# Research Papers — 2026-06 · bundle page 203 of 260 > Full-text markdown of papers #10101–10150 of 12993 published in 2026-06, newest first. AI-readable mirrors of the full 12993-paper 2026-06 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-06/page/203 [← Previous bundle](https://aipod.dev/papers/archive/2026-06/page/202.md) [Next bundle →](https://aipod.dev/papers/archive/2026-06/page/204.md) ### DALE-CT: Depth-Aware Foundation Models for Computed Tomography Authors: Evan W. Damron, Mahmut S. Gokmen, Mitchell A. Klusty, Caroline N. Leach, Emily B. Collier, V. K. Cody Bumgardner Published: 2026-06-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.07775v1 Recent breakthroughs in self-supervised learning (SSL), such as the Latent-Euclidean Joint-Embedding Predictive Architecture (LeJEPA), alongside successes in integrating visual encoders with language models, have driven the demand for adaptable, high-capacity vision encoders in Computed Tomography (CT). In this work, we explore 2D slice-based architectures as a flexible alternative to native 3D models for processing volumetric CT data. Using the CT-RATE dataset, we trained DALE-CT (Depth-Aware Latent-Euclidean Computed Tomography), a 2D model family built entirely from scratch using LeJEPA, and compared its performance against a continually pre-trained DINOv2 baseline. To enhance representation quality, we developed a novel 3D depth-aware pre-training strategy anchored by dense auxiliary supervision from both automated anatomical masks and human-annotated abnormalities. Under linear probe evaluation with Multiple Instance Learning (MIL) for multi-abnormality detection, the frozen backbone of this dual-supervised model (DALE-CT-2S) achieves a Macro AUROC of 0.833. This performance demonstrates near-parity with state-of-the-art 3D vision-language models, achieved entirely from scratch with significantly less data and no textual supervision. To ensure reproducibility, all training code, evaluation scripts, and model weights have been made publicly available. --- ### Beyond Point Estimates: Benchmarking Uncertainty Quantification Methods on the AION-1 Astronomical Foundation Model Authors: Karla Tame-Narvaez, Aleksandra Ćiprijanović, Shubhendu Trivedi Published: 2026-06-05 Categories: astro-ph.IM, astro-ph.GA, cs.AI arXiv: https://arxiv.org/abs/2606.07771v1 Foundation models for astronomical surveys offer powerful learned representations that can be transferred to downstream regression tasks such as galaxy property estimation. However, point predictions alone are insufficient for scientific inference; reliable uncertainty quantification (UQ) is essential. We compare seven UQ methods on galaxy property regression using frozen AION-1 foundation-model embeddings, predicting redshift, stellar mass, stellar-population age, gas-phase metallicity, and specific star-formation rate, from Legacy Survey photometry/imaging and DESI spectra, with PROVABGS-derived labels. Distribution-free conformal methods achieve marginal coverage within $\sim$1\,pp of the nominal 90\% across all properties, while non-conformal baselines (Deep Ensembles, MC~Dropout) fail to calibrate reliably. Among conformal approaches, Conformalized Quantile Regression (CQR) delivers the best coverage in the bin with the poorest model predictions. More importantly, only the Locally Valid and Discriminative (LVD) framework -- particularly when operating on AION-1 embeddings -- also provides finite-sample \emph{local validity}, producing intervals that adapt to each galaxy's local prediction difficulty rather than relying on marginal guarantees alone. These results establish conformal prediction, and LVD in particular, as the preferred UQ framework for uncertainty-aware inference on foundation-model embeddings in astrophysics. --- ### Contrast encodes inductive bias: separating slow noise from dynamics in predictive representation learning Authors: Paarth Gulati, Ilya Nemenman Published: 2026-06-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2606.07770v1 Self-supervised methods that learn representations and predict dynamics fully in the latent space, such as JEPA, have been shown to confuse slowly varying noise with the dynamical signals they aim to capture. Specifically, when noise features remain approximately constant within each trajectory, contrastive predictive objectives preferentially encode these features instead of the true latent variables governing the system. The learned representation then becomes dominated by trajectory-specific noise, so downstream performance degrades with noise strength and does not improve even as the number and duration of training trajectories increase. We argue that this failure is a property of the objective itself, shared by a long line of contrastive predictive objectives that sample negatives across trajectories. To illustrate this generality, we study the failure mode and its remedy in two settings: a standard SimCLR-style JEPA on a synthetic moving-dot dataset, and DySIB, a recently introduced method designed for extracting physically interpretable representations of dynamics, on movies of a rigid-body pendulum. When negatives are instead sampled within a single trajectory, the slow noise can no longer distinguish frames within that trajectory, removing the predictive shortcut. Training one encoder simultaneously on many such trajectories then forces it to encode the variables relevant for the dynamics, with longer trajectories yielding better representations even for strong slow noise. Our results point toward principles for designing contrastive predictive objectives in dynamical representation learning, especially for physical systems with noisy experimental observations. --- ### Quantum-Enhanced Similarity Measures for Polarimetric Materials Classification Authors: Sara Shojaei, Seyed Mohamad Ali Tousi, Emma Bennett, Param Sangani, Ali Shiri Sichani, Ilker Ersoy, Hadi Ali-Akbarpour, Filiz Bunyak, G. N. DeSouza Published: 2026-06-05 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2606.07766v1 We present a quantum--classical hybrid pipeline for polarimetric material classification that casts this as a point-matching problem. Voxel cubes, containing polarized light reflections, are used to train an encoder to produce 32-dimensional embeddings for the voxels of the cubes. At inference, the encoder head is discarded and the embeddings are encoded as probability amplitudes of quantum states. Next, a SWAP-test circuit estimates the fidelity between each of the 32D embeddings from the query cube and a dataset of anchor cubes. The aggregated fidelity serves as materials similarity scores, and the class of the anchor with highest aggregated fidelity is deemed as the class of the queried material. We evaluate our approach on a dataset of 23 materials ($\approx$800 samples each) derived from their Mueller matrices. The point-matching approaches from the proposed quantum SWAP-test and a classical classifier using Optimal Transport are compared. Our results demonstrate the competitive classification accuracy alongside open-set discrimination potential, establishing it as a viable path toward NISQ-based material recognition. --- ### scCBGM: Interpretable Single-Cell Counterfactual Editing Authors: Alma Andersson, Aya Abdelsalam Ismail, Edward De Brouwer, Doron Haviv, Tommaso Biancalani, Kyunghyun Cho, Gabriele Scalia, Aïcha BenTaieb, Hector Corrada Bravo Published: 2026-06-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2606.07760v1 Understanding cellular phenotypes and how they respond to perturbations is critical for disease biology and therapeutic design. Single-cell RNA sequencing enables characterization at cellular resolution, yet the combinatorial space of conditions makes exhaustive experimental mapping infeasible. We introduce single-cell Concept Bottleneck Generative Models (scCBGM), a framework for interpretable and precise counterfactual editing of individual cells. scCBGM adapts concept bottleneck architectures for single-cell data through decoder skip connections and a cross-covariance penalty that promotes disentanglement without dimensional constraints. We extend the framework to flow matching models, enabling concept-guided editing in both encoding-decoding and generation regimes. To enable rigorous evaluation, we develop a synthetic benchmark with ground-truth counterfactuals. Across multiple real datasets, scCBGM demonstrates superior performance in combinatorial generalization and counterfactual prediction, supported by cell-level validation on synthetic data and population-level benchmarks on real datasets. --- ### DroneDAR: Long-Range Drone Distance Estimation Using Monocular Vision and Bounding-Box Features Authors: Knut Peterson, Zaid Mayers, David Han Published: 2026-06-05 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2606.07756v1 Accurate distance estimation for small drones in long-range imagery is important for tracking and situational awareness, yet remains challenging due to extreme target scale variation, background clutter, and noisy visual cues. This paper studies monocular drone distance estimation using image crops together with bounding-box geometry, a practical setting in which a detector provides a candidate drone region and the model predicts range from appearance and box-derived features. We evaluate a Droneranger-style baseline, and introduce a new DroneDAR (Drone Detection And Ranging) model that combines a convolutional backbone with explicit bounding-box cues through a lightweight gating mechanism. Experiments analyze how backbone capacity, crop resolution, and regression loss functions affect performance across distance regimes. We further examine common failure modes at long distances, including sensitivity to bounding-box noise and reduced texture detail in the crop. The results provide guidance for designing and training range estimators that remain robust under real-world long-range conditions and highlight directions for improving reliability when drones occupy only a few pixels. --- ### ReadingMachine: A Computational Methodology for Structured Corpus Reading and Large-Scale Synthesis Authors: James Morrissey Published: 2026-06-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2606.07753v2 ReadingMachine is a computational methodology for structured corpus reading that uses large language models to perform bounded reading operations over entire document collections. Rather than relying on retrieval or recursive summarization, the approach decomposes analysis into inspectable stages including insight extraction, semantic clustering, theme generation, and iterative omission detection. By delaying irreversible compression