# Research Papers — 2026-03 · bundle page 177 of 243 > Full-text markdown of papers #8801–8850 of 12123 published in 2026-03, newest first. AI-readable mirrors of the full 12123-paper 2026-03 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-03/page/177 [← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/176.md) [Next bundle →](https://aipod.dev/papers/archive/2026-03/page/178.md) ### Functional Bias and Tangent-Space Geometry in Variational Inference Authors: Sean Plummer Published: 2026-03-09 Categories: math.ST, stat.ML arXiv: https://arxiv.org/abs/2603.08925v1 Variational inference approximates Bayesian posterior distributions by projecting onto a tractable family of distributions. While most theoretical analyses evaluate the quality of this approximation using global divergence measures, many applications rely on specific posterior summaries such as expectations, variances, or tail probabilities. We develop a geometric framework for analyzing the bias of posterior functionals under variational approximations. We show that the leading-order bias of a posterior functional is determined by its component orthogonal to the variational tangent space induced by the variational family. Functionals aligned with this space incur only second-order bias. For structured mean-field variational families we characterize the tangent space explicitly and show that it consists of block-additive functions of the parameter blocks, while interaction components determine the leading-order bias. Under standard local asymptotic normality conditions we further derive explicit asymptotic expansions for the bias of posterior functionals and show that omitted interaction directions produce first-order distortion of cross-block dependence measures. These results provide a geometric explanation for several well-known properties of mean-field variational inference, including the systematic distortion of cross-block dependencies. --- ### Quantifying Uncertainty in AI Visibility: A Statistical Framework for Generative Search Measurement Authors: Ronald Sielinski Published: 2026-03-09 Categories: stat.AP, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2603.08924v3 AI-powered answer engines are inherently non-deterministic: identical queries submitted at different times can produce different responses and cite different sources. Despite this stochastic behavior, current approaches to measuring domain visibility in generative search typically rely on single-run point estimates of citation share and prevalence, implicitly treating them as fixed values. This paper argues that citation visibility metrics should be treated as sample estimators of an underlying response distribution rather than fixed values. We conduct an empirical study of citation variability across three generative search platforms--Perplexity Search, OpenAI SearchGPT, and Google Gemini--using repeated sampling across three consumer product topics. Two sampling regimes are employed: daily collections over nine days and high-frequency sampling at ten-minute intervals. We show that citation distributions follow a power-law form and exhibit substantial variability across repeated samples. Bootstrap confidence intervals reveal that many apparent differences between domains fall within the noise floor of the measurement process. Distribution-wide rank stability analysis further demonstrates that citation rankings are unstable across samples, not only among top-ranked domains but throughout the frequently cited domain set. These findings demonstrate that single-run visibility metrics provide a misleadingly precise picture of domain performance in generative search. We argue that citation visibility must be reported with uncertainty estimates and provide practical guidance for sample sizes required to achieve interpretable confidence intervals. --- ### Quantifying Uncertainty in AI Visibility: A Statistical Framework for Generative Search Measurement Authors: Ronald Sielinski Published: 2026-03-09 Categories: stat.AP, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2603.08924v2 AI-powered answer engines are inherently non-deterministic: identical queries submitted at different times can produce different responses and cite different sources. Despite this stochastic behavior, current approaches to measuring domain visibility in generative search typically rely on single-run point estimates of citation share and prevalence, implicitly treating them as fixed values. This paper argues that citation visibility metrics should be treated as sample estimators of an underlying response distribution rather than fixed values. We conduct an empirical study of citation variability across three generative search platforms--Perplexity Search, OpenAI SearchGPT, and Google Gemini--using repeated sampling across three consumer product topics. Two sampling regimes are employed: daily collections over nine days and high-frequency sampling at ten-minute intervals. We show that citation distributions follow a power-law form and exhibit substantial variability across repeated samples. Bootstrap confidence intervals reveal that many apparent differences between domains fall within the noise floor of the measurement process. Distribution-wide rank stability analysis further demonstrates that citation rankings are unstable across samples, not only among top-ranked domains but throughout the frequently cited domain set. These findings demonstrate that single-run visibility metrics provide a misleadingly precise picture of domain performance in generative search. We argue that citation visibility must be reported with uncertainty estimates and provide practical guidance for sample sizes required to achieve interpretable confidence intervals. --- ### Vision-Language Models Encode Clinical Guidelines for Concept-Based Medical Reasoning Authors: Mohamed Harmanani, Bining Long, Zhuoxin Guo, Paul F. R. Wilson, Amirhossein Sabour, Minh Nguyen Nhat To, Gabor Fichtinger, Purang Abolmaesumi, Parvin Mousavi Published: 2026-03-09 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2603.08921v1 Concept Bottleneck Models (CBMs) are a prominent framework for interpretable AI that map learned visual features to a set of meaningful concepts for task-specific downstream predictions. Their sequential structure enhances transparency by connecting model predictions to the underlying concepts that support them. In medical imaging, where transparency is essential, CBMs offer an appealing foundation for explainable model design. However, discrete concept representations often overlook broader clinical context such as diagnostic guidelines and expert heuristics, reducing reliability in complex cases. We propose MedCBR, a concept-based reasoning framework that integrates clinical guidelines with vision-language and reasoning models. Labeled clinical descriptors are transformed into guideline-conformant text, and a concept-based model is trained with a multitask objective combining multimodal contrastive alignment, concept supervision, and diagnostic classification to jointly ground image features, concepts, and pathology. A reasoning model then converts these predictions into structured clinical narratives that explain the diagnosis, emulating expert reasoning based on established guidelines. MedCBR achieves superior diagnostic and concept-level performance, with AUROCs of 94.2% on ultrasound and 84.0% on mammography. Further experiments on non-medical datasets achieve 86.1% accuracy. Our framework enhances interpretability and forms an end-to-end bridge from medical image analysis to decision-making. --- ### Uncovering a Winning Lottery Ticket with Continuously Relaxed Bernoulli Gates Authors: Itamar Tsayag, Ofir Lindenbaum Published: 2026-03-09 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.08914v1 Over-parameterized neural networks incur prohibitive memory and computational costs for resource-constrained deployment. The Strong Lottery Ticket (SLT) hypothesis suggests that randomly initialized networks contain sparse subnetworks achieving competitive accuracy without weight training. Existing SLT methods, notably edge-popup, rely on non-differentiable score-based selection, limiting optimization efficiency and scalability. We propose using continuously relaxed Bernoulli gates to discover SLTs through fully differentiable, end-to-end optimization - training only gating parameters while keeping all network weights frozen at their initialized values. Continuous relaxation enables direct gradient-based optimization of an $\ell_0$-regularization objective, eliminating the need for non-differentiable gradient estimators or iterative pruning cycles. To our knowledge, this is the first fully differentiable approach for SLT discovery that avoids straight-through estimator approximations. Experiments across fully connected networks, CNNs (ResNet, Wide-ResNet), and Vision Transformers (ViT, Swin-T) demonstrate up to 90% sparsity with minimal accuracy loss - nearly double the sparsity achieved by edge-popup at comparable accuracy - establishing a scalable framework for pre-training network sparsification. --- ### Quantifying Memorization and Privacy Risks in Genomic Language Models Authors: Alexander Nemecek, Wenbiao Li, Xiaoqian Jiang, Jaideep Vaidya, Erman Ayday Published: 2026-03-09 Categories: cs.LG, cs.CR, q-bio.GN arXiv: https://arxiv.org/abs/2603.08913v2 Genomic language models (GLMs) have emerged as powerful tools for learning representations of DNA sequences, enabling advances in variant prediction, regulatory element identification, and cross-task transfer learning. However, as these models are increasingly trained or fine-tuned on sensitive genomic cohorts, they risk memorizing specific sequences from their training data, raising serious concerns around privacy, data leakage, and regulatory compliance. Despite growing awareness of memorization risks in general-purpose language models, little systematic evaluation exists for these risks in the genomic domain, where data exhibit unique properties such as a fixed nucleotide alphabet, strong biological structure, and individual identifiability. We present a comprehensive, multi-vector privacy evaluation framework designed to quantify memorization risks in GLMs. Our approach integrates three complementary risk assessment methodologies: perplexity-based detection, canary sequence extraction, and membership