns. At a modest scale of 8,000 images, our approach achieves superior results compared to manually curated images (at the same scale) sourced from games and real-world objects. When scaled to 352,000 images, our method yields performance comparable to--and in several benchmarks, exceeding--models trained on over 692,000 manually curated images. The source code and the data are available at https://github.com/princeton-vl/SimpleProc. --- ### Your Pre-trained Diffusion Model Secretly Knows Restoration Authors: Sudarshan Rajagopalan, Vishal M. Patel Published: 2026-04-06 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.04924v1 Pre-trained diffusion models have enabled significant advancements in All-in-One Restoration (AiOR), offering improved perceptual quality and generalization. However, diffusion-based restoration methods primarily rely on fine-tuning or Control-Net style modules to leverage the pre-trained diffusion model's priors for AiOR. In this work, we show that these pre-trained diffusion models inherently possess restoration behavior, which can be unlocked by directly learning prompt embeddings at the output of the text encoder. Interestingly, this behavior is largely inaccessible through text prompts and text-token embedding optimization. Furthermore, we observe that naive prompt learning is unstable because the forward noising process using degraded images is misaligned with the reverse sampling trajectory. To resolve this, we train prompts within a diffusion bridge formulation that aligns training and inference dynamics, enforcing a coherent denoising path from noisy degraded states to clean images. Building on these insights, we introduce our lightweight learned prompts on the pre-trained WAN video model and FLUX image models, converting them into high-performing restoration models. Extensive experiments demonstrate that our approach achieves competitive performance and generalization across diverse degradations, while avoiding fine-tuning and restoration-specific control modules. --- ### Stratifying Reinforcement Learning with Signal Temporal Logic Authors: Justin Curry, Alberto Speranzon Published: 2026-04-06 Categories: cs.LG, cs.LO, eess.SY, math.AT arXiv: https://arxiv.org/abs/2604.04923v1 In this paper, we develop a stratification-based semantics for Signal Temporal Logic (STL) in which each atomic predicate is interpreted as a membership test in a stratified space. This perspective reveals a novel correspondence principle between stratification theory and STL, showing that most STL formulas can be viewed as inducing a stratification of space-time. The significance of this interpretation is twofold. First, it offers a fresh theoretical framework for analyzing the structure of the embedding space generated by deep reinforcement learning (DRL) and relates it to the geometry of the ambient decision space. Second, it provides a principled framework that both enables the reuse of existing high-dimensional analysis tools and motivates the creation of novel computational techniques. To ground the theory, we (1) illustrate the role of stratification theory in Minigrid games and (2) apply numerical techniques to the latent embeddings of a DRL agent playing such a game where the robustness of STL formulas is used as the reward. In the process, we propose computationally efficient signatures that, based on preliminary evidence, appear promising for uncovering the stratification structure of such embedding spaces. --- ### TriAttention: Efficient Long Reasoning with Trigonometric KV Compression Authors: Weian Mao, Xi Lin, Wei Huang, Yuxin Xie, Tianfu Fu, Bohan Zhuang, Song Han, Yukang Chen Published: 2026-04-06 Categories: cs.CL, cs.CV arXiv: https://arxiv.org/abs/2604.04921v1 Extended reasoning in large language models (LLMs) creates severe KV cache memory bottlenecks. Leading KV cache compression methods estimate KV importance using attention scores from recent post-RoPE queries. However, queries rotate with position during RoPE, making representative queries very few, leading to poor top-key selection and unstable reasoning. To avoid this issue, we turn to the pre-RoPE space, where we observe that Q and K vectors are highly concentrated around fixed non-zero centers and remain stable across positions -- Q/K concentration. We show that this concentration causes queries to preferentially attend to keys at specific distances (e.g., nearest keys), with the centers determining which distances are preferred via a trigonometric series. Based on this, we propose TriAttention to estimate key importance by leveraging these centers. Via the trigonometric series, we use the distance preference characterized by these centers to score keys according to their positions, and also leverage Q/K norms as an additional signal for importance estimation. On AIME25 with 32K-token generation, TriAttention matches Full Attention reasoning accuracy while achieving 2.5x higher throughput or 10.7x KV memory reduction, whereas leading baselines achieve only about half the accuracy at the same efficiency. TriAttention enables OpenClaw deployment on a single consumer GPU, where long context would otherwise cause out-of-memory with Full Attention. --- ### PINNs