and explicitly tracking intermediate representations, the method prioritizes coverage, traceability, and preservation of disagreement across large corpora. The system is demonstrated on a heterogeneous corpus of 152 industrial policy documents, producing more than 17,500 extracted insights and a structured thematic map. ReadingMachine is released as an open-source experimental framework for large-scale qualitative synthesis and corpus analysis. --- ### ReadingMachine: A Computational Methodology for Structured Corpus Reading and Large-Scale Synthesis Authors: James Morrissey Published: 2026-06-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2606.07753v1 ReadingMachine is a computational methodology for structured corpus reading that uses large language models to perform bounded reading operations over entire document collections. Rather than relying on retrieval or recursive summarization, the approach decomposes analysis into inspectable stages including insight extraction, semantic clustering, theme generation, and iterative omission detection. By delaying irreversible compression and explicitly tracking intermediate representations, the method prioritizes coverage, traceability, and preservation of disagreement across large corpora. The system is demonstrated on a heterogeneous corpus of 152 industrial policy documents, producing more than 17,500 extracted insights and a structured thematic map. ReadingMachine is released as an open-source experimental framework for large-scale qualitative synthesis and corpus analysis. --- ### Characterizing the Discrete Geometry of ReLU Networks Authors: Blake B. Gaines, Jinbo Bi Published: 2026-06-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2606.07728v1 It is well established that ReLU networks define continuous piecewise-linear functions, and that their linear regions are polyhedra in the input space. These regions form a complex that fully partitions the input space. The way these regions fit together is fundamental to the behavior of the network, as nonlinearities occur only at the boundaries where these regions connect. However, relatively little is known about the geometry of these complexes beyond bounds on the total number of regions, and calculating the complex exactly is intractable for most networks. In this work, we prove new theoretical results about these complexes that hold for all fully-connected ReLU networks, specifically about their connectivity graphs in which nodes correspond to regions and edges exist between each pair of regions connected by a face. We find that the average degree of this graph is upper bounded by twice the input dimension regardless of the width and depth of the network, and that the diameter of this graph has an upper bound that does not depend on input dimension, despite the number of regions increasing exponentially with input dimension. We corroborate our findings through experiments with networks trained on both synthetic and real-world data, which provide additional insight into the geometry of ReLU networks. Code to reproduce our results can be found at https://github.com/bl-ake/ICLR-2026. --- ### Benchmarking Quantum Algorithmic Resilience for CVaR Portfolio Optimization: The Expressibility-Coherence Trade-off Authors: Prashik N. Somkuwar, K. Srinivasan, G. Raghavan Published: 2026-06-05 Categories: quant-ph, cs.CL, math.OC, q-fin.PM arXiv: https://arxiv.org/abs/2606.07727v1 Quantum combinatorial optimization offers theoretical advantages for complex financial modeling, but physical implementation on Noisy Intermediate Scale Quantum (NISQ) devices is severely constrained by hardware topology. This study presents a hardware benchmarking analysis between a Hardware Efficient Variational Quantum Neural Network (HE-VQNN) and the Warm Start Quantum Approximate Optimization Algorithm (WS-QAOA) for a hybrid Mean Variance and Conditional Value at Risk (CVaR) portfolio objective. By implementing a novel classical quantum hybrid proxy matrix to bypass the CVaR auxiliary qubit bottleneck, we map up to 16 assets from the NIFTY 50 index onto an IBM heavy hex processor. We systematically quantify algorithmic resilience to the "SWAP tax" incurred during routing. Empirical results reveal a critical operational trade-off: WS-QAOA provides exact theoretical mapping but suffers catastrophic hardware decoherence due to exponential nonlocal gate overhead. Conversely, HE-VQNN preserves hardware coherence but lacks the mathematical expressibility to capture dense tail risk asset correlations. This study exposes the limitations of dense financial optimization on current architectures forces an nonviable choice between algorithmic inexpressibility and hardware decoherence. This is indicative of a deeper limitation as to what can and cannot be done with NISQ computers lacking in all-to-all connectivity. --- ### Cutting LLM Evaluation Costs with SySRs: A Bandit Algorithm that Provably Exploits Model Similarity Authors: Zifan Lyu, Chahine Nejma, Tobias Wegel, Fanny Yang, Florian E. Dorner Published: 2026-06-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2606.07726v1 Large Language Models are typically benchmarked by evaluating every model on every test query. For practitioners seeking the best model to deploy, this is often wasteful: if a model clearly performs worse than others, there is no need to precisely estimate its performance. Best-arm identification algorithms can be naturally applied to drastically reduce costs by adaptively allocating evaluation budget. Further, language models often respond similarly to the same prompt-a property previous work has tried to leverage with mixed success. We propose Synchronized Successive Rejects (SySRs), augmenting the classical Successive Rejects algorithm with paired comparisons. Unlike prior attempts to leverage model similarity in best-model identification, our approach is hyperparameter-free and enjoys performance guarantees that improve with the degree of similarity between evaluated models. Empirically, our method outperforms all baselines in terms of average error rate across 15 standard benchmarks, and in terms of worst-case budget for reliably identifying the best model. --- ### GNSS-FM: A Self-Supervised Foundation Model for Daily GNSS Displacement Time Series Authors: Nick Teutschmann, Laura Crocetti, Fanny Lehmann, Leonardo Trentini, Benedikt Soja Published: 2026-06-05 Categories: physics.geo-ph, cs.LG arXiv: https://arxiv.org/abs/2606.07725v1 Displacement time series from Global Navigation Satellite Systems (GNSS) are essential for a wide range of applications, including monitoring tectonic crustal deformations and investigating the different stages of the earthquake cycle. Machine learning methods have proven promising for GNSS applications; however, most remain fully supervised. This creates a bottleneck as labeled data are scarce, even though large amounts of unlabeled GNSS data are freely available. We present GNSS-FM, a self-supervised foundation model for daily GNSS time series. The model uses a dual-stream input combining displacement and velocity-like increments, and is pretrained using a masked latent prediction objective with vector-quantized targets adapted from wav2vec 2.0, with several modifications for geodetic data. Pretrained on data from over 17,000 globally distributed GNSS stations, an analysis of the learned codebook suggests that the representations capture the main signal types in GNSS displacement data, including seismic offsets, tectonic drift, and seasonal patterns. The foundation model is later fine-tuned on two downstream tasks, namely 90-day displacement forecasting and seismic step localization, where it outperforms strong task-specific baselines in both cases. These results show that self-supervised pretraining is a promising approach for GNSS time series analysis. --- ### A Geometry-Aware Triplane Field Network for Vehicle Aerodynamic Prediction Authors: Kangkang Qi, Huiyu Yang, Keqi Ding, Yunpeng Wang, Yuntian Chen, Yuanwei Bin, Rikui Zhang, Jianchun Wang Published: 2026-06-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2606.07724v2 High-fidelity computational fluid dynamics (CFD) is crucial to vehicle aerodynamic analysis, but its cost still constrains early-stage design exploration. Machine-learning-based surface-field prediction offers a faster alternative if the model can efficiently capture both global flow context and local geometric detail. This work proposes a machine-learning-based method, named the geometry-aware triplane field network (GTF-Net), for vehicle aerodynamic pressure and wall shear stress prediction. GTF-Net constructs triplane features directly from sampled surface points through a shared multilayer perceptron (MLP) and smooth bilinear rasterization. The planes are then processed by a dual-stream backbone that combines adaptive Fourier neural operator (AFNO) spectral mixing with convolutional neural network (CNN) refinement, so long-range aerodynamic coupling and local geometry-induced variations are modeled in the same representation. At query stage, sampled triplane features are combined with vehicle-aligned directional coordinates, normal-projection features, and a voxel-based curvature proxy. GTF-Net is compared with Transolver, geometry-informed neural operator (GINO), and TripNet, a triplane-based surrogate model. GTF-Net improves the relative L2 error from the strongest baseline value of 0.157 to 0.145 for pressure prediction and from 0.237 to 0.226 for wall shear stress prediction. Ablation results show that AFNO mixing, local CNN refinement, and query-side geometric encoding each contribute to accuracy, supporting the proposed mechanism of combining structured triplane representation with explicit aerodynamic geometry cues. --- ### A Geometry-Aware Triplane Field Network for Vehicle Aerodynamic Prediction Authors: Kangkang Qi, Huiyu Yang, Keqi Ding, Yunpeng Wang, Yuntian Chen, Yuanwei Bin, Rikui Zhang, Jianchun Wang Published: 2026-06-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2606.07724v1 High-fidelity computational fluid dynamics (CFD) is crucial to vehicle aerodynamic analysis, but its cost still constrains early-stage design exploration. Machine-learning-based surface-field prediction offers a faster alternative if the model can efficiently capture both global flow context and local geometric detail. This work proposes a machine-learning-based method, named the geometry-aware triplane field network (GTF-Net), for vehicle aerodynamic pressure and wall shear stress prediction. GTF-Net constructs triplane features directly from sampled surface points through a shared multilayer perceptron (MLP) and smooth bilinear