inference. These are combined into a unified evaluation pipeline that produces a worst-case memorization risk score. To enable controlled evaluation, we plant canary sequences at varying repetition rates into both synthetic and real genomic datasets, allowing precise quantification of how repetition and training dynamics influence memorization. We evaluate our framework across multiple GLM architectures, examining the relationship between sequence repetition, model capacity, and memorization risk. Our results establish that GLMs exhibit measurable memorization and that the degree of memorization varies across architectures and training regimes. These findings reveal that no single attack vector captures the full scope of memorization risk, underscoring the need for multi-vector privacy auditing as a standard practice for genomic AI systems. --- ### Quantifying Memorization and Privacy Risks in Genomic Language Models Authors: Alexander Nemecek, Wenbiao Li, Xiaoqian Jiang, Jaideep Vaidya, Erman Ayday Published: 2026-03-09 Categories: cs.LG, cs.CR, q-bio.GN arXiv: https://arxiv.org/abs/2603.08913v1 Genomic language models (GLMs) have emerged as powerful tools for learning representations of DNA sequences, enabling advances in variant prediction, regulatory element identification, and cross-task transfer learning. However, as these models are increasingly trained or fine-tuned on sensitive genomic cohorts, they risk memorizing specific sequences from their training data, raising serious concerns around privacy, data leakage, and regulatory compliance. Despite growing awareness of memorization risks in general-purpose language models, little systematic evaluation exists for these risks in the genomic domain, where data exhibit unique properties such as a fixed nucleotide alphabet, strong biological structure, and individual identifiability. We present a comprehensive, multi-vector privacy evaluation framework designed to quantify memorization risks in GLMs. Our approach integrates three complementary risk assessment methodologies: perplexity-based detection, canary sequence extraction, and membership inference. These are combined into a unified evaluation pipeline that produces a worst-case memorization risk score. To enable controlled evaluation, we plant canary sequences at varying repetition rates into both synthetic and real genomic datasets, allowing precise quantification of how repetition and training dynamics influence memorization. We evaluate our framework across multiple GLM architectures, examining the relationship between sequence repetition, model capacity, and memorization risk. Our results establish that GLMs exhibit measurable memorization and that the degree of memorization varies across architectures and training regimes. These findings reveal that no single attack vector captures the full scope of memorization risk, underscoring the need for multi-vector privacy auditing as a standard practice for genomic AI systems. --- ### FedLECC: Cluster- and Loss-Guided Client Selection for Federated Learning under Non-IID Data Authors: Daniel M. Jimenez-Gutierrez, Giovanni Giunta, Mehrdad Hassanzadeh, Aris Anagnostopoulos, Ioannis Chatzigiannakis, Andrea Vitaletti Published: 2026-03-09 Categories: cs.DC, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.08911v1 Federated Learning (FL) enables distributed Artificial Intelligence (AI) across cloud-edge environments by allowing collaborative model training without centralizing data. In cross-device deployments, FL systems face strict communication and participation constraints, as well as strong non-independent and identically distributed (non-IID) data that degrades convergence and model quality. Since only a subset of devices (a.k.a clients) can participate per training round, intelligent client selection becomes a key systems challenge. This paper proposes FedLECC (Federated Learning with Enhanced Cluster Choice), a lightweight, cluster-aware, and loss-guided client selection strategy for cross-device FL. FedLECC groups clients by label-distribution similarity and prioritizes clusters and clients with higher local loss, enabling the selection of a small yet informative and diverse set of clients. Experimental results under severe label skew show that FedLECC improves test accuracy by up to 12%, while reducing communication rounds by approximately 22% and overall communication overhead by up to 50% compared to strong baselines. These results demonstrate that informed client selection improves the efficiency and scalability of FL workloads in cloud-edge systems. --- ### SciTaRC: A Plan-Annotated Scientific Tabular QA Benchmark for Language Reasoning and Complex Computation Authors: Hexuan Wang, Yaxuan Ren, Srikar Bommireddypalli, Shuxian Chen, Adarsh Prabhudesai, Rongkun Zhou, Elina Baral, Philipp Koehn Published: 2026-03-09 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.08910v2 We introduce SciTaRC, an expert-authored benchmark for question answering over scientific tables that targets composite, multi-step reasoning. To enable fine-grained diagnostic analysis beyond end-task accuracy, SciTaRC pairs each question with a manually constructed reasoning plan and explicit complexity metrics. State-of-the-art models fail on at least 23% of these questions, while highly capable open-weight models like Llama-3.3-70B collapse on 65.5% of the benchmark. Error analysis shows that, in zero-shot settings, failures are driven primarily by question comprehension, where models misinterpret the scientific query and derive the wrong reasoning objective. To determine whether overcoming this gap is sufficient, we use the structured plans to decouple strategy formulation from execution. Surprisingly, providing oracle step-by-step plans yields only limited gains and fails to eliminate the performance gap. This reveals a substantial execution bottleneck: both natural language and code-based methods struggle to reliably carry out long-horizon computational chains over structured data. Ultimately, SciTaRC serves as a rigorous diagnostic testbed for studying both planning and execution in scientific table reasoning. --- ### SciTaRC: Benchmarking QA on Scientific Tabular Data that Requires Language Reasoning and Complex Computation Authors: Hexuan Wang, Yaxuan Ren, Srikar Bommireddypalli, Shuxian Chen, Adarsh Prabhudesai, Rongkun Zhou, Elina Baral, Philipp Koehn Published: 2026-03-09 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.08910v1 We introduce SciTaRC, an expert-authored benchmark of questions about tabular data in scientific papers requiring both deep language reasoning and complex computation. We show that current state-of-the-art AI models fail on at least 23% of these questions, a gap that remains significant even for highly capable open-weight models like Llama-3.3-70B-Instruct, which fails on 65.5% of the tasks. Our analysis reveals a universal "execution bottleneck": both code and language models struggle to faithfully execute plans, even when provided with correct strategies. Specifically, code-based methods prove brittle on raw scientific tables, while natural language reasoning primarily fails due to initial comprehension issues and calculation errors. --- ### Cross-Domain Uncertainty Quantification for Selective Prediction: A Comprehensive Bound Ablation with Transfer-Informed Betting Authors: Abhinaba Basu Published: 2026-03-09 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2603.08907v1 We present a comprehensive ablation of nine finite-sample bound families for selective prediction with risk control, combining concentration inequalities (Hoeffding, Empirical Bernstein, Clopper-Pearson, Wasserstein DRO, CVaR) with multiple-testing corrections (union bound, Learn Then Test fixed-sequence) and betting-based confidence sequences (WSR). Our main theoretical contribution is Transfer-Informed Betting (TIB), which warm-starts the WSR wealth process using a source domain's risk profile, achieving tighter bounds in data-scarce settings with a formal dominance guarantee. We prove that the TIB wealth process remains a valid supermartingale under all source-target divergences, that TIB dominates standard WSR when domains match, and that no data-independent warm-start can achieve better convergence. The combination of betting-based confidence sequences, LTT monotone testing, and cross-domain transfer is, to our knowledge, a three-way novelty not present in the literature. We evaluate all nine bound families on four benchmarks-MASSIVE (n=1,102), NyayaBench (n=280), CLINC-150 (n=22.5K), and Banking77 (n=13K)-across 18 (alpha, delta) configurations. On MASSIVE at alpha=0.10, LTT eliminates the ln(K) union-bound penalty, achieving 94.0% guaranteed coverage versus 73.8% for Hoeffding-a 27% relative improvement. On NyayaBench, where the small calibration set makes Hoeffding-family bounds infeasible below alpha=0.20, Transfer-Informed Betting achieves 18.5% coverage at alpha=0.10, a 5.4x improvement over LTT + Hoeffding. We additionally compare with split-conformal prediction, showing that conformal methods produce prediction sets (avg. 1.67 classes) whereas selective prediction provides single-prediction risk guarantees. We apply these methods to agentic caching systems, formalizing a progressive trust model where the guarantee determines when cached responses can be served autonomously. --- ### Multi-Kernel Gated Decoder Adapters for Robust Multi-Task Thyroid Ultrasound under Cross-Center Shift Authors: Maziar Sabouri, Nourhan Bayasi, Arman Rahmim Published: 2026-03-09 Categories: cs.CV, physics.med-ph arXiv: https://arxiv.org/abs/2603.08906v1 Thyroid ultrasound (US) automation couples two competing requirements: global, geometry-driven reasoning for nodule delineation and local, texture-driven reasoning for malignancy risk assessment. Under cross-center domain shift, these cues degrade asymmetrically, yet most multi-task pipelines rely on a single shared backbone, often inducing negative transfer. In this paper, we characterize this interference across CNN (ResNet34) and medical ViT (MedSAM) backbones, and observe a consistent trend: ViTs transfer geometric priors that benefit segmentation, whereas CNNs more reliably preserve texture cues for malignancy discrimination under strong shift and artifacts. Motivated by this failure mode, we propose a lightweight family of decoder-side adapters, the Multi-Kernel