in PDE Constrained Optimal Control Problems: Direct vs Indirect Methods Authors: Zhen Zhang, Shanqing Liu, Alessandro Alla, Jerome Darbon, George Em Karniadakis Published: 2026-04-06 Categories: math.OC, cs.LG arXiv: https://arxiv.org/abs/2604.04920v1 We study physics-informed neural networks (PINNs) as numerical tools for the optimal control of semilinear partial differential equations. We first recall the classical direct and indirect viewpoints for optimal control of PDEs, and then present two PINN formulations: a direct formulation based on minimizing the objective under the state constraint, and an indirect formulation based on the first-order optimality system. For a class of semilinear parabolic equations, we derive the state equation, the adjoint equation, and the stationarity condition in a form consistent with continuous-time Pontryagin-type optimality conditions. We then specialize the framework to an Allen-Cahn control problem and compare three numerical approaches: (i) a discretize-then-optimize adjoint method, (ii) a direct PINN, and (iii) an indirect PINN. Numerical results show that the PINN parameterization has an implicit regularizing effect, in the sense that it tends to produce smoother control profiles. They also indicate that the indirect PINN more faithfully preserves the PDE contraint and optimality structure and yields a more accurate neural approximation than the direct PINN. --- ### Vero: An Open RL Recipe for General Visual Reasoning Authors: Gabriel Sarch, Linrong Cai, Qunzhong Wang, Haoyang Wu, Danqi Chen, Zhuang Liu Published: 2026-04-06 Categories: cs.CV, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.04917v3 What does it take to build a visual reasoner that works across charts, science, spatial understanding, and open-ended tasks? The strongest vision-language models (VLMs) suggest that broad visual reasoning is within reach, yet their closed data and reinforcement learning (RL) pipelines make their gains difficult to study, reproduce, or extend. We introduce Vero, a family of fully open VLMs that match or exceed existing open-weight models across diverse visual reasoning tasks. We scale RL data and rewards across six broad task categories, constructing Vero-600K, a 600K-sample dataset from 59 datasets, and designing task-routed rewards that handle heterogeneous answers. Across VeroEval, our 30-benchmark suite, Vero-600K outperforms existing RL datasets under controlled comparisons. Applied to five starting models, Vero variants gain 2.9-5.4 points on average over their initial models. Notably, Vero-Qwen3I-8B, trained on the Instruct model, surpasses Qwen3-VL-8B-Thinking by 3.8 points on average without additional distillation. Systematic ablations reveal that different task categories elicit distinct reasoning patterns and that broad gains depend on learning them jointly rather than in isolation. All data, code, and models are publicly available. --- ### Empowering Power Outage Prediction with Spatially Aware Hybrid Graph Neural Networks and Contrastive Learning Authors: Xuyang Shen, Zijie Pan, Diego Cerrai, Xinxuan Zhang, Christopher Colorio, Emmanouil N. Anagnostou, Dongjin Song Published: 2026-04-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.04916v1 Extreme weather events, such as severe storms, hurricanes, snowstorms, and ice storms, which are exacerbated by climate change, frequently cause widespread power outages. These outages halt industrial operations, impact communities, damage critical infrastructure, profoundly disrupt economies, and have far-reaching effects across various sectors. To mitigate these effects, the University of Connecticut and Eversource Energy Center have developed an outage prediction modeling (OPM) system to provide pre-emptive forecasts for electric distribution networks before such weather events occur. However, existing predictive models in the system do not incorporate the spatial effect of extreme weather events. To this end, we develop Spatially Aware Hybrid Graph Neural Networks (SA-HGNN) with contrastive learning to enhance the OPM predictions for extreme weather-induced power outages. Specifically, we first encode spatial relationships of both static features (e.g., land cover, infrastructure) and event-specific dynamic features (e.g., wind speed, precipitation) via Spatially Aware Hybrid Graph Neural Networks (SA-HGNN). Next, we leverage contrastive learning to handle the imbalance problem associated with different types of extreme weather events and generate location-specific embeddings by minimizing intra-event distances between similar locations while maximizing inter-event distances across all locations. Thorough empirical studies in four utility service territories, i.e., Connecticut, Western Massachusetts, Eastern Massachusetts, and New Hampshire, demonstrate that SA-HGNN can achieve state-of-the-art performance for power outage prediction. --- ### Analyzing Symbolic Properties for DRL Agents in Systems and Networking Authors: Mohammad Zangooei, Jannis Weil, Amr Rizk, Mina Tahmasbi Arashloo, Raouf Boutaba Published: 2026-04-06 Categories: cs.NI, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.04914v1 Deep reinforcement learning (DRL) has