rasterization. The planes are then processed by a dual-stream backbone that combines adaptive Fourier neural operator (AFNO) spectral mixing with convolutional neural network (CNN) refinement, so long-range aerodynamic coupling and local geometry-induced variations are modeled in the same representation. At query stage, sampled triplane features are combined with vehicle-aligned directional coordinates, normal-projection features, and a voxel-based curvature proxy. GTF-Net is compared with Transolver, geometry-informed neural operator (GINO), and TripNet, a triplane-based surrogate model. GTF-Net improves the relative L2 error from the strongest baseline value of 0.157 to 0.145 for pressure prediction and from 0.237 to 0.226 for wall shear stress prediction. Ablation results show that AFNO mixing, local CNN refinement, and query-side geometric encoding each contribute to accuracy, supporting the proposed mechanism of combining structured triplane representation with explicit aerodynamic geometry cues. --- ### Some hypotheses on how chatbots work in problem-solution-driven conversations: Large Language Models as confirmation of the Innovation Illusion Authors: S. F. M. van Vlijmen, H. D. Lethe Published: 2026-06-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2606.07722v5 We discuss the nature of chatbots as conversation partners in problem-solving conversations. What can chatbots do and what can't they do? Our analysis draws on insights from Aggregation Dynamics, Cognitive Linguistics, Neuropsychology and Psychology. We establish that chatbots are multifaceted and composite systems. Our argument focuses on basic chatbots in the hope of thereby making statements about the core functionality of more advanced chatbots. Basic chatbots are assumed to consist of a Large Language Model (LLM) with a simple interface. The main results of our analysis are: a description of human imagination, understanding and thinking based on so-called metaphorical problem propagations; that the texts in text datasets used for training LLMs have specific characteristics and that these texts only partially imitate human thinking and understanding; that the LLM training process encodes artificial metaphorical problem propagations into an LLM from these text datasets. Our conclusions are that a basic chatbot cannot be a thinking partner capable of matching the cognitive flexibility of humans, and that further development of LLMs will not lead to this either. But chatbots exist, they are being used on a massive scale, by both individuals and organisations. It is therefore socially and politically important to understand them. Our article aims to contribute to the discussion on the functioning, benefits and drawbacks of chatbots. Cognitive Linguistics shows how the use of metaphor is an expression of our thinking. Aggregation Dynamics, is an attempt at a comprehensive systems theory. We believe that the concept of metaphorical problem propagation could provide an interesting addition for both. Chatbots a solution? For what? --- ### Some hypotheses on how chatbots work in problem-solving-driven conversations. Large Language Models as confirmation of the Innovation Illusion Authors: S. F. M. van Vlijmen, H. D. Lethe Published: 2026-06-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2606.07722v4 We discuss the nature of chatbots as conversation partners in problem-solving. What can chatbots do and what can't they do? We develop hypotheses on how this can this be explained. Our argument draws on insights from Aggregation Dynamics, Cognitive Linguistics, Neuropsychology and Psychology. We establish that chatbots are multifaceted and composite systems. Our argument focuses on basic chatbots in the hope of thereby making statements about the core functionality of more advanced chatbots. Basic chatbots are assumed to consist of a Large Language Model (LLM) with a simple interface. The main results of our research are: a description of human imagination, understanding and thinking based on so-called metaphorical problem propagations; the hypothesis that the texts in the text dataset used for training LLMs have specific characteristics and that these texts only partially imitate human thinking and understanding; the hypothesis that the LLM training process encodes artificial metaphorical problem propagations into an LLM from these text datasets. Our conclusions are that a basic chatbot cannot be a thinking partner capable of matching the cognitive flexibility of humans, and that further development of the Large Language Model will not lead to this either. But chatbots exist, they are being used on a massive scale, by both individuals and organisations. It is therefore socially and politically important to understand them. Our article aims to contribute to the discussion on the functioning, benefits and drawbacks of chatbots. Cognitive Linguistics shows how the use of metaphor is an expression of our thinking. Aggregation Dynamics, is an attempt at a comprehensive systems theory. We believe that the concept of metaphorical problem propagation could provide an interesting addition for both. Chatbots a solution? For what? --- ### Automatic Extraction of Structured Information from Brain MRI Reports Using an Open-Weight Large Language Model Authors: Kaouther Mouheb, Amos Pomp, Antoine Manenti, Romy de Haan, Farog Faghir, Joy Martens, Harro Seelaar, Francesco Mattace-Raso, Meike W. Vernooij, Frank J. Wolters, Stefan Klein, Esther E. Bron Published: 2026-06-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2606.07721v1 Objectives: Automatic data extraction from free-text radiology reports enables large-scale research, but few studies assessed the performance of large language models (LLMs) on Dutch neuroradiology reports. Methods: We analyzed 947 brain MRI reports from a tertiary memory clinic (2016-2021), authored by consultant neuroradiologists. Trained medical students annotated thirty variables; 100 reports were double-annotated to assess inter-rater reliability. We evaluated the performance of the open-weight LLM LLaMA 3.1 using different languages (Dutch vs. English translation) and few-shot prompting with different example selection strategies. Performance was evaluated using balanced accuracy for categorical variables, accuracy and mean absolute error for counts, and text similarity for free-text. Metrics were computed across 10 random splits of the 947 reports. Results: LLaMA 3.1 demonstrated high zero-shot performance for visual rating scores (mean (95%-CI)): Medial Temporal Atrophy: 90% (77-100%) on the left and 96% (94-99%) on the right, Global Cortical Atrophy: 87% (83-91%), and Fazekas: 94% (93-96%). Microbleed mentions were detected with 93% accuracy (92-95%) and infarct mentions with 82% (80-84%). Text similarity for lesion location reached 0.95 (0.95-0.96). Performance was lower for numerical variables: 80% (78-82%) for the number of microbleeds and 66% (63-68%) for infarcts. English translation yielded comparable results. Few-shot prompting improved performance for numerical variables, achieving 92% (90-93%) for microbleeds and 81% (77-85%) for infarcts using structural similarity-based selection. Conclusion: LLaMA 3.1 shows strong potential for extracting data from Dutch neuroradiology reports. Few-shot prompting enhances performance for numerical variables, whereas challenges remain for location-specific variables. --- ### Why Limit the Residual Stream to Layers and Not Tokens? Persistent Memory for Continuous Latent Reasoning Authors: Mujtaba Farhan, Maheep Chaudhary Published: 2026-06-05 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2606.07720v1 Large language models (LLMs) have demonstrated remarkable reasoning abilities on mathematical and multi-hop planning tasks. The CoCoNuT (Chain of Continuous Thought) paradigm~\cite{hao2024coconut} extends this by enabling models to reason in latent space, exploring multiple reasoning paths simultaneously rather than committing to a single chain early on. However, we identify a limitation we term the \textbf{concept bottleneck}. At each reasoning pass, intermediate hidden states are overwritten, causing the model to lose critical facts computed in earlier steps as reasoning depth increases. We observe this empirically. On HotpotQA, vanilla CoCoNuT (10.4\% EM) fails to improve over the CoT baseline (11.0\% EM), and performance degrades with curriculum depth on GSM8K. To address this, we propose \textbf{AGCLR} (Adaptive Gated Continuous Latent Reasoning), which augments CoCoNuT with a \textit{Gated Concept Stream}. A persistent residual memory maintained across all reasoning passes, controlled by three learned gates: a \textit{write} gate that commits intermediate facts to memory, a \textit{read} gate that retrieves relevant prior states, and a \textit{forget} gate that prunes irrelevant context. Evaluated on GSM8K, HotpotQA, and ProsQA using GPT-2 as our base model, AGCLR achieves consistent improvements across all types of datasets. With the performance gap compounding as curriculum depth increases, directly resolving the concept bottleneck. Code available at https://anonymous.4open.science/r/JJJJ/README.md --- ### A case study of evaluating AI agents on a neuroscience data-to-discovery pipeline Authors: Kai A. Horstmann, Ethan Lin, Alice A. Robie, Jennifer J. Sun, Kristin Branson Published: 2026-06-05 Categories: cs.AI, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2606.07718v2 Agentic AI offers a promising path to automating software development bottlenecks in scientific research pipelines, particularly for stages that take domain experts days to months to build and where correctness and robustness matter more than implementation details. We present an empirical study of general-purpose coding agents on a fly optogenetics data-to-discovery pipeline. We assess agents on tasks and datasets substantially larger than existing benchmarks and evaluation criteria grounded in domain expert standards. We show that agents can solve several pipeline stages, suggesting stage-level automation is tractable. By analyzing agents' code iterations, we show they struggle most without a pre-defined criterion, when they must instead use their scientific judgment to assess their current solution. Mirroring scientists, they sometimes attempt visual inspection of intermediate outputs for self-evaluation, but largely fail to interpret what they see or act on it appropriately. Solving the end-to-end pipeline requires stringing together successes across all stages, which is beyond agents' current abilities. We identify challenges largely absent from existing benchmarks, including computational resource management and generalization