Gated Adapter (MKGA) and a residual variant (ResMKGA), which refine multi-scale skip features using complementary receptive fields and apply semantic, context-conditioned gating to suppress artifact-prone content before fusion. Across two US benchmarks, the proposed adapters improve cross-center robustness: they strengthen out-of-domain segmentation and, in the CNN setting, yield clear gains in clinical TI-RADS diagnostic accuracy compared to standard multi-task baselines. Code and models will be released. --- ### NetDiffuser: Deceiving DNN-Based Network Attack Detection Systems with Diffusion-Generated Adversarial Traffic Authors: Pratyay Kumar, Abu Saleh Md Tayeen, Satyajayant Misra, Huiping Cao, Jiefei Liu, Qixu Gong, Jayashree Harikumar Published: 2026-03-09 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2603.08901v1 Deep learning (DL)-based Network Intrusion Detection System (NIDS) has demonstrated great promise in detecting malicious network traffic. However, they face significant security risks due to their vulnerability to adversarial examples (AEs). Most existing adversarial attacks maliciously perturb data to maximize misclassification errors. Among AEs, natural adversarial examples (NAEs) are particularly difficult to detect because they closely resemble real data, making them challenging for both humans and machine learning models to distinguish from legitimate inputs. Creating NAEs is crucial for testing and strengthening NIDS defenses. This paper proposes NetDiffuser1, a novel framework for generating NAEs capable of deceiving NIDS. NetDiffuser consists of two novel components. First, a new feature categorization algorithm is designed to identify relatively independent features in network traffic. Perturbing these features minimizes changes while preserving network flow validity. The second component is a novel application of diffusion models to inject semantically consistent perturbations for generating NAEs. NetDiffuser performance was extensively evaluated using three benchmark NIDS datasets across various model architectures and state-of-the-art adversarial detectors. Our experimental results show that NetDiffuser achieves up to a 29.93% higher attack success rate and reduces AE detection performance by at least 0.267 (in some cases up to 0.534) in the Area under the Receiver Operating Characteristic Curve (AUC-ROC) score compared to the baseline attacks. --- ### A New Modeling to Feature Selection Based on the Fuzzy Rough Set Theory in Normal and Optimistic States on Hybrid Information Systems Authors: Mohammad Hossein Safarpour, Seyed Majid Alavi, Mohammad Izadikhah, Hossein Dibachi Published: 2026-03-09 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.08900v2 Considering the high volume, wide variety, and rapid speed of data generation, investigating feature selection methods for big data presents various applications and advantages. By removing irrelevant and redundant features, feature selection reduces data dimensions, thereby facilitating optimal decision-making within decision systems. One of the key tools for feature selection in hybrid information systems is fuzzy rough set theory. However, this theory faces two significant challenges: First, obtaining fuzzy equivalence relations through intersection operations in high-dimensional spaces can be both time-consuming and memory-intensive. Additionally, this method may produce noisy data, complicating the feature selection process. The purpose and innovation of this paper are to address these issues. We proposed a new feature selection model that calculates the combined distance between objects and subsequently used this information to derive the fuzzy equivalence relation. Rather than directly solving the feature selection problem, this approach reformulates it into an optimization problem that can be tackled using appropriate meta-heuristic algorithms. We have named this new approach FSbuHD. The FSbuHD model operates in two modes - normal and optimistic - based on the selection of one of the two introduced fuzzy equivalence relations. The model is then tested on standard datasets from the UCI repository and compared with other algorithms. The results of this research demonstrate that FSbuHD is one of the most efficient and effective methods for feature selection when compared to previous methods and algorithms. --- ### ConFu: Contemplate the Future for Better Speculative Sampling Authors: Zongyue Qin, Raghavv Goel, Mukul Gagrani, Risheek Garrepalli, Mingu Lee, Yizhou Sun Published: 2026-03-09 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2603.08899v3 Speculative decoding has emerged as a powerful approach to accelerate large language model (LLM) inference by employing lightweight draft models to propose candidate tokens that are subsequently verified by the target model. The effectiveness of this paradigm critically depends on the quality of the draft model. While recent advances such as the EAGLE series achieve state-of-the-art speedup, existing draft models remain limited by error accumulation: they condition only on the current prefix, causing their predictions to drift from the target model over steps. In this work, we propose \textbf{ConFu} (Contemplate the Future), a novel speculative decoding framework that enables draft models to anticipate the future direction of generation. ConFu introduces (i) contemplate tokens and soft prompts that allow the draft model to leverage future-oriented signals from the target model at negligible cost, (ii) a dynamic contemplate token mechanism with MoE to enable context-aware future prediction, and (iii) a training framework with anchor token sampling and future prediction replication that learns robust future prediction. ConFu improves token acceptance rates and generation speed over EAGLE-3 by 8--11\% on Llama-3 3B/8B and by approximately 20\% on Qwen-3 4B across downstream tasks. We believe our work is the first to bridge speculative decoding with continuous reasoning tokens, offering a new direction for accelerating LLM inference. --- ### Towards Visual Query Segmentation in the Wild Authors: Bing Fan, Minghao Li, Hanzhi Zhang, Shaohua Dong, Naga Prudhvi Mareedu, Weishi Shi, Yunhe Feng, Yan Huang, Heng Fan Published: 2026-03-09 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.08898v1 In this paper, we introduce visual query segmentation (VQS), a new paradigm of visual query localization (VQL) that aims to segment all pixel-level occurrences of an object of interest in an untrimmed video, given an external visual query. Compared to existing VQL locating only the last appearance of a target using bounding boxes, VQS enables more comprehensive (i.e., all object occurrences) and precise (i.e., pixel-level masks) localization, making it more practical for real-world scenarios. To foster research on this task, we present VQS-4K, a large-scale benchmark dedicated to VQS. Specifically, VQS-4K contains 4,111 videos with more than 1.3 million frames and covers a diverse set of 222 object categories. Each video is paired with a visual query defined by a frame outside the search video and its target mask, and annotated with spatial-temporal masklets corresponding to the queried target. To ensure high quality, all videos in VQS-4K are manually labeled with meticulous inspection and iterative refinement. To the best of our knowledge, VQS-4K is the first benchmark specifically designed for VQS. Furthermore, to stimulate future research, we present a simple yet effective method, named VQ-SAM, which extends SAM 2 by leveraging target-specific and background distractor cues from the video to progressively evolve the memory through a novel multi-stage framework with an adaptive memory generation (AMG) module for VQS, significantly improving the performance. In our extensive experiments on VQS-4K, VQ-SAM achieves promising results and surpasses all existing approaches, demonstrating its effectiveness. With the proposed VQS-4K and VQ-SAM, we expect to go beyond the current VQL paradigm and inspire more future research and practical applications on VQS. Our benchmark, code, and results will be made publicly available. --- ### Comparative Analysis of Patch Attack on VLM-Based Autonomous Driving Architectures Authors: David Fernandez, Pedram MohajerAnsari, Amir Salarpour, Long Cheng, Abolfazl Razi, Mert D. Pesé Published: 2026-03-09 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.08897v1 Vision-language models are emerging for autonomous driving, yet their robustness to physical adversarial attacks remains unexplored. This paper presents a systematic framework for comparative adversarial evaluation across three VLM architectures: Dolphins, OmniDrive (Omni-L), and LeapVAD. Using black-box optimization with semantic homogenization for fair comparison, we evaluate physically realizable patch attacks in CARLA simulation. Results reveal severe vulnerabilities across all architectures, sustained multi-frame failures, and critical object detection degradation. Our analysis exposes distinct architectural vulnerability patterns, demonstrating that current VLM designs inadequately address adversarial threats in safety-critical autonomous driving applications. --- ### From Word2Vec to Transformers: Text-Derived Composition Embeddings for Filtering Combinatorial Electrocatalysts Authors: Lei Zhang, Markus Stricker Published: 2026-03-09 Categories: cond-mat.mtrl-sci, cs.CL arXiv: https://arxiv.org/abs/2603.08881v1 Compositionally complex solid solution electrocatalysts span vast composition spaces, and even one materials system can contain more candidate compositions than can be measured exhaustively. Here we evaluate a label-free screening strategy that represents each composition using embeddings derived from scientific texts and prioritizes candidates based on similarity to two property concepts. We compare a corpus-trained Word2Vec baseline with transformer-based embeddings, where compositions are encoded either by linear element-wise mixing or by short composition prompts. Similarities to `concept directions', the terms conductivity and dielectric, define a 2-dimensional descriptor space, and a symmetric Pareto-front selection is used to filter candidate subsets without using electrochemical labels. Performance is assessed on 15 materials libraries including noble metal alloys and multicomponent oxides. In this setting, the lightweight