shown remarkable performance on complex control problems in systems and networking, including adaptive video streaming, wireless resource management, and congestion control. For safe deployment, however, it is critical to reason about how agents behave across the range of system states they encounter in practice. Existing verification-based methods in this domain primarily focus on point properties, defined around fixed input states, which offer limited coverage and require substantial manual effort to identify relevant input-output pairs for analysis. In this paper, we study symbolic properties, that specify expected behavior over ranges of input states, for DRL agents in systems and networking. We present a generic formulation for symbolic properties, with monotonicity and robustness as concrete examples, and show how they can be analyzed using existing DNN verification engines. Our approach encodes symbolic properties as comparisons between related executions of the same policy and decomposes them into practically tractable sub-properties. These techniques serve as practical enablers for applying existing verification tools to symbolic analysis. Using our framework, diffRL, we conduct an extensive empirical study across three DRL-based control systems, adaptive video streaming, wireless resource management, and congestion control. Through these case studies, we analyze symbolic properties over broad input ranges, examine how property satisfaction evolves during training, study the impact of model size on verifiability, and compare multiple verification backends. Our results show that symbolic properties provide substantially broader coverage than point properties and can uncover non-obvious, operationally meaningful counterexamples, while also revealing practical solver trade-offs and limitations. --- ### A Frame is Worth One Token: Efficient Generative World Modeling with Delta Tokens Authors: Tommie Kerssies, Gabriele Berton, Ju He, Qihang Yu, Wufei Ma, Daan de Geus, Gijs Dubbelman, Liang-Chieh Chen Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04913v1 Anticipating diverse future states is a central challenge in video world modeling. Discriminative world models produce a deterministic prediction that implicitly averages over possible futures, while existing generative world models remain computationally expensive. Recent work demonstrates that predicting the future in the feature space of a vision foundation model (VFM), rather than a latent space optimized for pixel reconstruction, requires significantly fewer world model parameters. However, most such approaches remain discriminative. In this work, we introduce DeltaTok, a tokenizer that encodes the VFM feature difference between consecutive frames into a single continuous "delta" token, and DeltaWorld, a generative world model operating on these tokens to efficiently generate diverse plausible futures. Delta tokens reduce video from a three-dimensional spatio-temporal representation to a one-dimensional temporal sequence, for example yielding a 1,024x token reduction with 512x512 frames. This compact representation enables tractable multi-hypothesis training, where many futures are generated in parallel and only the best is supervised. At inference, this leads to diverse predictions in a single forward pass. Experiments on dense forecasting tasks demonstrate that DeltaWorld forecasts futures that more closely align with real-world outcomes, while having over 35x fewer parameters and using 2,000x fewer FLOPs than existing generative world models. Code and weights: https://deltatok.github.io. --- ### SpatialEdit: Benchmarking Fine-Grained Image Spatial Editing Authors: Yicheng Xiao, Wenhu Zhang, Lin Song, Yukang Chen, Wenbo Li, Nan Jiang, Tianhe Ren, Haokun Lin, Wei Huang, Haoyang Huang, Xiu Li, Nan Duan et al. Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04911v2 Image spatial editing performs geometry-driven transformations, allowing precise control over object layout and camera viewpoints. Current models are insufficient for fine-grained spatial manipulations, motivating a dedicated assessment suite. Our contributions are listed: (i) We introduce SpatialEdit-Bench, a complete benchmark that evaluates spatial editing by jointly measuring perceptual plausibility and geometric fidelity via viewpoint reconstruction and framing analysis. (ii) To address the data bottleneck for scalable training, we construct SpatialEdit-500k, a synthetic dataset generated with a controllable Blender pipeline that renders objects across diverse backgrounds and systematic camera trajectories, providing precise ground-truth transformations for both object- and camera-centric operations. (iii) Building on this data, we develop SpatialEdit-16B, a baseline model for fine-grained spatial editing. Our method achieves competitive performance on general editing while substantially outperforming prior methods on spatial manipulation tasks. All resources will be made public at https://github.com/EasonXiao-888/SpatialEdit. --- ### HI-MoE: Hierarchical Instance-Conditioned Mixture-of-Experts for Object Detection Authors: Vadim Vashkelis, Natalia Trukhina Published: 2026-04-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.04908v1 