The paper introduces DMFL-SQ, a decentralized multi-task learning algorithm that integrates graph-based personalization, agnostic fairness, and compressed event-triggered communication. It provides convergence guarantees for non-convex objectives, achieving an ≠O(T^{-1/2}) stationarity rate despite sparse, quantized, and event-triggered communication, and offers PAC-Bayes generalization bounds for the fairness objective. Experiments on CIFAR-10 and the MUSMET EEG dataset show that DMFL-SQ reduces communication while preserving predictive performance and improving fairness across clients.
By Krishnendu S. Tharakan, Carlo Fischione
The paper introduces a new approach to safety in contextual bandits with continuous actions, focusing on high‑probability constraints on the realized cost rather than expected cost. It presents the High‑Probability Constrained UCB algorithm, which balances reward exploration with conservative safety estimation, and provides theoretical regret guarantees for linear models and extensions to general function classes. Experiments demonstrate that this realized‑cost safety framework significantly reduces safety violations compared to expected‑cost constrained methods.
By Spyros Dragazis, Aldo Pacchiano
The paper evaluates large language models (LLMs) for annotating German parliamentary debates on migration, achieving macro‑F1 scores comparable to human agreement, particularly with GPT‑5 and gpt‑oss‑120B. It combines soft‑label outputs with Design‑based Supervised Learning to mitigate systematic bias and applies the method to a 150‑plus‑year corpus, revealing high solidarity post‑war and a sharp rise in anti‑solidarity since 2015. The study demonstrates that LLMs can enable large‑scale social‑scientific analysis while highlighting the need for rigorous validation and bias correction.
By Aida Kostikova, Ole P\"utz, Steffen Eger, Olga Sabelfeld, Benjamin Paassen
The paper introduces Reward‑Informed Sparse Autoencoders (RI‑SAEs), which use reinforcement‑learning rewards to curate data for training sparse autoencoders on language‑model activations. On Llama‑3.1‑8B, a sparse subset of features separates high‑reward from low‑reward reasoning continuations, but control experiments show this separation largely reflects solution completeness rather than true reasoning quality. The authors conclude that reward filtering can cheaply reuse RL signals for interpretability, though most of the discovered features capture completion form rather than deep reasoning.
By Tanvi Nagilla, Alexander Jameson, Daniel Manta, Shayaan Uddin
The paper introduces a method to embed a controllable latent variable into natural-looking text by steering a teacher LLM along eight sparse autoencoder directions that follow a ring-shaped Markov chain. A small transformer trained on this data successfully tracks the Bayesian posterior of the planted variable and arranges the eight states in the same ring order, linking belief states to concept geometry. This demonstrates that LLMs can model latent variables and that concept geometry may arise from the statistical dynamics of these variables.
By Alexandru-Iulius Jerpelea
The paper demonstrates that the multilingual voice cloning model XTTSv2 can be repurposed for speaker anonymization without retraining. By conditioning on a pseudo-speaker and using an iterative refinement strategy, the authors balance privacy and intelligibility, achieving near‑optimal privacy (EER ≈ 0.49) and competitive speech quality across seven European languages. The method outperforms dedicated anonymization baselines and requires no language‑specific training.
By Romolo Muletta, Felix Matthias Saaro, Mark Cieliebak, Jan Deriu
SkillChain is a system designed to automate the lifecycle of skills for image-based AI assistants on e-commerce platforms. It operates in three stages—Skill Creator, Route Optimizer, and Body Refiner—to bootstrap skills from task specifications, align routing, and iteratively refine skill bodies using dual-path LLM-Judge evaluation. In production, SkillChain improves response quality, especially in structural compliance and content quality, and an online A/B test shows increased user engagement, content consumption, and long‑term retention.
By Yimin Hu, Mengtao Xu, Hao Guo, Yuheng Song, Xiaoyong Zhu, Bo Zheng
The paper presents a deep learning approach for detecting woody clearing using bitemporal Sentinel‑2 imagery from New South Wales, Australia. By introducing a loss‑scaling coefficient, the authors align the model’s objective with end‑user metrics, boosting precision and recall. They further demonstrate zero‑shot transfer to woody regrowth and segmentation tasks, achieving significant error reductions and high F1 scores through image augmentation and generation techniques.