Word2Vec baseline, which uses a simple linear combination of element embeddings, often achieves the highest number of reductions of possible candidate compositions while staying close to the best measured performance. --- ### MultiGraSCCo: A Multilingual Anonymization Benchmark with Annotations of Personal Identifiers Authors: Ibrahim Baroud, Christoph Otto, Vera Czehmann, Christine Hovhannisyan, Lisa Raithel, Sebastian Möller, Roland Roller Published: 2026-03-09 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.08879v3 Accessing sensitive patient data for machine learning is challenging due to privacy concerns. Datasets with annotations of personally identifiable information are crucial for developing and testing anonymization systems to enable safe data sharing that complies with privacy regulations. Since accessing real patient data is a bottleneck, synthetic data offers an efficient solution for data scarcity, bypassing privacy regulations that apply to real data. Moreover, neural machine translation can help to create high-quality data for low-resource languages by translating validated real or synthetic data from a high-resource language. In this work, we create a multilingual anonymization benchmark in ten languages, using a machine translation methodology that preserves the original annotations and renders names of cities and people in a culturally and contextually appropriate form in each target language. Our evaluation study with medical professionals confirms the quality of the translations, both in general and with respect to the translation and adaptation of personal information. Our benchmark with over 2,500 annotations of personal information can be used in many applications, including training annotators, validating annotations across institutions without legal complications, and helping improve the performance of automatic personal information detection. We make our benchmark and annotation guidelines available for further research. --- ### Quantifying the Accuracy and Cost Impact of Design Decisions in Budget-Constrained Agentic LLM Search Authors: Kyle McCleary, James Ghawaly Published: 2026-03-09 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.08877v1 Agentic Retrieval-Augmented Generation (RAG) systems combine iterative search, planning prompts, and retrieval backends, but deployed settings impose explicit budgets on tool calls and completion tokens. We present a controlled measurement study of how search depth, retrieval strategy, and completion budget affect accuracy and cost under fixed constraints. Using Budget-Constrained Agentic Search (BCAS), a model-agnostic evaluation harness that surfaces remaining budget and gates tool use, we run comparisons across six LLMs and three question-answering benchmarks. Across models and datasets, accuracy improves with additional searches up to a small cap, hybrid lexical and dense retrieval with lightweight re-ranking produces the largest average gains in our ablation grid, and larger completion budgets are most helpful on HotpotQA-style synthesis. These results provide practical guidance for configuring budgeted agentic retrieval pipelines and are accompanied by reproducible prompts and evaluation settings. --- ### One Language, Two Scripts: Probing Script-Invariance in LLM Concept Representations Authors: Sripad Karne Published: 2026-03-09 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.08869v1 Do the features learned by Sparse Autoencoders (SAEs) represent abstract meaning, or are they tied to how text is written? We investigate this question using Serbian digraphia as a controlled testbed: Serbian is written interchangeably in Latin and Cyrillic scripts with a near-perfect character mapping between them, enabling us to vary orthography while holding meaning exactly constant. Crucially, these scripts are tokenized completely differently, sharing no tokens whatsoever. Analyzing SAE feature activations across the Gemma model family (270M-27B parameters), we find that identical sentences in different Serbian scripts activate highly overlapping features, far exceeding random baselines. Strikingly, changing script causes less representational divergence than paraphrasing within the same script, suggesting SAE features prioritize meaning over orthographic form. Cross-script cross-paraphrase comparisons provide evidence against memorization, as these combinations rarely co-occur in training data yet still exhibit substantial feature overlap. This script invariance strengthens with model scale. Taken together, our findings suggest that SAE features can capture semantics at a level of abstraction above surface tokenization, and we propose Serbian digraphia as a general evaluation paradigm for probing the abstractness of learned representations. --- ### Why Channel-Centric Models are not Enough to Predict End-to-End Performance in Private 5G: A Measurement Campaign and Case Study Authors: Nils Jörgensen Published: 2026-03-09 Categories: cs.NI, cs.LG, cs.RO arXiv: https://arxiv.org/abs/2603.08865v1 Communication-aware robot planning requires accurate predictions of wireless network performance. Current approaches rely on channel-level metrics such as received signal strength and signal-to-noise ratio, assuming these translate reliably into end-to-end throughput. We challenge this assumption through a measurement campaign in a private 5G industrial environment. We evaluate throughput predictions from a commercial ray-tracing simulator as well as data-driven Gaussian process regression models against measurements collected using a mobile robot. The study uses off-the-shelf user equipment in an underground, radio-shielded facility with detailed 3D modeling, representing a best-case scenario for prediction accuracy. The ray-tracing simulator captures the spatial structure of indoor propagation and predicts channel-level metrics with reasonable fidelity. However, it systematically over-predicts throughput, even in line-of-sight regions. The dominant error source is shown to be over-estimation of sustainable MIMO spatial layers: the simulator assumes near-uniform four-layer transmission while measurements reveal substantial adaptation between one and three layers. This mismatch inflates predicted throughput even when channel metrics appear accurate. In contrast, a Gaussian process model with a rational quadratic kernel achieves approximately two-thirds reduction in prediction error with near-zero bias by learning end-to-end throughput directly from measurements. These findings demonstrate that favorable channel conditions do not guarantee high throughput; communication-aware planners relying solely on channel-centric predictions risk overly optimistic trajectories that violate reliability requirements. Accurate throughput prediction for 5G systems requires either extensive calibration of link-layer models or data-driven approaches that capture real system behavior. --- ### APPLV: Adaptive Planner Parameter Learning from Vision-Language-Action Model Authors: Yuanjie Lu, Beichen Wang, Zhengqi Wu, Yang Li, Xiaomin Lin, Chengzhi Mao, Xuesu Xiao Published: 2026-03-09 Categories: cs.RO, cs.LG arXiv: https://arxiv.org/abs/2603.08862v2 Autonomous navigation in highly constrained environments remains challenging for mobile robots. Classical navigation approaches offer safety assurances but require environment-specific parameter tuning; end-to-end learning bypasses parameter tuning but struggles with precise control in constrained spaces. To this end, recent robot learning approaches automate parameter tuning while retaining classical systems' safety, yet still face challenges in generalizing to unseen environments. Recently, Vision-Language-Action (VLA) models have shown promise by leveraging foundation models' scene understanding capabilities, but still struggle with precise control and inference latency in navigation tasks. In this paper, we propose Adaptive Planner Parameter Learning from Vision-Language-Action Model (\textsc{applv}). Unlike traditional VLA models that directly output actions, \textsc{applv} leverages pre-trained vision-language models with a regression head to predict planner parameters that configure classical planners. We develop two training strategies: supervised learning fine-tuning from collected navigation trajectories and reinforcement learning fine-tuning to further optimize navigation performance. We evaluate \textsc{applv} across multiple motion planners on the simulated Benchmark Autonomous Robot Navigation (BARN) dataset and in physical robot experiments. Results demonstrate that \textsc{applv} outperforms existing methods in both navigation performance and generalization to unseen environments. --- ### Expressivity-Efficiency Tradeoffs for Hybrid Sequence Models Authors: John Cooper, Ilias Diakonikolas, Mingchen Ma, Frederic Sala Published: 2026-03-09 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.08859v1 Hybrid sequence models--combining Transformer and state-space model layers--seek to gain the expressive versatility of attention as well as the computational efficiency of state-space model layers. Despite burgeoning interest in hybrid models, we lack a basic understanding of the settings where--and underlying mechanisms through which--they offer benefits over their constituent models. In this paper, we study this question, focusing on a broad family of core synthetic tasks. For this family of tasks, we prove the existence of fundamental limitations for non-hybrid models. Specifically, any Transformer or state-space model that solves the underlying task requires either a large number of parameters or a large working memory. On the other hand, for two prototypical tasks within this family--namely selective copying and associative recall--we construct hybrid models of small size and working memory that provably solve these tasks, thus achieving the best of both worlds. Our experimental evaluation empirically validates our theoretical findings. Importantly, going beyond the settings in our theoretical analysis, we empirically show that learned--rather than constructed--hybrids outperform non-hybrid models with up to 6x as many parameters. We additionally demonstrate that hybrid models exhibit stronger length generalization and out-of-distribution robustness than non-hybrids. --- ### Unpacking Interpretability: Human-Centered Criteria for Optimal Combinatorial