Mixture-of-Experts (MoE) architectures enable conditional computation by activating only a subset of model parameters for each input. Although sparse routing has been highly effective in language models and has also shown promise in vision, most vision MoE methods operate at the image or patch level. This granularity is poorly aligned with object detection, where the fundamental unit of reasoning is an object query corresponding to a candidate instance. We propose Hierarchical Instance-Conditioned Mixture-of-Experts (HI-MoE), a DETR-style detection architecture that performs routing in two stages: a lightweight scene router first selects a scene-consistent expert subset, and an instance router then assigns each object query to a small number of experts within that subset. This design aims to preserve sparse computation while better matching the heterogeneous, instance-centric structure of detection. In the current draft, experiments are concentrated on COCO with preliminary specialization analysis on LVIS. Under these settings, HI-MoE improves over a dense DINO baseline and over simpler token-level or instance-only routing variants, with especially strong gains on small objects. We also provide an initial visualization of expert specialization patterns. We present the method, ablations, and current limitations in a form intended to support further experimental validation. --- ### How AI Aggregation Affects Knowledge Authors: Daron Acemoglu, Tianyi Lin, Asuman Ozdaglar, James Siderius Published: 2026-04-06 Categories: econ.TH, cs.AI, cs.CY, cs.SI arXiv: https://arxiv.org/abs/2604.04906v1 Artificial intelligence (AI) changes social learning when aggregated outputs become training data for future predictions. To study this, we extend the DeGroot model by introducing an AI aggregator that trains on population beliefs and feeds synthesized signals back to agents. We define the learning gap as the deviation of long-run beliefs from the efficient benchmark, allowing us to capture how AI aggregation affects learning. Our main result identifies a threshold in the speed of updating: when the aggregator updates too quickly, there is no positive-measure set of training weights that robustly improves learning across a broad class of environments, whereas such weights exist when updating is sufficiently slow. We then compare global and local architectures. Local aggregators trained on proximate or topic-specific data robustly improve learning in all environments. Consequently, replacing specialized local aggregators with a single global aggregator worsens learning in at least one dimension of the state. --- ### ClickAIXR: On-Device Multimodal Vision-Language Interaction with Real-World Objects in Extended Reality Authors: Dawar Khan, Alexandre Kouyoumdjian, Xinyu Liu, Omar Mena, Dominik Engel, Ivan Viola Published: 2026-04-06 Categories: cs.CV, cs.GR, cs.HC arXiv: https://arxiv.org/abs/2604.04905v1 We present ClickAIXR, a novel on-device framework for multimodal vision-language interaction with objects in extended reality (XR). Unlike prior systems that rely on cloud-based AI (e.g., ChatGPT) or gaze-based selection (e.g., GazePointAR), ClickAIXR integrates an on-device vision-language model (VLM) with a controller-based object selection paradigm, enabling users to precisely click on real-world objects in XR. Once selected, the object image is processed locally by the VLM to answer natural language questions through both text and speech. This object-centered interaction reduces ambiguity inherent in gaze- or voice-only interfaces and improves transparency by performing all inference on-device, addressing concerns around privacy and latency. We implemented ClickAIXR in the Magic Leap SDK (C API) with ONNX-based local VLM inference. We conducted a user study comparing ClickAIXR with Gemini 2.5 Flash and ChatGPT 5, evaluating usability, trust, and user satisfaction. Results show that latency is moderate and user experience is acceptable. Our findings demonstrate the potential of click-based object selection combined with on-device AI to advance trustworthy, privacy-preserving XR interactions. The source code and supplementary materials are available at: nanovis.org/ClickAIXR.html --- ### Are Latent Reasoning Models Easily Interpretable? Authors: Connor Dilgren, Sarah Wiegreffe Published: 2026-04-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.04902v2 Latent reasoning models (LRMs) have attracted significant research interest due to their low inference cost (relative to explicit reasoning models) and theoretical ability to explore multiple reasoning paths in parallel. However, these benefits come at the cost of reduced interpretability: LRMs are difficult to monitor because they do not reason in natural language. This paper presents an investigation into LRM interpretability by examining two state-of-the-art LRMs. First, we find that latent reasoning tokens are often unnecessary for LRMs' predictions; on logical reasoning datasets, LRMs can almost always produce the same final answers without using latent reasoning at all. This underutilization of reasoning tokens may partially explain why LRMs do not consistently outperform