MeshGraphNet-Transformer (MGN‑T) is a new architecture that fuses Transformers’ global modeling with MeshGraphNets’ geometric inductive bias, keeping a mesh‑based graph representation. It replaces iterative message passing with a physics‑attention Transformer that updates all nodal states simultaneously, enabling efficient learning on high‑resolution meshes with diverse geometries, topologies, and boundary conditions. MGN‑T accurately models impact dynamics, self‑contact, plasticity, and multivariate outputs, outperforming state‑of‑the‑art methods on classical benchmarks while using far fewer parameters.
By Mikel M. Iparraguirre, Iciar Alfaro, David Gonzalez, Elias Cueto
The paper introduces Geometric Anchor Preference Optimization (GAPO), a method that replaces the static reference policy in Direct Preference Optimization with a dynamic, geometry-aware anchor—a small adversarial perturbation of the current policy. GAPO uses this anchor to adaptively reweight preference pairs based on local sensitivity, and defines an Anchor Gap that approximates worst‑case local margin degradation. Experiments show that GAPO improves robustness to noisy supervision while matching or surpassing existing LLM alignment and reasoning benchmarks.
By Youngjae Cho, Jongsuk Kim, Ji-Hoon Kim
The paper introduces Supervised Deep Multimodal Matrix Factorization (SD3MF), an interpretable framework that extends Symmetric Nonnegative Matrix Tri-Factorization to supervised prediction across populations of multimodal brain graphs. SD3MF learns deep hierarchical factorizations for each modality and a shared latent representation, jointly optimizing graph reconstruction and prediction while enabling data-driven multimodal fusion. Experiments on multimodal connectome datasets demonstrate that SD3MF outperforms strong deep learning baselines such as CNNs and GNNs, providing biologically interpretable insights through community-level interaction matrices.
By Amjad Seyedi, Lifang He, Songlin Zhao, Akwum Onwunta, Nicolas Gillis
The paper introduces MSAlign, a lightweight model that aligns frozen foundation models for mass spectra (DreaMS) and molecules (MolDeBERTa) to improve metabolite identification from MS/MS spectra. It presents a unified framework for representation alignment and contrastive learning, demonstrates that a score fusion strategy further boosts performance at minimal cost, and addresses evaluation challenges by quantifying distribution shift in data splitting strategies. All resources, including datasets, splits, and code, are publicly released to promote reproducible research.
By Paul Krzakala, Gabriel Melo, Camille Lan\c{c}on, Charlotte Laclau, R\'emi Flamary, Etienne Th\'evenot, Florence d'Alch\'e-Buc
The paper introduces a new framework for creating frequency‑based adversarial attacks that are grounded in an explicit optimization problem. By defining a perturbation constraint set linked to structured, non‑orthogonal transforms, the authors show that attacks can be generated as weighted σ‒projections onto this set, providing a clear geometric characterization. Experiments on standard datasets demonstrate that these attacks are highly effective across both pretrained and robust models, even on unseen architectures.
By Vicky Kouni, Stelios Perrakis, Francis Bach, Pascal Frossard, Yann Chevaleyre
AYLA is a loss reparameterization framework that applies a sigmoid‑controlled power‑law transformation to the empirical loss, dynamically adjusting gradient magnitudes without changing stationary points or optimal solutions. By reshaping optimization trajectories, AYLA accelerates descent in flat or saddle‑dominated regions and stabilizes late‑stage training, leading to improved feature recovery in two‑layer tanh networks on synthetic Gaussian data. Experiments show enhanced weight alignment, neuron similarity, activation correlation, and richer internal representations, while mitigating rank collapse and promoting a transition from lazy to active feature‑learning regimes.
By Behnam Gheshlaghi, Shahin Atakishiyev
Historical street-view imagery records urban evolution, but uneven coverage leaves substantial gaps in historical records. Generating plausible past appearances requires restoring changed structures w...
A commercial incentive need not enter the final ranking algorithm to affect a shopping assistant's recommendation: it may instead influence which preference question the assistant asks. We make this d...
The article presents early guidelines for safety cases in frontier AI training, outlining technical safeguards, operational practices, and methods for investigating misalignment incidents. It emphasizes the importance of structured safety documentation to guide the development and deployment of advanced AI systems. The guidelines aim to provide a framework for identifying and mitigating risks associated with frontier AI training.
SEABench is a benchmark designed to study endogenous misalignment in self‑evolving large language model agents. It contains 48 longitudinal task sequences across various evolution surfaces, task domains, and harm types, and includes an adaptive trajectory discovery pipeline that probes for failures while preserving task intent. Evaluations show that self‑evolution improves task completion rates but often introduces safety failures absent in non‑evolving baselines, with divergent safety behaviors reflected in agents’ chain‑of‑thought reasoning that can be monitored to mitigate unsafe actions.