Solutions Authors: Dominik Pegler, Frank Jäkel, David Steyrl, Frank Scharnowski, Filip Melinscak Published: 2026-03-09 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2603.08856v1 Algorithmic support systems often return optimal solutions that are hard to understand. Effective human-algorithm collaboration, however, requires interpretability. When machine solutions are equally optimal, humans must select one, but a precise account of what makes one solution more interpretable than another remains missing. To identify structural properties of interpretable machine solutions, we present an experimental paradigm in which participants chose which of two equally optimal solutions for packing items into bins was easier to understand. We show that preferences reliably track three quantifiable properties of solution structure: alignment with a greedy heuristic, simple within-bin composition, and ordered visual representation. The strongest associations were observed for ordered representations and heuristic alignment, with compositional simplicity also showing a consistent association. Reaction-time evidence was mixed, with faster responses observed primarily when heuristic differences were larger, and aggregate webcam-based gaze did not show reliable effects of complexity. These results provide a concrete, feature-based account of interpretability in optimal packing solutions, linking solution structure to human preference. By identifying actionable properties (simple compositions, ordered representation, and heuristic alignment), our findings enable interpretability-aware optimization and presentation of machine solutions, and outline a path to quantify trade-offs between optimality and interpretability in real-world allocation and design tasks. --- ### LDP: An Identity-Aware Protocol for Multi-Agent LLM Systems Authors: Sunil Prakash Published: 2026-03-09 Categories: cs.AI, cs.MA, cs.SE arXiv: https://arxiv.org/abs/2603.08852v1 As multi-agent AI systems grow in complexity, the protocols connecting them constrain their capabilities. Current protocols such as A2A and MCP do not expose model-level properties as first-class primitives, ignoring properties fundamental to effective delegation: model identity, reasoning profile, quality calibration, and cost characteristics. We present the LLM Delegate Protocol (LDP), an AI-native communication protocol introducing five mechanisms: (1) rich delegate identity cards with quality hints and reasoning profiles; (2) progressive payload modes with negotiation and fallback; (3) governed sessions with persistent context; (4) structured provenance tracking confidence and verification status; (5) trust domains enforcing security boundaries at the protocol level. We implement LDP as a plugin for the JamJet agent runtime and evaluate against A2A and random baselines using local Ollama models and LLM-as-judge evaluation. Identity-aware routing achieves ~12x lower latency on easy tasks through delegate specialization, though it does not improve aggregate quality in our small delegate pool; semantic frame payloads reduce token count by 37% (p=0.031) with no observed quality loss; governed sessions eliminate 39% token overhead at 10 rounds; and noisy provenance degrades synthesis quality below the no-provenance baseline, arguing that confidence metadata is harmful without verification. Simulated analyses show architectural advantages in attack detection (96% vs. 6%) and failure recovery (100% vs. 35% completion). This paper contributes a protocol design, reference implementation, and initial evidence that AI-native protocol primitives enable more efficient and governable delegation. --- ### HECTOR: Hybrid Editable Compositional Object References for Video Generation Authors: Guofeng Zhang, Angtian Wang, Jacob Zhiyuan Fang, Liming Jiang, Haotian Yang, Alan Yuille, Chongyang Ma Published: 2026-03-09 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.08850v1 Real-world videos naturally portray complex interactions among distinct physical objects, effectively forming dynamic compositions of visual elements. However, most current video generation models synthesize scenes holistically and therefore lack mechanisms for explicit compositional manipulation. To address this limitation, we propose HECTOR, a generative pipeline that enables fine-grained compositional control. In contrast to prior methods,HECTOR supports hybrid reference conditioning, allowing generation to be simultaneously guided by static images and/or dynamic videos. Moreover, users can explicitly specify the trajectory of each referenced element, precisely controlling its location, scale, and speed (see Figure1). This design allows the model to synthesize coherent videos that satisfy complex spatiotemporal constraints while preserving high-fidelity adherence to references. Extensive experiments demonstrate that HECTOR achieves superior visual quality, stronger reference preservation, and improved motion controllability compared with existing approaches. --- ### A Lightweight Multi-Cancer Tumor Localization Framework for Deployable Digital Pathology Authors: Brian Isett, Rebekah Dadey, Aofei Li, Ryan C. Augustin, Kate Smith, Aatur D. Singhi, Qiangqiang Gu, Riyue Bao Published: 2026-03-09 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.08844v1 Accurate localization of tumor regions from hematoxylin and eosin-stained whole-slide images is fundamental for translational research including spatial analysis, molecular profiling, and tissue architecture investigation. However, deep learning-based tumor detection trained within specific cancers may exhibit reduced robustness when applied across different tumor types. We investigated whether balanced training across cancers at modest scale can achieve high performance and generalize to unseen tumor types. A multi-cancer tumor localization model (MuCTaL) was trained on 79,984 non-overlapping tiles from four cancers (melanoma, hepatocellular carcinoma, colorectal cancer, and non-small cell lung cancer) using transfer learning with DenseNet169. The model achieved a tile-level ROC-AUC of 0.97 in validation data from the four training cancers, and 0.71 on an independent pancreatic ductal adenocarcinoma cohort. A scalable inference workflow was built to generate spatial tumor probability heatmaps compatible with existing digital pathology tools. Code and models are publicly available at https://github.com/AivaraX-AI/MuCTaL. --- ### MASEval: Extending Multi-Agent Evaluation from Models to Systems Authors: Cornelius Emde, Alexander Rubinstein, Anmol Goel, Ahmed Heakl, Sangdoo Yun, Seong Joon Oh, Martin Gubri Published: 2026-03-09 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2603.08835v1 The rapid adoption of LLM-based agentic systems has produced a rich ecosystem of frameworks (smolagents, LangGraph, AutoGen, CAMEL, LlamaIndex, i.a.). Yet existing benchmarks are model-centric: they fix the agentic setup and do not compare other system components. We argue that implementation decisions substantially impact performance, including choices such as topology, orchestration logic, and error handling. MASEval addresses this evaluation gap with a framework-agnostic library that treats the entire system as the unit of analysis. Through a systematic system-level comparison across 3 benchmarks, 3 models, and 3 frameworks, we find that framework choice matters as much as model choice. MASEval allows researchers to explore all components of agentic systems, opening new avenues for principled system design, and practitioners to identify the best implementation for their use case. MASEval is available under the MIT licence https://github.com/parameterlab/MASEval. --- ### Computer Vision-Based Vehicle Allotment System using Perspective Mapping Authors: Prachi Nandi, Sonakshi Satapathy, Suchismita Chinara Published: 2026-03-09 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.08827v1 Smart city research envisions a future in which data-driven solutions and sustainable infrastructure work together to define urban living at the crossroads of urbanization and technology. Within this framework, smart parking systems play an important role in reducing urban congestion and supporting sustainable transportation. Automating parking solutions have considerable benefits, such as increased efficiency and less reliance on human involvement, but obstacles such as sensor limitations and integration complications remain. To overcome them, a more sophisticated car allotment system is required, particularly in heavily populated urban areas. Computer vision, with its higher accuracy and adaptability, outperforms traditional sensor-based systems for recognizing vehicles and vacant parking spaces. Unlike fixed sensor technologies, computer vision can dynamically assess a wide range of visual inputs while adjusting to changing parking layouts. This research presents a cost-effective, easy-to-implement smart parking system utilizing computer vision and object detection models like YOLOv8. Using inverse perspective mapping (IPM) to merge images from four camera views, we extract data on vacant spaces. The system simulates a 3D parking environment, representing available spots with a 3D Cartesian plot to guide users. --- ### Local Message-Passing for Discrete Graph Generation Authors: Jay Revolinsky, Harry Shomer, Jiliang Tang Published: 2026-03-09 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.08825v2 Discrete graph generation has emerged as a powerful paradigm for modeling graph-structured data, yet state of the art models often rely on Graph Transformers or higher order architectures. We revisit this design assumption by introducing GenGNN, a modular message passing backbone for graph generation. GenGNN enables powerful generation by persisting edge fields through latent refinement of coupled node edge graph states, all without requiring global attention. Diffusion models integrating GenGNN achieve over 90 percent validity on standard benchmark datasets, performing within margins of Graph Transformer backbones and achieving up to 2x or even 5x faster inference. Systematic ablations isolate how GenGNN is resilient to oversmoothing during generative denoising, indicating each GenGNN component is necessary for downstream generation quality. Finally, representation-space analysis suggests GenGNN learns functionally similar representations to more