By Kal Backman, Jared Wood, Adam Roff
The paper introduces FAN-LoRA, a Fourier‑adaptive nonlinear low‑rank adaptor designed to improve domain adaptation of vision foundation models like SAM for medical imaging. By decoupling frequency components into a low‑pass B‑spline branch for global structure and a high‑pass Fourier branch for local texture, FAN‑LoRA addresses performance drops caused by domain shifts. Experiments on three cross‑modality and cross‑center benchmarks show that FAN‑LoRA outperforms existing PEFT methods, achieving higher Dice scores and lower boundary errors while remaining computationally efficient.
By Ziquan Liu, Zhewei Zhu, Xuyang Shi
The paper introduces the Multi-Instruction Multi-Shot Long-Video Editing (MMLVE) task, aiming to enable consistent editing of long videos with multiple instructions. It proposes an agentic editing framework that combines Large Language Models and Vision-Language Models for shot-level decoupling and precise instruction parsing. The authors also present the MMLVE-Bench dataset and evaluation metrics, showing that their MMLVE-Agent outperforms existing state‑of‑the‑art methods by eliminating hallucinations and preserving temporal consistency.
By Chenyang Wu, Fuchen Long, Binyuan Huang, Xinlong Sun, Xi Chen, Chun-Le Guo, Chongyi Li
RubricRM introduces a pairwise generative reward modeling framework that generates an input‑specific rubric—comprising evaluation dimensions, weights, and scoring criteria—to score candidate images. The method is trained in two stages: supervised fine‑tuning to learn the rubric‑based scoring paradigm and GRPO to refine dimension‑level rewards. Experiments on text‑to‑image generation and instruction‑based image editing benchmarks demonstrate that RubricRM outperforms existing specialized reward models and competes with strong proprietary MLLM judges while using smaller backbones.
By Zijian Kan, Wei Wang, Long Luo, Bing Zhao, Xuan Ren, Weixu Qiao, Wenbo Li, Hu Wei, Lin Qu
The paper tackles the problem of calibrating a stationary camera’s yaw, pitch, and roll using a drone’s GPS trajectory, which suffers from altitude bias, time offset, and discrete sampling. It formulates a parameter estimation problem that jointly estimates the GPS altitude bias, time offset, and camera orientation biases, and proposes a maximum likelihood estimator based on Iterated Least Squares to handle the asynchronous, discrete GPS data. Simulation results show the estimator achieves accuracy close to the Cramér–Rao Lower Bound, with a recommended drone trajectory yielding calibration errors within 14% of the measurement error standard deviation.
By R. Yang, Y. Bar-Shalom, H. A. J. Huang
The paper presents a method for accurately measuring 3D and 2D circular centers, a key geometric sensing problem for instruments like cameras and LiDARs. It integrates a conformal‑geometric‑algebra estimator with RANSAC to jointly recover the 3D center, normal, and radius from noisy LiDAR points, and uses a chord‑length‑variance criterion to estimate the 2D projected center, resolving ambiguity with homography validation or a quasi‑RANSAC fallback. Experiments on synthetic and real sensors demonstrate improved center measurement accuracy and reduced extrinsic calibration error across various target and sensor setups.
By Jiajun Jiang, Xiao Hu, Wancheng Liu, Wei Jiang
The paper introduces Disjoint Parameter Training (DPT), a framework that addresses Skill Conflict—where shared encoder parameters hinder separate tasks of motion prediction and safety planning—by training tasks on distinct parameter subsets before merging. DPT employs sparse merging to integrate only the most influential parameters, reducing interference and enhancing representational capacity. Experiments on JRDB and JTA benchmarks show that DPT outperforms existing unified models, demonstrating its effectiveness for safe, resource‑efficient robot navigation.