to held-out data. Finally, we distill principles for constructing scientific tasks and rigorous evaluation criteria for open-ended problems. --- ### A case study of evaluating AI agents on a neuroscience data-to-discovery pipeline Authors: Kai A. Horstmann, Ethan Lin, Alice A. Robie, Jennifer J. Sun, Kristin Branson Published: 2026-06-05 Categories: cs.AI, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2606.07718v1 Agentic AI tools offer a promising path to automating software development bottlenecks in scientific research pipelines, particularly for stages that take domain experts days to months to build, where scientists care about correctness and robustness, not implementation details. We present an empirical study of general-purpose coding agents on a fly optogenetics data-to-discovery pipeline. We assess agents on tasks substantially larger than existing benchmarks, datasets orders of magnitude bigger, and evaluation criteria grounded in domain expert standards. We show that agents can solve several individual pipeline stages, suggesting stage-level automation is tractable. By analyzing agents' code iterations, we show that they struggle most when there is not a pre-defined criterion to iterate on, and they must instead use their scientific judgment to assess their current solution, a key open challenge. Mirroring scientific practice, they sometimes attempt visual inspection of intermediate outputs for self-evaluation, but largely fail to interpret what they see or act on it appropriately. Solving the end-to-end pipeline correctly requires stringing together successes across all pipeline stages, and this is beyond agents' current abilities. We identify challenges largely absent from existing benchmarks, including computational resource management and generalization to large held-out data collections. Finally, we distill principles for constructing scientific tasks and rigorous evaluation criteria for open-ended problems. --- ### Multi-planar 2D-U-Net Segmentation of 3D-CT Abdominal Organs augmented by Spatial Occurrence Maps Authors: Daria Kern, Negar Chabi, Souraj Adhikary, Andre Mastmeyer Published: 2026-06-05 Categories: eess.IV, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2606.07717v1 This work proposes a lightweight 2D-U-Net-based framework for segmenting five abdominal organs in large field-of-view 3D CT scans. The method combines coarse-to-fine segmentation, predictions from multiple anatomical planes, and additional fuzzy 3D spatial maps that provide anatomical location cues to improve segmentation accuracy. We combine multi-planar 2D-U-Net models augmented by a spatial occurrence map. The approach involves two main stages. First, the abdominal volume of interest region is detected by traversing the whole scan axially with a 2D-U-Net and determining the x-y-z-minimum and -maximum extents of the 5 abdominal organs of interest. Second, we use spatial occurrence maps to enhance our multi-planar 2D-U-net architecture inside the bounds from the former stage. The method is evaluated on 80 CT scans from various public sources. The results show Dice improvements of about 4% at maximum compared to the same model trained without spatial occurrence maps. --- ### SHIELD-IDS: Structurally Heterogeneous Ensemble with Integrated Layered Defense for Intrusion Detection Systems Authors: Maryam Zaman, Muhammad Khuram Shahzad Published: 2026-06-05 Categories: cs.CR, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2606.07716v1 Adversarial attacks pose a serious and growing threat to Machine Learning (ML)-based Intrusion Detection Systems (IDS), where imperceptible perturbations to network flow features can systematically mislead classifiers into accepting malicious traffic as benign. The IDS-Anta framework partially addresses this through Z-score normalization, Singular Value Decomposition (SVD), and Multi-Armed Bandit (MAB) classifier selection with Thompson Sampling, yet its classifier pool lacks sufficient structural diversity for robust adversarial resistance. This work introduces IDS-Anta++, which incorporates XGBoost and LightGBM gradient boosting models into the ensemble and wraps the extended pool in a three-layer black-box defense: Isolation Forest anomaly screening, median feature smoothing, and six-way majority voting. Experiments conducted on CIC-IDS-2017, CEC-CIC-IDS-2018, and CIC-DDoS-2019 under both Fast Gradient Sign Method (FGSM) and Zeroth Order Optimization (ZOO) attacks confirm detection accuracy above 99% on clean data, with measurable robustness gains under adversarial conditions relative to the baseline IDS-Anta configuration. --- ### Beyond Accuracy: Interpreting Topic Representation in Suicide Ideation Detection Models Authors: Hamideh Ghanadian, Isar Nejadgholi, Hussein Al Osman Published: 2026-06-05 Categories: cs.LG, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2606.07714v1 Suicide ideation detection models are typically evaluated using aggregate performance metrics, yet little is known about how they internally represent psychologically meaningful risk factors. In high-stakes mental health applications, understanding these internal representations is essential for safety, transparency, and responsible deployment. In this work, we move beyond accuracy and analyze how suicide detection models trained on original and topic-augmented datasets encode psychological risk factors in their internal representation space. Using visualization and geometric analysis, we examine the coherence and separability of topic-related features. Our results show that topic-aware augmentation increases the clarity and distinctness of underrepresented psychosocial risk factors such as immigration, family issues, and financial crisis. These findings suggest that augmentation not only improves model performance but also leads to more structured and interpretable internal representations. --- ### Attention at the Theoretical Minimum: A Mathematics of Arrays Framework for Memory-Optimal Transformer Kernels Authors: Lenore Mullin, Gaetan Hains Published: 2026-06-05 Categories: cs.LG, cs.AI, cs.PF arXiv: https://arxiv.org/abs/2606.07713v1 The attention mechanism is the dominant computational bottleneck in modern transformer-based AI. Its standard implementation incurs quadratic memory traffic in the sequence length~$n$, and DRAM accesses cost 100--1000$\times$ more energy than arithmetic operations on contemporary hardware, so any analysis focused solely on FLOP counts fundamentally mischaracterises the bottleneck. We present a Mathematics of Arrays (MoA) reformulation of scaled dot-product attention and its numerically stable softmax, deriving a Denotational Normal Form (DNF) that eliminates all intermediate arrays -- including the implicit transposed-key buffer and every softmax temporary -- by algebraic construction rather than empirical tuning. The DNF achieves $O(n_{dk} + n{_{dv}})$ data movement versus $O(n^2 + n_{dk} + n_{dv})$ for the standard implementation, where $n$ is the sequence length, $dk$ is the key dimensionality and $dv$ the value dimensionality, and is verified numerically against PyTorch at full double-precision floating-point on concrete inputs. Unlike hardware-specific accelerators or empirical tiling schemes such as FlashAttention, MoA simultaneously provides array fusion, shape-transformation correctness, and predictive cost models from a single algebraic framework. Memory minimality is a theorem established before any code is written. A predictive performance model projects $2$--$100\times$ speedup and $2$--$50\times$ energy reduction, with the advantage widening at exascale. The derivation establishes a formally verified pipeline from Python specification through (ONF) Operational Normal Form, and dimension-lifted hardware mapping, providing performance-portable AI kernels of direct relevance to DARPA edge-deployment and DOE exascale priorities. --- ### MatMind: A Structure-Activity Knowledge-Driven Generative Foundation Model for Materials Science Authors: Zhan'ao Yao, Boxuan Zhang, Jingyuan Shu, Xiaoyu Wu, Rongyan Wang, Linjing Li, Dajun Zeng, Yudong Yao, Tingwei Chen, Youwei Wang, Xiaolin Zhao, Jiahui Shi et al. Published: 2026-06-05 Categories: cond-mat.mtrl-sci, cs.AI arXiv: https://arxiv.org/abs/2606.07712v1 Progress in AI-driven crystal materials science has so far been carried by narrow architectures purpose-built for individual tasks -- graph neural networks for property prediction, diffusion and flow-matching models for crystal generation -- each excelling within its niche yet unable to act as a shared backbone across the full spectrum of materials problems. Generative large language models offer a fundamentally different paradigm, in which structural representation, quantitative prediction, and structure-activity reasoning can be unified within one model, but the materials community has yet to see this paradigm realized at a level competitive with established narrow specialists. Here we present MatMind, a generative foundation model purpose-built for crystal materials science under this paradigm, developed through the coordinated activation of structure-activity knowledge and physics-informed feedback within a progressive training framework -- combining structure-activity knowledge injection, a dual-head architecture that jointly trains language reasoning and numerical regression in a shared representation space, and multi-objective physics-informed reinforcement learning over stability, novelty, and structural diversity. Across three task families, MatMind attains the lowest mean absolute error on energy above hull, bulk modulus, and band gap -- surpassing graph neural network predictors purpose-built for these tasks -- reaches an S.U.N. rate of 65.3% on unconditional crystal generation, and achieves a comparable multiplicative improvement on magnetization-density-conditioned generation, where only 21 positive samples exist within over 600000 training entries. By matching or surpassing narrow specialists on their own ground while operating within a single unified model, MatMind shows that the LLM-based paradigm can serve as a viable backbone for crystal materials science going forward. --- ### Rosetta Memory: Adaptive Memory for Cross-LLM Agents Authors: Hao Yang, Shiqi Shen, Haoxuan Li, Zhipeng Wang, Zhi Gong, Xu Chen Published: 2026-06-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07711v1 Memory is the key component for transforming a stateless LLM into a persistent, evolving agent through experience accumulation, long-horizon planning, and continual self-improvement. Existing memory systems typically take the LLM as the center and design memory operations tailored to a specific