theoretically-expressive architectures; even at deeper layers. As such, GenGNN uplifts local message-passing to challenge prevailing assumptions that performant discrete graph generation requires global attention or higher-order representations. Source Code Available Here --- ### SoftJAX & SoftTorch: Empowering Automatic Differentiation Libraries with Informative Gradients Authors: Anselm Paulus, A. René Geist, Vít Musil, Sebastian Hoffmann, Onur Beker, Georg Martius Published: 2026-03-09 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.08824v1 Automatic differentiation (AD) frameworks such as JAX and PyTorch have enabled gradient-based optimization for a wide range of scientific fields. Yet, many "hard" primitives in these libraries such as thresholding, Boolean logic, discrete indexing, and sorting operations yield zero or undefined gradients that are not useful for optimization. While numerous "soft" relaxations have been proposed that provide informative gradients, the respective implementations are fragmented across projects, making them difficult to combine and compare. This work introduces SoftJAX and SoftTorch, open-source, feature-complete libraries for soft differentiable programming. These libraries provide a variety of soft functions as drop-in replacements for their hard JAX and PyTorch counterparts. This includes (i) elementwise operators such as clip or abs, (ii) utility methods for manipulating Booleans and indices via fuzzy logic, (iii) axiswise operators such as sort or rank -- based on optimal transport or permutahedron projections, and (iv) offer full support for straight-through gradient estimation. Overall, SoftJAX and SoftTorch make the toolbox of soft relaxations easily accessible to differentiable programming, as demonstrated through benchmarking and a practical case study. Code is available at github.com/a-paulus/softjax and github.com/a-paulus/softtorch. --- ### Fish Audio S2 Technical Report Authors: Shijia Liao, Yuxuan Wang, Songting Liu, Yifan Cheng, Ruoyi Zhang, Tianyu Li, Shidong Li, Yisheng Zheng, Xingwei Liu, Qingzheng Wang, Zhizhuo Zhou, Jiahua Liu et al. Published: 2026-03-09 Categories: cs.SD, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2603.08823v2 We introduce Fish Audio S2, an open-sourced text-to-speech system featuring multi-speaker, multi-turn generation, and, most importantly, instruction-following control via natural-language descriptions. To scale training, we develop a multi-stage training recipe together with a staged data pipeline covering video captioning and speech captioning, voice-quality assessment, and reward modeling. To push the frontier of open-source TTS, we release our model weights, fine-tuning code, and an SGLang-based inference engine. The inference engine is production-ready for streaming, achieving an RTF of 0.195 and a time-to-first-audio below 100 ms.Our code and weights are available on GitHub (https://github.com/fishaudio/fish-speech) and Hugging Face (https://huggingface.co/fishaudio/s2-pro). We highly encourage readers to visit https://fish.audio to try custom voices. --- ### Beyond Relevance: On the Relationship Between Retrieval and RAG Information Coverage Authors: Saron Samuel, Alexander Martin, Eugene Yang, Andrew Yates, Dawn Lawrie, Ian Soboroff, Laura Dietz, Benjamin Van Durme Published: 2026-03-09 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2603.08819v4 Retrieval-augmented generation (RAG) systems combine document retrieval with a generative model to address complex information seeking tasks like report generation. While the relationship between retrieval quality and generation effectiveness seems intuitive, it has not been systematically studied. We investigate whether upstream retrieval metrics can serve as reliable early indicators of the final generated response's information coverage. Through experiments across two text RAG benchmarks (TREC NeuCLIR 2024 and TREC RAG 2024) and one multimodal benchmark (WikiVideo), we analyze 15 text retrieval stacks and 10 multimodal retrieval stacks across four RAG pipelines and multiple evaluation frameworks (Auto-ARGUE and MiRAGE). Our findings demonstrate strong correlations between coverage-based retrieval metrics and nugget coverage in generated responses at both topic and system levels. This relationship holds most strongly when retrieval objectives align with generation goals, though more complex iterative RAG pipelines can partially decouple generation quality from retrieval effectiveness. These findings provide empirical support for using retrieval metrics as proxies for RAG performance. --- ### Scale-Plan: Scalable Language-Enabled Task Planning for Heterogeneous Multi-Robot Teams Authors: Piyush Gupta, Sangjae Bae, Jiachen Li, David Isele Published: 2026-03-09 Categories: cs.RO, cs.AI, cs.ET, cs.MA arXiv: https://arxiv.org/abs/2603.08814v2 Long-horizon task planning for heterogeneous multi-robot systems is essential for deploying collaborative teams in real-world environments; yet, it remains challenging due to the large volume of perceptual information, much of which is irrelevant to task objectives and burdens planning. Traditional symbolic planners rely on manually constructed problem specifications, limiting scalability and adaptability, while recent large language model (LLM)-based approaches often suffer from hallucinations and weak grounding-i.e., poor alignment between generated plans and actual environmental objects and constraints-in object-rich settings. We present Scale-Plan, a scalable LLM-assisted framework that generates compact, task-relevant problem representations from natural language instructions. Given a PDDL domain specification, Scale-Plan constructs an action graph capturing domain structure and uses shallow LLM reasoning to guide a structured graph search that identifies a minimal subset of relevant actions and objects. By filtering irrelevant information prior to planning, Scale-Plan enables efficient decomposition, allocation, and long-horizon plan generation. We evaluate our approach on complex multi-agent tasks and introduce MAT2-THOR, a cleaned benchmark built on AI2-THOR for reliable evaluation of multi-robot planning systems. Scale-Plan outperforms pure LLM and hybrid LLM-PDDL baselines across all metrics, improving scalability and reliability. Project website: https://github.com/honda-research-institute/Scale_Plan --- ### Scale-Plan: Scalable Language-Enabled Task Planning for Heterogeneous Multi-Robot Teams Authors: Piyush Gupta, Sangjae Bae, Jiachen Li, David Isele Published: 2026-03-09 Categories: cs.RO, cs.AI, cs.ET, cs.MA arXiv: https://arxiv.org/abs/2603.08814v1 Long-horizon task planning for heterogeneous multi-robot systems is essential for deploying collaborative teams in real-world environments; yet, it remains challenging due to the large volume of perceptual information, much of which is irrelevant to task objectives and burdens planning. Traditional symbolic planners rely on manually constructed problem specifications, limiting scalability and adaptability, while recent large language model (LLM)-based approaches often suffer from hallucinations and weak grounding-i.e., poor alignment between generated plans and actual environmental objects and constraints-in object-rich settings. We present Scale-Plan, a scalable LLM-assisted framework that generates compact, task-relevant problem representations from natural language instructions. Given a PDDL domain specification, Scale-Plan constructs an action graph capturing domain structure and uses shallow LLM reasoning to guide a structured graph search that identifies a minimal subset of relevant actions and objects. By filtering irrelevant information prior to planning, Scale-Plan enables efficient decomposition, allocation, and long-horizon plan generation. We evaluate our approach on complex multi-agent tasks and introduce MAT2-THOR, a cleaned benchmark built on AI2-THOR for reliable evaluation of multi-robot planning systems. Scale-Plan outperforms pure LLM and hybrid LLM-PDDL baselines across all metrics, improving scalability and reliability. --- ### VisionCreator-R1: A Reflection-Enhanced Native Visual-Generation Agentic Model Authors: Jinxiang Lai, Wenzhe Zhao, Zexin Lu, Hualei Zhang, Qinyu Yang, Rongwei Quan, Zhimin Li, Shuai Shao, Song Guo, Qinglin Lu Published: 2026-03-09 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.08812v1 Visual content generation has advanced from single-image to multi-image workflows, yet existing agents remain largely plan-driven and lack systematic reflection mechanisms to correct mid-trajectory visual errors. To address this limitation, we propose VisionCreator-R1, a native visual generation agent with explicit reflection, together with a Reflection-Plan Co-Optimization (RPCO) training methodology. Through extensive experiments and trajectory-level analysis, we uncover reflection-plan optimization asymmetry in reinforcement learning (RL): planning can be reliably optimized via plan rewards, while reflection learning is hindered by noisy credit assignment. Guided by this insight, our RPCO first trains on the self-constructed VCR-SFT dataset with reflection-strong single-image trajectories and planning-strong multi-image trajectories, then co-optimization on VCR-RL dataset via RL. This yields our unified VisionCreator-R1 agent, which consistently outperforms Gemini2.5Pro on existing benchmarks and our VCR-bench covering single-image and multi-image tasks. --- ### Where, What, Why: Toward Explainable 3D-GS Watermarking Authors: Mingshu Cai, Jiajun Li, Osamu Yoshie, Yuya Ieiri, Yixuan Li Published: 2026-03-09 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.08809v2 As 3D Gaussian Splatting becomes the de facto representation for interactive 3D assets, robust yet imperceptible watermarking is critical. We present a representation-native framework that separates where to write from how to preserve quality. A Trio-Experts module operates directly on Gaussian primitives to derive priors for carrier selection, while a Safety and Budget Aware Gate (SBAG) allocates Gaussians to watermark carriers, optimized for bit resilience under perturbation and bitrate budgets, and to visual compensators that are insulated from watermark loss. To maintain fidelity, we introduce a channel-wise group mask that controls gradient propagation for carriers and compensators, thereby limiting Gaussian parameter