By Taewon Seo, Seonae Jeon, Giwon Lee, Kuk-Jin Yoon, Daehee Park
The paper investigates federated adversarial training (AT) for vision transformers, a topic not previously explored in federated learning (FL). It evaluates various transformer architectures and aggregation strategies, and introduces FedWAvg, an extension of FedAvg that weights client updates based on similarity of their last-layer representations. Experiments demonstrate that FedWAvg yields higher robust accuracy than existing aggregation methods in non‑IID settings.
By Ahmed Aldahdooh, Wassim Hamidouche, Olivier D\'eforges
The paper investigates whether large language models (LLMs) can construct knowledge graphs (KGs) from documents in a zero‑shot, schema‑free manner. It introduces the Detail‑to‑Abstract Hierarchical Knowledge Graph (D2A‑HKG) framework, which splits KG construction into extraction, splitting, and abstraction stages, and evaluates the resulting graphs semantically and structurally. Using seven leading LLMs, the authors benchmark zero‑shot KG construction on CMW‑Lit—a dataset of pediatric mental‑wellness research—and find that state‑of‑the‑art LLMs generally produce relevant, document‑faithful triples with limited hallucination, though their extraction behaviors vary across stages. The study releases CMW‑Lit and the generated graphs as resources for future research and downstream knowledge‑intensive applications.
By Ruirui Chen, Weifeng Jiang, Chengwei Qin, Bo Xiong, Kaiwen Wei, Fiona Liausvia, Pei Fang Tan, Ker Yung Chua, Dongkyu Choi, Mukkesh Kumar, Evelyn C. Law, Dennis Wang, Boon Kiat Quek
The paper proposes a new framework for automated fault diagnostics in modern vehicles by treating diagnostic trouble codes (DTCs) as a high‑dimensional language. It introduces Transformer‑based models for predictive maintenance, scalable causal discovery methods, and a multi‑agent system that automatically generates Boolean error‑pattern rules. The approach aims to replace costly manual grouping of DTCs with scalable, data‑driven techniques.
By Hugo Math
The authors present the Cross-Platform Fairness Evaluation (CPFE) framework, a five‑axis audit protocol that assesses discriminative performance, calibration, statistical significance, prediction equity, and attribution stability of transformer models. Applying CPFE to four models trained on a Kaggle mental‑health corpus and tested on Reddit and Twitter, they find substantial cross‑platform degradation in AUC (30–40%) and severe calibration failures (ECE rising to 0.5 on Twitter). The study demonstrates that platform‑specific temperature scaling can largely fix calibration without harming discrimination, while prediction equity and attribution stability analyses reveal significant disparities and vocabulary divergence across platforms. The results argue that cross‑platform validation across all CPFE axes should become a standard requirement for mental‑health NLP systems deployed in heterogeneous environments.
By Rajveer Singh Pall, Sameer Yadav
CGS‑SLAM is a hybrid decentralized/centralized SLAM system that enables multi‑agent 3D Gaussian Splatting reconstruction using only RGB images and inertial data. Each agent locally tracks motion with inertial priors, builds a scaled map via a monocular depth estimator, and shares keyframe encodings to facilitate dynamic keyframing and submap alignment. A central server then aligns submaps with a view‑alignment model, keeping communication costs low while achieving competitive tracking, higher rendering quality, and accurate submap alignment in GNSS‑denied environments.
By Jean-Daniel de Ambrogi, Aladine Chetouani, Vincent Nguyen, Aur\'elien Chateigner
Geo-LoRA introduces a geometry‑aware framework for low‑rank adaptation in rehearsal‑free class‑incremental learning. It regulates the evolution of shared and task‑specific LoRA subspaces using Subspace Projection Preservation, Adaptive Core‑Slack Alignment, and Median‑Calibrated Block Overlap, ensuring smooth trajectories on the Grassmann manifold and balanced stability‑plasticity trade‑offs. The method achieves state‑of‑the‑art performance across multiple benchmarks without adding new adapter types.
By Yibo Feng