explicit reasoning methods and raises doubts about the stated role of these tokens in prior work. Second, we demonstrate that when latent reasoning tokens are necessary for performance, we can decode gold reasoning traces up to 65-93% of the time for correctly predicted instances. This suggests LRMs often implement the expected solution rather than an uninterpretable reasoning process. Finally, we present a method to decode a verified natural language reasoning trace from latent tokens without knowing a gold reasoning trace a priori, demonstrating that it is possible to find a verified trace for a majority of correct predictions but only a minority of incorrect predictions. Our findings highlight that current LRMs largely encode interpretable processes, and interpretability itself can be a signal of prediction correctness. --- ### Are Latent Reasoning Models Easily Interpretable? Authors: Connor Dilgren, Sarah Wiegreffe Published: 2026-04-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.04902v1 Latent reasoning models (LRMs) have attracted significant research interest due to their low inference cost (relative to explicit reasoning models) and theoretical ability to explore multiple reasoning paths in parallel. However, these benefits come at the cost of reduced interpretability: LRMs are difficult to monitor because they do not reason in natural language. This paper presents an investigation into LRM interpretability by examining two state-of-the-art LRMs. First, we find that latent reasoning tokens are often unnecessary for LRMs' predictions; on logical reasoning datasets, LRMs can almost always produce the same final answers without using latent reasoning at all. This underutilization of reasoning tokens may partially explain why LRMs do not consistently outperform explicit reasoning methods and raises doubts about the stated role of these tokens in prior work. Second, we demonstrate that when latent reasoning tokens are necessary for performance, we can decode gold reasoning traces up to 65-93% of the time for correctly predicted instances. This suggests LRMs often implement the expected solution rather than an uninterpretable reasoning process. Finally, we present a method to decode a verified natural language reasoning trace from latent tokens without knowing a gold reasoning trace a priori, demonstrating that it is possible to find a verified trace for a majority of correct predictions but only a minority of incorrect predictions. Our findings highlight that current LRMs largely encode interpretable processes, and interpretability itself can be a signal of prediction correctness. --- ### FileGram: Grounding Agent Personalization in File-System Behavioral Traces Authors: Shuai Liu, Shulin Tian, Kairui Hu, Yuhao Dong, Zhe Yang, Bo Li, Jingkang Yang, Chen Change Loy, Ziwei Liu Published: 2026-04-06 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.04901v1 Coworking AI agents operating within local file systems are rapidly emerging as a paradigm in human-AI interaction; however, effective personalization remains limited by severe data constraints, as strict privacy barriers and the difficulty of jointly collecting multimodal real-world traces prevent scalable training and evaluation, and existing methods remain interaction-centric while overlooking dense behavioral traces in file-system operations; to address this gap, we propose FileGram, a comprehensive framework that grounds agent memory and personalization in file-system behavioral traces, comprising three core components: (1) FileGramEngine, a scalable persona-driven data engine that simulates realistic workflows and generates fine-grained multimodal action sequences at scale; (2) FileGramBench, a diagnostic benchmark grounded in file-system behavioral traces for evaluating memory systems on profile reconstruction, trace disentanglement, persona drift detection, and multimodal grounding; and (3) FileGramOS, a bottom-up memory architecture that builds user profiles directly from atomic actions and content deltas rather than dialogue summaries, encoding these traces into procedural, semantic, and episodic channels with query-time abstraction; extensive experiments show that FileGramBench remains challenging for state-of-the-art memory systems and that FileGramEngine and FileGramOS are effective, and by open-sourcing the framework, we hope to support future research on personalized memory-centric file-system agents. --- ### QED-Nano: Teaching a Tiny Model to Prove Hard Theorems Authors: LM-Provers, Yuxiao Qu, Amrith Setlur, Jasper Dekoninck, Edward Beeching, Jia Li, Ian Wu, Lewis Tunstall, Aviral Kumar Published: 2026-04-06 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.04898v1 Proprietary AI systems have recently demonstrated impressive capabilities on complex proof-based problems, with gold-level performance reported at the 2025 International Mathematical Olympiad (IMO). However, the training pipelines behind these systems remain largely undisclosed, and their reliance on large "internal" models and scaffolds makes them expensive to run, difficult to reproduce, and hard to study or improve upon. This raises a central question: can small, open models also be trained to achieve