backbone. In practice, however, users frequently switch between LLMs, for example using Claude for coding and GPT for writing across tasks, or routing different steps to different backbones within a single task for cost-effective trade-offs. As a result, memory written by one model often needs to be consumed by another. Making upstream memory effectively adapt to and activate downstream LLMs remains a critical yet underexplored problem. To bridge this gap, we shift the perspective from LLM-centric memory design to \emph{memory-centric LLM adaptation}. Specifically, we approach the above upstream-downstream memory adaptation problem from both the write and read sides, and design two profile-conditioned operators that are jointly trained to optimize how memory is stored and presented for better task completion. To ensure the learned operators generalize across a broad set of LLMs, we propose a minimum-gain sampling curriculum that prioritizes the least-served LLMs during training. To better measure the operators' actual contribution rather than the LLM's own capability, we design a performance-gap reward that compares against a naive memory baseline. Experiments on HotpotQA, 2WikiMultihopQA, and MuSiQue demonstrate that our model consistently outperforms baselines and remains robust under unseen-model replacement. --- ### WhiFlash: Accelerating Speculative Decoding with Token-Level Cross-Paradigm Routing Authors: Young D. Kwon, Miles Williams, Rui Li, Alexandros Kouris, Stylianos I. Venieris Published: 2026-06-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07710v2 The autoregressive nature of large language models (LLMs) remains a significant bottleneck for inference, particularly in complex agentic workloads. While speculative decoding (SD) accelerates inference, current approaches rely on static drafting paradigms, utilising either autoregressive drafting models for reasoning or diffusion-based parallel drafting models for structured outputs. We empirically find that drafting accuracy fluctuates dramatically within a single sequence, leaving significant performance unrealised by static paradigms and coarse-grained routing. To address this volatility, we introduce WhiFlash, the first cross-paradigm SD method that unifies autoregressive and diffusion-based parallel drafting under a single token-level controller. WhiFlash adopts a fine-grained routing mechanism that employs either a lightweight entropy-based or a learned neural policy, both parametrised to provide a tunable balance between expected token gain and latency. To make high-frequency switching computationally viable, we introduce novel cache-management optimisations, Lazy Catch-up and KV-only Prefill, reducing switching overhead to below 7% of per-round latency. By capitalising on the complementary strengths of fundamentally distinct drafting architectures, WhiFlash achieves significantly higher acceptance lengths, yielding category-specific throughput gains of up to 69.6% over the state-of-the-art autoregressive EAGLE-3 and 37.3% over the diffusion-based DFlash. --- ### WhiFlash: Accelerating Speculative Decoding with Token-Level Cross-Paradigm Routing Authors: Young D. Kwon, Miles Williams, Rui Li, Alexandros Kouris, Stylianos I. Venieris Published: 2026-06-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07710v1 The autoregressive nature of large language models (LLMs) remains a significant bottleneck for inference, particularly in complex agentic workloads. While speculative decoding (SD) accelerates inference, current approaches rely on static drafting paradigms, utilising either autoregressive drafting models for reasoning or diffusion-based parallel drafting models for structured outputs. We empirically find that drafting accuracy fluctuates dramatically within a single sequence, leaving significant performance unrealised by static paradigms and coarse-grained routing. To address this volatility, we introduce WhiFlash, the first cross-paradigm SD method that unifies autoregressive and diffusion-based parallel drafting under a single token-level controller. WhiFlash adopts a fine-grained routing mechanism that employs either a lightweight entropy-based or a learned neural policy, both parametrised to provide a tunable balance between expected token gain and latency. To make high-frequency switching computationally viable, we introduce novel cache-management optimisations, Lazy Catch-up and KV-only Prefill, reducing switching overhead to below 7% of per-round latency. By capitalising on the complementary strengths of fundamentally distinct drafting architectures, WhiFlash achieves significantly higher acceptance lengths, yielding category-specific throughput gains of up to 69.6% over the state-of-the-art autoregressive EAGLE-3 and 37.3% over the diffusion-based DFlash. --- ### Cross-View Urban Traffic Dataset: Drone-Supervised Ground Truth for Monocular Bird's-Eye View Localization Authors: Prakhar Bhardwaj, Simone Weikl, Kilian Mang, Elia Jonas Sandtner Published: 2026-06-05 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2606.07708v1 We introduce a dataset and benchmark for cross-view urban traffic perception built from synchronized ego-centric bicycle videos and aerial drone videos recorded at real urban intersections. The benchmark targets two linked tasks: cross-view identity matching between street-view and drone-view object tracks, and ego-to-bird's-eye-view prediction using aerial supervision. In contrast to prior urban driving and V2X datasets, our benchmark provides identity-level alignment across radically different viewpoints together with standardized evaluation, annotation tooling, and baseline implementations. This setting is motivated by intersection-centric traffic analysis, where identity preservation, local interactions, and global spatial structure must be reasoned about jointly across views. We evaluate methods at both the track and frame levels, including cross-view ID precision/recall/IDF1, near--far breakdowns, temporal stability, and consistency metrics. We also provide baseline results for wedge-based cross-view matching and for three BEV prediction baselines: inverse perspective mapping, a MonoLayout-style learned baseline, and a regression baseline. The results show that the benchmark is feasible but challenging: cross-view matching achieves strong recall yet remains limited by over-assignment and temporal inconsistency, while ego-to-BEV prediction benefits from aerial supervision but remains far from saturated under lightweight monocular sensing. We hope that this benchmark will support future research on cross-view perception, urban scene alignment, and ego-to-global traffic understanding. --- ### Decoding Naturalistic Emotion Dynamics from the Brain: An LLM-Enhanced Regression Framework Authors: Lemei Zhang, Peng Liu, Hans Dahle Kvadsheim, August Sætre Aasvær, Shuer Ye, Reza Bonyadi, Maryam Ziaei, Jon Atle Gulla Published: 2026-06-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2606.07707v1 Decoding emotional states from neural signals has been typically framed as a discrete, single-label classification task based on emotionally stable stimuli, a formulation that oversimplifies the continuous, fluid, and co-occurring nature of human affect. This study reconceptualizes emotion decoding by adopting a multi-target regression framework to track multiple overlapping emotional dimensions as continuous trajectories over time. Leveraging the robust generalization capabilities of Large Language Models (LLMs), we extracted fine-grained, continuous sentiment profiles from a naturalistic auditory narrative, Alice in Wonderland, to serve as scalable proxies for subjective affect from human fMRI dataset. Departing from standard classification paradigms or mass-univariate subtractive contrasts that filter out network dynamics, we leverage regularized and kernel-based machine learning algorithms as continuous estimators to track the magnitude of macroscale neural state variations. We demonstrate that models trained on temporal snapshots of Dynamic Functional Connectivity (DFC) significantly outperform static region-of-interest (ROI) amplitude representations, effectively capturing continuous emotional trajectories under rapidly fluctuating narrative input. Furthermore, by implementing graph-theoretical Explainable AI (XAI) techniques, we deconstruct the underlying predictive features to reveal highly interpretable, emotion-specific topological configurations. Collectively, these results highlight the utility of LLM-automated annotation in affective neuroscience and provide compelling empirical evidence for psychological constructionist frameworks, demonstrating that dynamic, distributed network interactions offer superior explanatory power over strictly locationist accounts of emotion. --- ### MLingualFC: Evaluating Jailbreak Vulnerabilities in Multilingual Vision-Language Models Authors: Rishabh Makwana, Mamta, Deeksha Varshney, Oana Cocarascu Published: 2026-06-05 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2606.07706v1 Vision-Language Models (VLMs) have demonstrated strong performance across multimodal tasks, yet their safety robustness remains an open challenge. While prior work has shown that structured visual prompts such as flowcharts can effectively jailbreak VLMs, existing studies are largely limited to English-centric settings. In this paper, we introduce MLingualFC, a multilingual multimodal benchmark designed to evaluate jailbreak vulnerabilities of VLMs across diverse languages using structured flowchart representations. MLingualFC encodes harmful instructions into flowchart images across five languages (Hindi, Punjabi, Spanish, Romanian, and German). We evaluate state-of-the-art multilingual VLMs, including Qwen2.5-VL, Gemma-4, and Pangea, under a black-box threat model. Our results reveal significant multilingual safety gaps. Flowchart-based attacks achieve high attack success rates (ASR) in case of Latin script languages, demonstrating that visual encoding of harmful content effectively bypasses safety alignment across languages. In contrast, non-Latin script languages such as Punjabi exhibit substantially lower ASR, suggesting potential limitations in visual text recognition rather than stronger safety alignment. These findings highlight that current VLM safety mechanisms fail to generalize across languages and modalities. Resources are available at https://github.com/Rishabhpm23/MLingualFC --- ### SAW: Stage-Aware Dynamic Weighting for Multi-Objective Reinforcement Learning in Large Language Models Authors: Yuchen He, Baolong Bi, Shenghua Liu, Huaming Liao, Yuyao Ge, Bolin Wan, Siqian Tong, Juan Chen, Jiafeng Guo, Xueqi