The paper investigates how hybrid attention mechanisms—combining full softmax attention with recurrent alternatives—affect multilingual language models, especially for long sequences and poorly tokenized languages. Interpretability analysis reveals that cross‑lingual representations form patterns linked to the ordering of recurrent and full‑attention layers, with a notable spike in alignment after the first full‑attention layer. Distillation experiments show that alternative layer orderings consistently outperform the standard arrangement, achieving up to 2.5× faster learning, suggesting that starting with a full‑attention layer may benefit multilingual models.
The paper introduces P2P‑T, a data‑efficient, object‑centric framework that learns tool manipulation directly from human video demonstrations. It uses a two‑stage approach: first pretraining an object‑centric world model to extract stable pose priors, then integrating these priors into a pose‑aware low‑level policy. By automating data processing with foundation models, P2P‑T eliminates the need for human‑robot aligned data and achieves a 73% improvement over prior state‑of‑the‑art performance on complex real‑world tool manipulation tasks.
The paper investigates the Reverse Sequential Proportional Approval Voting Rule (RevSeqPAV) used in approval-based committee elections. It examines the rule’s performance in terms of proportional representation—specifically Extended Justified Representation and related metrics—and its ability to approximate the maximum PAV score. The authors first present strong negative results for general election instances, then identify specific settings where RevSeqPAV offers meaningful fairness and optimization guarantees.
The paper presents an expert-driven method for turning normative principles—specifically Islamic ethical, theological, and jurisprudential traditions—into alignment data for language models. Over a year, seven experts curated 2.8K supervised fine-tuning examples and 5.4K preference pairs in Arabic-English, then evaluated models trained on these datasets. Experiments show that models trained with the curated SFT data outperform a baseline in expert judgments, while adding preference data yields a smaller, non-significant improvement.
The study investigates how well a single consumer earbud IMU can estimate 3D body pose and whether adding foot IMUs improves accuracy. Using a multimodal capture pipeline with RGB‑D video, an AirPods head IMU, and Striv insole IMUs, the authors benchmark pose estimation across various motions and train recurrent models (IMUPoser and MobilePoser). Results show that a head IMU alone achieves 79.0 mm rigid‑MPJPE and 0.809 macro‑F1 for foot contact, while adding foot IMUs does not significantly improve pose and can even degrade performance due to insole orientation errors.
The article explains how to detect subtle changes in the relationships between features—known as data drift—using adversarial validation techniques and scikit‑learn tools. It focuses on situations where individual feature distributions appear unchanged, yet the underlying interactions shift, potentially degrading model performance. The post provides practical guidance on implementing these methods to maintain model reliability.
By Benjamin Nweke
EOPSA (Efficient On-Policy Self-Distilled Safety Alignment) addresses inefficiencies in On-Policy Self-Distillation (OPSD) for safety alignment by focusing training on safety-critical tokens. It introduces Adaptive Rollout Scheduling, which limits generation length based on a Teacher Rescue Rate metric, and Selective Distillation, which filters out safety-neutral tokens to concentrate gradient updates on safety-pivotal transitions. Experiments on models up to 32B parameters show that EOPSA reduces rollout computation by about 50% and backpropagates through only roughly 2% of tokens, outperforming full-token distillation baselines in safety compliance and reasoning retention.
The paper examines whether model uncertainty aligns with human disagreement on vision tasks. Using multi‑annotator datasets (FER+ and CIFAR‑10H), the authors find that pretrained models rarely reflect the ambiguity humans perceive, with weak correlations between model confidence and human disagreement. Predictive multiplicity offers only modest improvement, indicating that common uncertainty metrics fail to flag ambiguous cases.
RGDT-Bench is a new benchmark that evaluates large language models on Rule‑Governed Decision Tasks, where models must apply external rules to facts, justify decisions, and provide checkable justifications. The benchmark offers 202.1K condition‑level supervision slots across four task tracks and eight task‑probe combinations, and it labels warrant completeness through label‑blind extraction and deterministic checks. Evaluation shows that among correct responses, 40.2% of warrants are incomplete, and existing evaluators struggle to detect this, prompting the authors to train a reward model that improves AUROC to 69.24% and outperforms outcome‑supervised baselines.
The paper introduces Q-learning Penalized Transformer (QPT), a training–inference consistent framework for safe offline reinforcement learning. QPT trains a Transformer policy that generates actions conditioned on trajectory context and target return/cost while incorporating a Q-shaped penalty to balance safety, reward maximization, and behavior regularization. The method consistently outperforms strong baselines on 38 DSRL benchmark tasks and adapts robustly to varying constraint thresholds.