updates, repairing local artifacts, and preserving high-frequency details without increasing runtime. Our design yields view-consistent watermark persistence and strong robustness against common image distortions such as compression and noise, while achieving a favorable robustness-quality trade-off compared with prior methods. In addition, decoupled finetuning provides per-Gaussian attributions that reveal where the message is carried and why those carriers are selected, enabling auditable explainability. Compared with state-of-the-art methods, our approach achieves a PSNR improvement of +0.83 dB and a bit-accuracy gain of +1.24%. --- ### Test-Driven AI Agent Definition (TDAD): Compiling Tool-Using Agents from Behavioral Specifications Authors: Tzafrir Rehan Published: 2026-03-09 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2603.08806v1 We present Test-Driven AI Agent Definition (TDAD), a methodology that treats agent prompts as compiled artifacts: engineers provide behavioral specifications, a coding agent converts them into executable tests, and a second coding agent iteratively refines the prompt until tests pass. Deploying tool-using LLM agents in production requires measurable behavioral compliance that current development practices cannot provide. Small prompt changes cause silent regressions, tool misuse goes undetected, and policy violations emerge only after deployment. To mitigate specification gaming, TDAD introduces three mechanisms: (1) visible/hidden test splits that withhold evaluation tests during compilation, (2) semantic mutation testing via a post-compilation agent that generates plausible faulty prompt variants, with the harness measuring whether the test suite detects them, and (3) spec evolution scenarios that quantify regression safety when requirements change. We evaluate TDAD on SpecSuite-Core, a benchmark of four deeply-specified agents spanning policy compliance, grounded analytics, runbook adherence, and deterministic enforcement. Across 24 independent trials, TDAD achieves 92% v1 compilation success with 97% mean hidden pass rate; evolved specifications compile at 58%, with most failed runs passing all visible tests except 1-2, and show 86-100% mutation scores, 78% v2 hidden pass rate, and 97% regression safety scores. The implementation is available as an open benchmark at https://github.com/f-labs-io/tdad-paper-code. --- ### The Temporal Markov Transition Field Authors: Michael Leznik Published: 2026-03-09 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2603.08803v1 The Markov Transition Field (MTF), introduced by Wang and Oates (2015), encodes a time series as a two-dimensional image by mapping each pair of time steps to the transition probability between their quantile states, estimated from a single global transition matrix. This construction is efficient when the transition dynamics are stationary, but produces a misleading representation when the process changes regime over time: the global matrix averages across regimes and the resulting image loses all information about \emph{when} each dynamical regime was active. In this paper we introduce the \emph{Temporal Markov Transition Field} (TMTF), an extension that partitions the series into $K$ contiguous temporal chunks, estimates a separate local transition matrix for each chunk, and assembles the image so that each row reflects the dynamics local to its chunk rather than the global average. The resulting $T \times T$ image has $K$ horizontal bands of distinct texture, each encoding the transition dynamics of one temporal segment. We develop the formal definition, establish the key structural properties of the representation, work through a complete numerical example that makes the distinction from the global MTF concrete, analyse the bias--variance trade-off introduced by temporal chunking, and discuss the geometric interpretation of the local transition matrices in terms of process properties such as persistence, mean reversion, and trending behaviour. The TMTF is amplitude-agnostic and order-preserving, making it suitable as an input channel for convolutional neural networks applied to time series characterisation tasks. --- ### Large Language Model-Assisted Superconducting Qubit Experiments Authors: Shiheng Li, Jacob M. Miller, Phoebe J. Lee, Gustav Andersson, Christopher R. Conner, Yash J. Joshi, Bayan Karimi, Amber M. King, Howard L. Malc, Harsh Mishra, Hong Qiao, Minseok Ryu et al. Published: 2026-03-09 Categories: quant-ph, cs.AI arXiv: https://arxiv.org/abs/2603.08801v1 Superconducting circuits have demonstrated significant potential in quantum information processing and quantum sensing. Implementing novel control and measurement sequences for superconducting qubits is often a complex and time-consuming process, requiring extensive expertise in both the underlying physics and the specific hardware and software. In this work, we introduce a framework that leverages a large language model (LLM) to automate qubit control and measurement. Specifically, our framework conducts experiments by generating and invoking schema-less tools on demand via a knowledge base on instrumental usage and experimental procedures. We showcase this framework with two experiments: an autonomous resonator characterization and a direct reproduction of a quantum non-demolition (QND) characterization of a superconducting qubit from literature. This framework enables rapid deployment of standard control-and-measurement protocols and facilitates implementation of novel experimental procedures, offering a more flexible and user-friendly paradigm for controlling complex quantum hardware. --- ### Granulon: Awakening Pixel-Level Visual Encoders with Adaptive Multi-Granularity Semantics for MLLM Authors: Junyuan Mao, Qiankun Li, Linghao Meng, Zhicheng He, Xinliang Zhou, Kun Wang, Yang Liu, Yueming Jin Published: 2026-03-09 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.08800v1 Recent advances in multimodal large language models largely rely on CLIP-based visual encoders, which emphasize global semantic alignment but struggle with fine-grained visual understanding. In contrast, DINOv3 provides strong pixel-level perception yet lacks coarse-grained semantic abstraction, leading to limited multi-granularity reasoning. To address this gap, we propose Granulon, a novel DINOv3-based MLLM with adaptive granularity augmentation. Granulon introduces a text-conditioned granularity Controller that dynamically adjusts the visual abstraction level according to the semantic scope of the textual input, and an Adaptive Token Aggregation module that performs granularity-guided pooling and relation-aware clustering to produce compact, semantically rich visual tokens. This design enables unified "pixel-to-fine-to-coarse" reasoning within a single forward pass. Extensive and interpretable experiments demonstrate that Granulon improves accuracy by ~30% and reduces hallucination by ~20%, outperforming all visual encoders under identical settings. --- ### Multi-level meta-reinforcement learning with skill-based curriculum Authors: Sichen Yang, Mauro Maggioni Published: 2026-03-09 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2603.08773v1 We consider problems in sequential decision making with natural multi-level structure, where sub-tasks are assembled together to accomplish complex goals. Systematically inferring and leveraging hierarchical structure has remained a longstanding challenge; we describe an efficient multi-level procedure for repeatedly compressing Markov decision processes (MDPs), wherein a parametric family of policies at one level is treated as single actions in the compressed MDPs at higher levels, while preserving the semantic meanings and structure of the original MDP, and mimicking the natural logic to address a complex MDP. Higher-level MDPs are themselves independent MDPs with less stochasticity, and may be solved using existing algorithms. As a byproduct, spatial or temporal scales may be coarsened at higher levels, making it more efficient to find long-term optimal policies. The multi-level representation delivered by this procedure decouples sub-tasks from each other and usually greatly reduces unnecessary stochasticity and the policy search space, leading to fewer iterations and computations when solving the MDPs. A second fundamental aspect of this work is that these multi-level decompositions plus the factorization of policies into embeddings (problem-specific) and skills (including higher-order functions) yield new transfer opportunities of skills across different problems and different levels. This whole process is framed within curriculum learning, wherein a teacher organizes the student agent's learning process in a way that gradually increases the difficulty of tasks and and promotes transfer across MDPs and levels within and across curricula. The consistency of this framework and its benefits can be guaranteed under mild assumptions. We demonstrate abstraction, transferability, and curriculum learning in examples, including MazeBase+, a more complex variant of the MazeBase example. --- ### Micro-Diffusion Compression - Binary Tree Tweedie Denoising for Online Probability Estimation Authors: Roberto Tacconelli Published: 2026-03-09 Categories: stat.ML, cs.IT, cs.LG arXiv: https://arxiv.org/abs/2603.08771v3 We present Midicoth, a lossless compression system that introduces a micro-diffusion denoising layer for improving probability estimates produced by adaptive statistical models. In compressors such as Prediction by Partial Matching (PPM), probability estimates are smoothed by a prior to handle sparse observations. When contexts have been seen only a few times, this prior dominates the prediction and produces distributions that are significantly flatter than the true source distribution, leading to compression inefficiency. Midicoth addresses this limitation by treating prior smoothing as a shrinkage process and applying a reverse denoising step that corrects predicted probabilities using empirical calibration statistics. To make this correction data-efficient, the method decomposes each byte prediction into a hierarchy of binary decisions along a bitwise tree. This converts a single 256-way calibration problem into a sequence of binary calibration tasks, enabling reliable estimation of correction terms from relatively small numbers of observations. The denoising