Cheng Published: 2026-06-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07705v1 Although multi-objective reinforcement learning (MORL) is central to aligning large language models with complex human preferences, the prevailing practice of static weighted summation overlooks a more fundamental phenomenon: reward learning is markedly asynchronous across objectives. Well-learned dimensions quickly produce homogeneous, low-variance signals whose residual noise contaminates the aggregated reward (in GRPO) or occupies a fixed share of the advantage budget (in GDPO), interfering with the scarce yet high-value signals carried by under-learned dimensions. To address this asynchrony, we propose Stage-Aware Dynamic Weighting (SAW), a lightweight, algorithm-agnostic dynamic weighting mechanism. SAW utilizes the coefficient of variation (CV) as a scale-invariant proxy for real-time informativeness, reweighting each dimension's reward or advantage contribution by its relative informativeness within the batch. Unlike gradient-based methods that require multiple forward and backward passes, SAW relies solely on batch-level statistics, introducing nearly negligible computational overhead. Experiments on tool-calling and text summarization tasks demonstrate that SAW consistently improves both training efficiency and final performance under both GRPO and GDPO frameworks, confirming it as a general-purpose plug-in for multi-reward LLM alignment. Our code is available at https://github.com/Zhaolutuan/SAW --- ### FunctionEvolve: Structure-Guided Symbolic Regression with LLMs Authors: Zeyu Xia, Jun Zhu, Dong Yan Published: 2026-06-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07704v1 Symbolic regression aims to uncover explicit scientific laws from data. Recent methods use LLMs to guide mutation from background text, which is more directed than random genetic programming. However, exact symbolic recovery requires both semantic guidance and explicit structure, so that domain-informed search are carried out through valid symbolic representation. Current LLM-driven systems remain structure-blind: they select among opaque candidates, lack explicit mechanisms for local mutation, and rely on brittle coefficient fitting that can undervalue correct skeletons. We propose FunctionEvolve, an evolutionary framework using expression trees to organize the whole search: structural summaries promote diverse parent selection, local tree edits preserve useful subexpressions, and structure-aware fitting decomposes, constrains, and simplifies coefficients for more reliable scoring. It uses only elementary function families, without additional domain-specific rules limiting generalization. On the 129-task synthetic subset of LLM-SRBench, FunctionEvolve with \emph{Claude Opus 4.6} recovers 107 exact forms, reaching 82.9% SA@50, 4.5x above same-backbone baselines, and 55.8% SA@1, 3.6x above the strongest previously published top-1 result. Ablations show that structure-visible search is central to reliable recovery, with LLM-guided refinements and structure-aware coefficient optimization serving as essential proposal and scoring mechanisms. We also audit the benchmark and show that collinearity in its materials-science subset creates identifiability issues. --- ### How Much Dense Attention is Necessary? Oracle-Guided Sparse Prefill for Full/GQA Layers in Hybrid Long-Context Models Authors: Hongxing Wang, Harenome Razanajato, Zhen Zhang, Yujie Yuan, Hongsheng Liu Published: 2026-06-05 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2606.07703v1 Long-context prefill remains expensive because full/GQA layers still score the historical sequence, even in hybrid models with local, sparse, linear, or recurrent components. We study how much dense attention is needed to preserve task-level behavior under explicit support granularity and top-k budgets. We introduce an attention-mass top-k oracle for existing GQA checkpoints: for each layer and query position, it computes dense attention, selects head-averaged token support, and recomputes attention only on that support. The oracle is a diagnostic reference, not a deployable accelerator, and separates sparse-budget feasibility from indexer error and runtime realization effects. On Qwen-family retrieval-heavy evaluations, the longest per-query oracle rows stay within 1 point of dense, and a Qwen3.5-9B RULER-style sweep from 4K to 100K stays within 0.48 points. Guided by the oracle, we derive a head-collapsed auxiliary indexer trained by KL distillation from dense attention-mass distributions while keeping the backbone frozen. With separately distilled Qwen3.5-0.8B and Qwen3.5-9B indexers, the reported 16K/32K validation macro gaps are +2.04 and +1.13 points, treated as quality preservation rather than improvement; fused selection-block-shared support can introduce a larger realization gap. Preliminary single-card TTFT measurements show distilled-indexer sparse serving speedups of 1.71x for Qwen3.5-0.8B on NPU and 1.93x for Qwen3.5-9B on GPU against its dense FlashAttention-2 baseline. Additional random-init stress rows reach 3.44x, indicating sparse-runtime headroom but not validated output quality. This first release separates oracle feasibility, distilled-indexer quality, and runtime headroom, leaving a fully matched quality-latency frontier to future work. --- ### EvoCSFL: Surrogate-Assisted Evolutionary Client Selection for Efficient and Robust Federated Learning Authors: Lin Qiang, Sun Xiaoyan, Hu Yao, Fang Wei Published: 2026-06-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07702v1 The heterogeneity of client data and systems makes it difficult to achieve satisfactory convergence speed and robustness in federated learning with random client selection. To address this issue, this paper proposes a surrogate-assisted client evolutionary selection framework for federated learning. In this framework, some typical client selection strategies are first used to generate candidate sets, and a metric function that integrates model performance, communication latency, and energy consumption is developed to formulate the client selection problem as a combinatorial optimization one. Subsequently, a surrogate model is constructed using the candidate selections and metric to efficiently approximate the performance of selected client subsets. An evolutionary algorithm is employed to search the combinatorial space of client selections, guided by the surrogate model to accelerate convergence. Experiments on MNIST, CIFAR10, CINIC10, and TinyImageNet demonstrate that the proposed algorithm achieves faster convergence, lower energy consumption, and improved robustness compared to existing methods. --- ### EssentialGIN: a new approach for gene essentiality prediction based on graph isomorphism neural networks Authors: Sahar Mansouri-Rad, Zahra Narimani, Parvin Razzaghi, Nazanin Hosseinkhan Published: 2026-06-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07700v2 Background: Prediction of essential genes (proteins), is a basic and challenging problem but at the same time very costly and time-consuming in wet-lab experiments. Predicting essential genes, only based on computational methods (to introduce wet-lab candidates) using centrality measures are not accurate and result in large number of false positives; therefore, more complex models such as deep learning and also integration of biological information are used in recent research to identify essential genes. Methods: In this work we focus on graph isomorphism networks, in order to embed proteins as a node in PPI network to conserve topological features of PPI network, and also integrate biological data such as gene expression data, gene orthology information and gene subcellular localization information, and introduced a deep architecture for predicting essential genes. Graph isomorphism network architecture is modified in this work for embedding node information. Results: Our experiments proved that the proposed method outperforms baseline centrality-based methods and also machine learning based methods such as Node2Vec, MLP, and also graph attention networks (GAT). Conclusion: In this paper we observed that using graph isomorphism networks that integrate biological data (as node attributes) and preserve network topology can significantly improve the essential gene prediction accuracy. In simpler organisms such as E. coli and D. melanogaster, methods such as multi-layer perceptron using Node2Vec embedding also performs very good, but in H. sapiens the introduced architecture significantly outperforms deep learning and other graph neural network solutions. Keywords: Essential gene prediction, graph neural network, graph isomorphism network, PPI network, node embedding --- ### EssentialGIN: a new approach for gene essentiality prediction based on graph isomorphism neural networks Authors: Sahar Mansouri-Rad, Zahra Narimani, Parvin Razzaghi, Nazanin Hosseinkhan Published: 2026-06-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07700v1 Background: Prediction of essential genes (proteins), is a basic and challenging problem but at the same time very costly and time-consuming in wet-lab experiments. Predicting essential genes, only based on computational methods (to introduce wet-lab candidates) using centrality measures are not accurate and result in large number of false positives; therefore, more complex models such as deep learning and also integration of biological information are used in recent research to identify essential genes. Methods: In this work we focus on graph isomorphism networks, in order to embed proteins as a node in PPI network to conserve topological features of PPI network, and also integrate biological data such as gene expression data, gene orthology information and gene subcellular localization information, and introduced a deep architecture for predicting essential genes. Graph isomorphism network architecture is modified in this work for embedding node information. Results: Our experiments proved that the proposed method outperforms baseline centrality-based methods and also machine learning based methods such as Node2Vec, MLP, and also graph attention networks (GAT). Conclusion: In this paper we observed that using graph isomorphism networks that integrate biological data (as node attributes) and preserve network topology can significantly improve the essential gene prediction accuracy. In simpler organisms such as E. coli and D. melanogaster, methods such as multi-layer perceptron using Node2Vec embedding also performs very good, but in H. sapiens the introduced architecture significantly outperforms deep learning and other graph neural network