competitive reasoning performance on difficult Olympiad-level math? In this paper, we answer this question by building QED-Nano, a 4B model post-trained for Olympiad-level proofs. Our training recipe has three stages: (1) supervised fine-tuning to imbue good proof-writing styles by distilling from DeepSeek-Math-V2, (2) reinforcement learning (RL) with rubric-based rewards, and (3) expanding RL with a reasoning cache, which decomposes long proofs into iterative summarize-and-refine cycles and enables stronger test-time reasoning. QED-Nano surpasses the proof-generation performance of much larger open models, including Nomos-1 and GPT-OSS-120B, and approaches the performance of proprietary models like Gemini 3 Pro, at a fraction of the inference cost. To support further research on open mathematical reasoning, we release the full QED-Nano pipeline, including the QED-Nano and QED-Nano-SFT models, the FineProofs-SFT and FineProofs-RL datasets, and the training and evaluation code. --- ### Agentic Federated Learning: The Future of Distributed Training Orchestration Authors: Rafael O. Jarczewski, Gabriel U. Talasso, Leandro Villas, Allan M. de Souza Published: 2026-04-06 Categories: cs.MA, cs.AI arXiv: https://arxiv.org/abs/2604.04895v1 Although Federated Learning (FL) promises privacy and distributed collaboration, its effectiveness in real-world scenarios is often hampered by the stochastic heterogeneity of clients and unpredictable system dynamics. Existing static optimization approaches fail to adapt to these fluctuations, resulting in resource underutilization and systemic bias. In this work, we propose a paradigm shift towards Agentic-FL, a framework where Language Model-based Agents (LMagents) assume autonomous orchestration roles. Unlike rigid protocols, we demonstrate how server-side agents can mitigate selection bias through contextual reasoning, while client-side agents act as local guardians, dynamically managing privacy budgets and adapting model complexity to hardware constraints. More than just resolving technical inefficiencies, this integration signals the evolution of FL towards decentralized ecosystems, where collaboration is negotiated autonomously, paving the way for future markets of incentive-based models and algorithmic justice. We discuss the reliability (hallucinations) and security challenges of this approach, outlining a roadmap for resilient multi-agent systems in federated environments. --- ### Asymmetric Advantage Modulation Calibrates Entropy Dynamics in RLVR Authors: Hengrui Gu, Xiaotian Han, Yujing Bian, Feiyi Wang, Kaixiong Zhou Published: 2026-04-06 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.04894v2 Reinforcement learning with verifiable rewards (RLVR) has substantially improved the reasoning ability of large language models (LLMs), but it often suffers from \textit{restricted exploration}, where the policy rapidly concentrates on a narrow set of solutions. A common remedy is entropy regularization, which attempts to preserve exploration by increasing policy entropy. However, for LLM-RL, this intervention is highly sensitive to its coefficient, can introduce semantically weak uncertainty, and often yields limited accuracy gains. This motivates a more precise question: which entropy helps reasoning, and which entropy should be reduced? To study this, we parameterize the advantage estimator in Group Relative Policy Optimization (GRPO) into positive and negative outcome-conditioned channels and analyze their entropy dynamics. Our results show that positive-channel modulation raises \textit{productive entropy} associated with successful reasoning trajectories, while negative-channel modulation removes \textit{noisy entropy} associated with failed rollouts and reduces interference with correct paths. Guided by this channel-wise view, we propose \textbf{AsymGRPO}, which decouples the modulation strengths of positive and negative advantages. This enables flexible control over how the model updates across prompt difficulty levels, allowing stronger reinforcement of rare successes on harder prompts or stronger suppression of residual failures on easier prompts without forcing the two channels to share the same modulation strength. Experiments on five mathematical reasoning benchmarks show that AsymGRPO outperforms strong RLVR baselines, with consistent gains across model backbones. --- ### Data Attribution in Adaptive Learning Authors: Amit Kiran Rege Published: 2026-04-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.04892v1 Machine learning models increasingly generate their own training data -- online bandits, reinforcement learning, and post-training pipelines for language models are leading examples. In these adaptive settings, a single training observation both updates the learner and shifts the distribution of future data the learner will collect. Standard attribution methods, designed for static datasets, ignore this feedback. We formalize occurrence-level attribution for finite-horizon adaptive learning via a conditional interventional target, prove that replay-side information cannot recover it in general, and identify a structural class in which the target is identified from logged data. ---