process is applied in multiple successive steps, allowing each stage to refine residual prediction errors left by the previous one. The micro-diffusion layer operates as a lightweight post-blend calibration stage applied after all model predictions have been combined, allowing it to correct systematic biases in the final probability distribution. Midicoth combines five fully online components: an adaptive PPM model, a long-range match model, a trie-based word model, a high-order context model, and the micro-diffusion denoiser applied as the final stage. --- ### SPREAD: Subspace Representation Distillation for Lifelong Imitation Learning Authors: Kaushik Roy, Giovanni D'urso, Nicholas Lawrance, Brendan Tidd, Peyman Moghadam Published: 2026-03-09 Categories: cs.LG, cs.RO arXiv: https://arxiv.org/abs/2603.08763v1 A key challenge in lifelong imitation learning (LIL) is enabling agents to acquire new skills from expert demonstrations while retaining prior knowledge. This requires preserving the low-dimensional manifolds and geometric structures that underlie task representations across sequential learning. Existing distillation methods, which rely on L2-norm feature matching in raw feature space, are sensitive to noise and high-dimensional variability, often failing to preserve intrinsic task manifolds. To address this, we introduce SPREAD, a geometry-preserving framework that employs singular value decomposition (SVD) to align policy representations across tasks within low-rank subspaces. This alignment maintains the underlying geometry of multimodal features, facilitating stable transfer, robustness, and generalization. Additionally, we propose a confidence-guided distillation strategy that applies a Kullback-Leibler divergence loss restricted to the top-M most confident action samples, emphasizing reliable modes and improving optimization stability. Experiments on the LIBERO, lifelong imitation learning benchmark, show that SPREAD substantially improves knowledge transfer, mitigates catastrophic forgetting, and achieves state-of-the-art performance. --- ### Scale Space Diffusion Authors: Soumik Mukhopadhyay, Prateksha Udhayanan, Abhinav Shrivastava Published: 2026-03-09 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.08709v1 Diffusion models degrade images through noise, and reversing this process reveals an information hierarchy across timesteps. Scale-space theory exhibits a similar hierarchy via low-pass filtering. We formalize this connection and show that highly noisy diffusion states contain no more information than small, downsampled images - raising the question of why they must be processed at full resolution. To address this, we fuse scale spaces into the diffusion process by formulating a family of diffusion models with generalized linear degradations and practical implementations. Using downsampling as the degradation yields our proposed Scale Space Diffusion. To support Scale Space Diffusion, we introduce Flexi-UNet, a UNet variant that performs resolution-preserving and resolution-increasing denoising using only the necessary parts of the network. We evaluate our framework on CelebA and ImageNet and analyze its scaling behavior across resolutions and network depths. Our project website ( https://prateksha.github.io/projects/scale-space-diffusion/ ) is available publicly. --- ### FVG-PT: Adaptive Foreground View-Guided Prompt Tuning for Vision-Language Models Authors: Haoyang Li, Liang Wang, Siyu Zhou, Jiacheng Sun, Jing Jiang, Chao Wang, Guodong Long, Yan Peng Published: 2026-03-09 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.08708v1 CLIP-based prompt tuning enables pretrained Vision-Language Models (VLMs) to efficiently adapt to downstream tasks. Although existing studies have made significant progress, they pay limited attention to changes in the internal attention representations of VLMs during the tuning process. In this paper, we attribute the failure modes of prompt tuning predictions to shifts in foreground attention of the visual encoder, and propose Foreground View-Guided Prompt Tuning (FVG-PT), an adaptive plug-and-play foreground attention guidance module, to alleviate the shifts. Concretely, FVG-PT introduces a learnable Foreground Reliability Gate to automatically enhance the foreground view quality, applies a Foreground Distillation Compensation module to guide visual attention toward the foreground, and further introduces a Prior Calibration module to mitigate generalization degradation caused by excessive focus on the foreground. Experiments on multiple backbone models and datasets show the effectiveness and compatibility of FVG-PT. Codes are available at: https://github.com/JREion/FVG-PT --- ### Impermanent: A Live Benchmark for Temporal Generalization in Time Series Forecasting Authors: Azul Garza, Renée Rosillo, Rodrigo Mendoza-Smith, David Salinas, Andrew Robert Williams, Arjun Ashok, Mononito Goswami, José Martín Juárez Published: 2026-03-09 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.08707v2 Recent advances in time-series forecasting increasingly rely on pre-trained foundation-style models. While these models often claim broad generalization, existing evaluation protocols provide limited evidence. Indeed, most current benchmarks use static train-test splits that can easily lead to contamination as foundation models can inadvertently train on test data or perform model selection using test scores, which can inflate performance. We introduce Impermanent, a live benchmark that evaluates forecasting models under open-world temporal change by scoring forecasts sequentially over time on continuously updated data streams, enabling the study of temporal robustness, distributional shift, and performance stability rather than one-off accuracy on a frozen test set. Impermanent is instantiated on GitHub open-source activity, providing a naturally live and highly non-stationary dataset shaped by releases, shifting contributor behavior, platform/tooling changes, and external events. We focus on the top 400 repositories by star count and construct time series from issues opened, pull requests opened, push events, and new stargazers, evaluated over a rolling window with daily updates, alongside standardized protocols and leaderboards for reproducible, ongoing comparison. By shifting evaluation from static accuracy to sustained performance, Impermanent takes a concrete step toward assessing when and whether foundation-level generalization in time-series forecasting can be meaningfully claimed. Code and a live dashboard are available at https://github.com/TimeCopilot/impermanent and https://impermanent.timecopilot.dev. --- ### Agentic Critical Training Authors: Weize Liu, Minghui Liu, Sy-Tuyen Ho, Souradip Chakraborty, Xiyao Wang, Furong Huang Published: 2026-03-09 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2603.08706v1 Training large language models (LLMs) as autonomous agents often begins with imitation learning, but it only teaches agents what to do without understanding why: agents never contrast successful actions against suboptimal alternatives and thus lack awareness of action quality. Recent approaches attempt to address this by introducing self-reflection supervision derived from contrasts between expert and alternative actions. However, the training paradigm fundamentally remains imitation learning: the model imitates pre-constructed reflection text rather than learning to reason autonomously. We propose Agentic Critical Training (ACT), a reinforcement learning paradigm that trains agents to identify the better action among alternatives. By rewarding whether the model's judgment is correct, ACT drives the model to autonomously develop reasoning about action quality, producing genuine self-reflection rather than imitating it. Across three challenging agent benchmarks, ACT consistently improves agent performance when combined with different post-training methods. It achieves an average improvement of 5.07 points over imitation learning and 4.62 points over reinforcement learning. Compared to approaches that inject reflection capability through knowledge distillation, ACT also demonstrates clear advantages, yielding an average improvement of 2.42 points. Moreover, ACT enables strong out-of-distribution generalization on agentic benchmarks and improves performance on general reasoning benchmarks without any reasoning-specific training data, highlighting the value of our method. These results suggest that ACT is a promising path toward developing more reflective and capable LLM agents. --- ### Evaluating Financial Intelligence in Large Language Models: Benchmarking SuperInvesting AI with LLM Engines Authors: Akshay Gulati, Kanha Singhania, Tushar Banga, Parth Arora, Anshul Verma, Vaibhav Kumar Singh, Agyapal Digra, Jayant Singh Bisht, Danish Sharma, Varun Singla, Shubh Garg Published: 2026-03-09 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.08704v1 Large language models are increasingly used for financial analysis and investment research, yet systematic evaluation of their financial reasoning capabilities remains limited. In this work, we introduce the AI Financial Intelligence Benchmark (AFIB), a multi-dimensional evaluation framework designed to assess financial analysis capabilities across five dimensions: factual accuracy, analytical completeness, data recency, model consistency, and failure patterns. We evaluate five AI systems: GPT, Gemini, Perplexity, Claude, and SuperInvesting, using a dataset of 95+ structured financial analysis questions derived from real-world equity research tasks. The results reveal substantial differences in performance across models. Within this benchmark setting, SuperInvesting achieves the highest aggregate performance, with an average factual accuracy score of 8.96/10 and the highest completeness score of 56.65/70, while also demonstrating the lowest hallucination rate among evaluated systems. Retrieval-oriented systems such as Perplexity perform strongly on data recency tasks due to live information access but exhibit weaker analytical synthesis and consistency. Overall, the results highlight that financial intelligence in large language models is inherently multi-dimensional, and systems that combine structured financial data access with analytical reasoning capabilities provide the most reliable performance for complex investment research workflows. ---