solutions. Keywords: Essential gene prediction, graph neural network, graph isomorphism network, PPI network, node embedding --- ### Pharmacogenomic Knowledge Graph Augmentation for Graph Neural Network-Based Drug-Drug Interaction Prediction Authors: Juergen Dietrich Published: 2026-06-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07698v1 Graph neural networks (GNNs) applied to drug-drug interaction (DDI) prediction rely exclusively on molecular structure encoded as SMILES-derived graphs. Prior work in this series demonstrated that model performance is bounded by the structural information content of training labels -- an Information Ceiling -- that architectural refinements alone cannot overcome. The present study investigates whether pharmacogenomic prior knowledge from the PharmGKB database partially closes this ceiling by providing metabolic pathway context that is independent of, and complementary to, molecular structure. Cytochrome P450 (CYP) enzyme substrate, inhibitor, and inducer annotations for four clinically relevant isoforms (CYP2D6, CYP3A4, CYP2C19, CYP2C9) are extracted and incorporated as a 12-dimensional feature vector concatenated to the molecular embedding prior to interaction prediction. Experiments are conducted under both pair-level and drug-level data splits to quantify generalization to unseen drugs. Results indicate that knowledge graph (KG) augmentation substantially improves DDI type classification under pair-level split conditions (F1-macro: 0.532 vs. 0.241 baseline), while binary interaction detection and drug-level generalization remain bounded by the Information Ceiling (AUC inflation: 0.224 vs. 0.250 baseline). Mechanistic validation on strictly held-out compounds confirms that augmentation preferentially improves CYP2C9-mediated interaction prediction, with probabilities increasing from 0.033-0.117 (baseline) to 0.560-0.586 (KG-augmented). An extension to single-molecule toxicity prediction on the Tox21 benchmark confirms that the effect is contingent on pharmacogenomic annotation coverage. These findings motivate the multimodal framework proposed for the subsequent study in this series. --- ### TianJi-Environ: An Autonomous AI Scientist for Atmospheric Environmental Research Authors: Haoluo Zhao, Hongchun Zhang, Nan Li, Jing-Jia Luo, Kaikai Zhang, Mengyang Yu, Nan Chen, Tao Song, Fan Meng Published: 2026-06-05 Categories: physics.ao-ph, cs.AI arXiv: https://arxiv.org/abs/2606.07697v1 As atmospheric environmental prediction continues to improve, interpretable validation of pollution mechanisms and feedback processes has become a main challenge in atmospheric chemistry. Yet mechanism validation based on complex numerical models still relies heavily on expert knowledge: mechanistic hypotheses must be operationalized into executable experiments, and model outputs must be organized into traceable evidence. We present TianJi-Environ, an auditable AI Scientist for atmospheric-chemistry mechanism validation. TianJi-Environ establishes the first WRF-Chem-based multi-agent framework that autonomously drives complex atmospheric-chemistry simulations, converting mechanistic hypotheses into executable configurations, testing experiments, and evidence criteria. Using ozone response and particulate-matter feedback as two representative examples, we demonstrate TianJi-Environ's capability for mechanism validation. In a summertime ozone case over the North China Plain, the system detects directionally consistent aerosol-radiation-interaction signals in shortwave radiation and boundary-layer height, but judges the evidence for ozone response to NOx control to be incomplete. In a wintertime PM2.5 case over the Guanzhong Basin, it localizes the unsupported link to insufficient propagation from black-carbon perturbation to particulate response and missing diagnostics of vertical absorptive heating. These results show that TianJi-Environ makes expert-driven mechanism validation explicit, structured, and auditable, offering a reproducible paradigm for multi-agent systems coupled with complex atmospheric-chemistry models. --- ### Adversarial Robustness of Activation Steering in Large Language Models Authors: Kien Le, Thai Le Published: 2026-06-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07696v1 Activation steering has become a popular training-free method to control LLM behavior by injecting precomputed direction vectors into the model's residual stream at inference time. Yet its robustness to realistic input variation remains unstudied. We present the first systematic evaluation of activation steering robustness under adversarial text perturbations on the inputs, covering four extraction methods, three attack strategies, six personas from Anthropic Model-Written Evaluation Dataset, and five models ranging from 1.5B to 30B parameters. Attacks succeed broadly across all settings: directional robustness drops by up to 64%, post-attack confidence collapses near or below 0.25 across all methods and models, and steering strength degrades on nearly every steerable input. Layer selection is equally fragile, with the optimal layer identified by an automated method on clean inputs shifting by up to 17 positions under perturbation, a failure that compounds the vector-level breakdown. Extracting vectors from adversarially perturbed inputs partially recovers steerability for PCA and MD on mid-to-large models, but they consistently fail to locate the improved optimal layer, limiting the practical benefit of this mitigation. Together, these findings reveal that the brittleness of activation steering is structural rather than method-specific, and that current layer selection strategies are not robust enough for real-world deployment. --- ### DSFNet: Learning Dual-Domain Spectral Operators for Multi-Modality Spatio-Temporal Forecasting in Urban Transportation Systems Authors: Yongchao Li, Yang Li, Zhuoxuan Li, Jun Chen, Chu Zhang, Jinde Cao, Leszek Rutkowski Published: 2026-06-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07695v1 Multi-Modality Spatio-Temporal Forecasting (MoSTF) extends traditional spatio-temporal forecasting by incorporating diverse traffic modalities. Despite significant recent strides in spatio-temporal modeling, existing approaches often fail to explicitly model the coupling relationships between different modality variables. Accurate MoSTF is challenging, as it requires modeling (1) temporal dynamic heterogeneity under exogenous influences and (2) heterogeneous spatial dependencies alongside complex cross-variable couplings. To address these challenges, we propose the Dual-Domain Spectral Filtering Network (DSFNet). Our framework employs dual-domain spectral filtering to capture heterogeneous spatial patterns and explicitly model the relationships between variables. Unlike graph-based message passing or dense attention over node-modality pairs, DSFNet factorizes space-modality interactions into feature-domain and spatial-domain spectral operators, enabling scalable modeling of nonlocal dependencies and cross-modality couplings. Furthermore, we introduce an external gating mechanism to adaptively regulate temporal dynamics under external influences. We validate our method through extensive experiments on five representative real-world traffic datasets. Compared with the second-best baselines, DSFNet reduces MAE by 3.21%-10.16% across these datasets. The results demonstrate that DSFNet significantly outperforms existing state-of-the-art baselines in accuracy while exhibiting efficiency and robustness. --- ### Vessel Traffic Flow Prediction on Sparse Data via Spatio-Temporal Graph Neural Networks with a Learnable Tweedie Head Authors: Kyeongjun Lee, Heeyoung Kim Published: 2026-06-05 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2606.07694v1 Accurate vessel traffic flow prediction is crucial for smart port operations and navigational safety. However, maritime traffic flow data are often highly sparse with intermittent bursts, making robust forecasting challenging. Under such conditions, conventional spatio-temporal graph neural networks (ST-GNNs) can degrade toward conservative near-zero predictions and fail to capture non-zero activity. Although zero-inflated negative binomial (ZINB) models partially address excess zeros, their two-part formulation can still remain conservative around abrupt transitions. To address these issues, we propose a model-agnostic learnable Tweedie head that can be attached as a plug-and-play output module to arbitrary ST-GNN backbones. Instead of likelihood-based Tweedie training, which typically requires surrogate objectives, our approach optimizes the closed-form Tweedie unit deviance and predicts the mean for point forecasting while learning a node-level variance power to capture heterogeneous variability across port areas. Experiments on a maritime traffic graph constructed from real-world AIS data in the Port of Los Angeles and Long Beach show that the proposed head consistently improves RMSE across multiple ST-GNN backbones, especially on non-zero events, leading to more reliable forecasts for practical maritime traffic control. --- ### Transfer learning for causal forest Authors: Bérénice-Alexia Jocteur, Véronique Maume-Deschamps, Pierre Ribereau Published: 2026-06-05 Categories: stat.ML, cs.LG, math.PR arXiv: https://arxiv.org/abs/2606.07693v1 Transfer learning addresses the challenge of transfering knowledge from one domain to another. Traditional transfer learning focuses on adapting models trained on a source domain (with a lot of observations) to improve performance on a target domain (with few observations). In this work we consider the case of a model shift and we focus on the transfer learning applied to a causal forest namely HTERF. This causal forest aims to estimate the Conditional Average Treatment Effect (CATE). The approach considered is the offset method presented by Wang (2016) adapted to a causal context. This method relies on the use of intermediate models in order to estimate the offset between source and target distributions. Our main result is a bound on the CATE error of HTERF on target depending on the error of the intermediate models. Simulation studies show the good performances of this approach in different settings on simulations and on a real-world dataset. --- ### BCG-FM: A Foundation Model for Ambient Cardiac Health Sensing Authors: Magnus Ruud Kjaer, Haejun Han, Ashish Neupane, David Q. Sun Published: 2026-06-05 Categories: cs.LG, cs.AI, cs.ET arXiv: https://arxiv.org/abs/2606.07692v1 Foundation models for wearable biosignals have matched or exceeded supervised specialists across a range of clinical tasks, yet all rely on modalities that require deliberate user action--wearing a device or visiting a sleep lab. We introduce BCG-FM, the first foundation model for ambient mechanical biosignals. A piezoelectric sensor embedded in the bed surface records ballistocardiography (BCG) each night without user effort; we pretrain BCG-FM with participant-level contrastive learning and using a total of 2.75 million hours of nightly recordings from 145,985 individuals, the largest raw-waveform biosignal pretraining corpus to date. Frozen BCG-FM embeddings achieve 3.26-year MAE on biological-age estimation (the lowest reported for any ambient, contactless modality) and yield clinically relevant discrimination across 15 self-reported health conditions and three independent external cohorts. Pretrained representations from only 500 labeled participants outperform a fully supervised baseline trained on 3,372, and representation quality scales log-linearly with contrastive batch size. These results establish ambient, longitudinal mechanical biosignals as a viable modality for health foundation models. --- ### HARP: Efficient Data Selection for Finetuning Large Language Models Authors: Ning Wang, Zhengxin Zhang, Maosen Tang, Yitang Gao, Claire Cardie, Sainyam Galhotra Published: 2026-06-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07690v1 Finetuning data selection requires balancing two competing goals: selecting examples that improve the downstream objective, and doing so without repeatedly finetuning models. Train-free selectors are scalable but rely on proxies such as embedding similarity or clustering, which may not match the target objective. Train-based selectors better reflect downstream utility through gradient signals, subset evaluation, or Shapley attribution, but require many costly train--evaluate iterations. We propose Hierarchical Active Region Pruning (HARP), an efficient train-based selector that preserves downstream alignment while reducing selection cost. HARP organizes the training pool into a node--leaf hierarchy, evaluates only representative leaves, and infers unmeasured utilities with empirical Bayes posteriors. It then selects data using two complementary envelopes: HARP-C, which conservatively controls redundancy, and HARP-E, which additively rewards complementary regions. We theoretically show that, under local smoothness and bounded estimation error, HARP controls selection error while reducing train--evaluate cost. We further validate that HARP variants achieve the best result and outperform the strongest baseline by up to $+8.9$ points, while using roughly $7\times$ fewer training examples. --- ### Struct-Searcher: Agentic Structural Thinking Advances Multimodal Deep Information Seeking Authors: Fan Zhang, Vireo Zhang, Shengju Qian, Haoxuan Li, Zheng Lian, Hao Wu, Yuan Gao, Xinyu Geng, Xin Wang, Pheng-Ann Heng Published: 2026-06-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.07689v1 Deep research agents have attracted increasing attention for their ability to collect large-scale online information to acquire target knowledge, with recent efforts shifting from purely text-based information seeking to multimodal settings. However, existing agentic workflows are largely aligned with evidence accumulation models, which linearly aggregate evidence and lack principled mechanisms for handling contradictory information across heterogeneous modalities. Towards this end, we propose Struct-Searcher, a structural agentic workflow grounded in belief revision theory that explicitly maintains an evolving multimodal structural graph throughout the reasoning process, enabling effective conflict-aware multimodal deep information seeking. Extensive experiments across multiple benchmark datasets and backbone models demonstrate that Struct-Searcher is (1) plug-and-play and model-agnostic, yielding an average relative accuracy improvement of 17.2% on BrowseComp-VL across five different backbones. (2) top-performing, consistently outperforming state-of-the-art vision-language models (VLMs) and deep research agents, with relative accuracy improvements of 3.7% on MM-BrowseComp, 1.5% on HLE-VL, and 0.7% on BrowseComp-VL over the second-best competing approach. --- ### TRACER: Token ReAssignment for Concept ERasure in Generative Recommendation Authors: Ziheng Chen, Jiali Cheng, Zezhong Fan, Hadi Amiri, Diyuan Wu, Gabriele Tolomei, Yang Zhang Published: 2026-06-05 Categories: cs.IR, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2606.07688v1 Generative recommendation formulates next-item prediction as autoregressive generation over semantic ID (SID) sequences derived from users' historical interactions, making modern recommender systems structurally similar to large language models (LLMs). As privacy and safety concerns grow, these systems increasingly require concept unlearning to remove sensitive or harmful concepts associated with items. However, existing LLM unlearning methods cannot be directly applied to generative recommendation. Unlike word tokens with explicit semantics, SIDs are abstract identifiers that are often shared by both forget and retain items, leading to severe conflicts between concept removal and recommendation utility preservation. To address this challenge, we propose TRACER, an end-to-end concept unlearning framework based on token reassignment. Rather than directly suppressing shared SIDs, TRACER reassigns concept-related items to alternative tokens that better facilitate forgetting while minimizing side effects on retained items. We further introduce a coherence regularizer to preserve semantic consistency among retain items during unlearning. Experiments on real-world recommendation datasets demonstrate that TRACER effectively removes target concepts while substantially better preserving recommendation utility than existing unlearning baselines. --- ### What Makes Video World Model Latents Action-Relevant: Prediction over Reconstruction Authors: Jewon Yeom, Hanseul Kim, Jeongjae Park, Sungmok Jung, Jaejin Lee, Taesup Kim Published: 2026-06-05 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2606.07687v1 Video world models are increasingly used to provide predictive visual representations, yet it remains unclear which pretraining signals induce action-relevant structure in their latent spaces. We study this question through a unified probe-based evaluation across diverse encoder families, including image-only self-supervision, video pretraining with and without latent prediction, reconstruction-based autoencoders, diffusion models, and shortcut-forcing dynamics models. Using a common inverse-dynamics probing objective, we find that action-relevant structure is driven primarily by temporal video pretraining rather than pixel reconstruction fidelity: models with strong pixel decoding quality can exhibit near-zero action recoverability, while video-pretrained self-supervised encoders consistently achieve the best Pareto trade-off between visual fidelity and action prediction. Comparing V-JEPA and VideoMAE further shows that most gains arise from natural-video temporal context, with feature-level latent prediction providing a smaller additional benefit. These trends transfer across robotic benchmarks, though CALVIN reveals that static-environment tasks can partially mask the importance of temporal structure by allowing strong image priors to suffice. Finally, inverse-dynamics supervision substantially improves robustness to visual corruption, suggesting that action-aware objectives regularize latent geometry beyond clean-setting performance. Our results identify temporal predictive structure -- not reconstruction fidelity -- as the primary ingredient underlying action-relevant video representations. --- ### Knowledge-Inclusive Adaptive Physics-Informed Neural Network for Microbial Interaction Modelling Authors: Ravisha Rupasinghe, Rajith Vidanaarachchi, Asela Hevapathige, Sachith Seneviratne, Sen-Lin Tang, Saman Halgamuge Published: 2026-06-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07686v1 Physics-Informed Neural Network (PINN) is a way of including knowledge in the form of equations in Machine Learning methods. Beyond equations, knowledge exists in other forms, such as text and network structure. While existing PINN-based approaches discover equation parameters from data, they rely solely on experimental measurements. We propose a new PINN framework that enriches parameter discovery by incorporating auxiliary knowledge sources. We instantiate our framework for microbiology, where generalised Lotka-Volterra (gLV) serves as a biological foundation for modelling microbial communities. We demonstrate that incorporating knowledge improves microbial community modelling. Our framework enriches the gLV parameters using peer-reviewed metagenomics literature, as text provides biological context on external influences that gLV alone cannot capture. We combine this knowledge with experimental measurements of microbial abundance using a data-driven integration approach. We integrate network-based structural knowledge by explicitly modelling microbial interactions. Our knowledge-inclusive framework infers microbial networks, revealing ecological insights. We validate these findings against ecological roles documented in the literature. We evaluate on real and simulated datasets spanning human- and plant-associated microbial communities. Our framework improves over the state-of-the-art by up to 53%, even without knowledge. Knowledge addition yields gains of up to 23% in Bray-Curtis Dissimilarity-based accuracy and 47% in $\mathrm{R}^2$. --- ### Test-Time Adaptive Composition for Machine Learning as a Service (MLaaS) in IoT Environments Authors: Deepak Kanneganti, Sajib Mistry, Sheik Mohammad Mostakim Fattah, Aneesh Krishna Published: 2026-06-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07685v1 The dynamic nature of Internet of Things (IoT) environments affects the long-term effectiveness of Machine Learning as a Service (MLaaS) compositions. Existing adaptive composition methods are mainly based on service replacement or re-composition, where identifying suitable substitutes is difficult and time-consuming. To address this, we propose a novel Test-Time Adaptive (TTA) composition framework for MLaaS in IoT environments. First, we introduce a TTA-aware composability model to determine whether adapted services remain compatible with the existing composition. Next, we design a service-level adaptation model to adjust individual services during inference while preserving composition performance. Experimental results demonstrate that the proposed framework reduces computational time more effectively than traditional adaptive approaches. ---