The paper introduces a method for constructing L-Lipschitz deep residual networks (ResNets) using a Linear Matrix Inequality (LMI) framework. By reformulating the ResNet architecture as a pseudo-tridiagonal LMI and applying the Gershgorin circle theorem, the authors derive closed‑form constraints on network parameters that guarantee Lipschitz continuity. The work also presents a compositional framework for handling recursive systems in hierarchical architectures, while noting that the Gershgorin-based approximations can over‑constrain the system, reducing expressive capacity.
By Marius F. R. Juston, William R. Norris, Dustin Nottage, Ahmet Soylemezoglu
The paper discusses tensorizing neural networks by reshaping dense weight matrices into higher-order tensors and approximating them with low-rank tensor network decompositions. This approach offers promising model compression and introduces bond indices that create new latent spaces, potentially enhancing interpretability. Despite encouraging empirical results, tensorized neural networks remain underused, and the authors call for more research to address practical scaling and adoption challenges.
By Safa Hamreras, Sukhbinder Singh, Rom\'an Or\'us
The paper argues that diffusion-based action policies can use a frozen, observation‑free backbone as a reusable trajectory prior, with task adaptation handled entirely by the conditioning pathway. By pretraining a general action head on forward‑kinematics data and then freezing it, the authors show that a single backbone can match or outperform normally trained models on MimicGen and LIBERO. Their experiments reveal that a small 5 M‑parameter MLP backbone can rival large U‑Net and transformer backbones, indicating that action backbones are often over‑parameterized and that image‑style architectures may not be the best fit for low‑dimensional action generation.
By Jian Zhou, Sihao Lin, Shuai Fu, Zerui Li, Gengze Zhou, Qi WU
DuoTok is a source‑aware dual‑track music tokenizer designed for vocal‑accompaniment generation. It first learns a semantic audio representation via self‑supervised pretraining, then refines source‑aware structure with feature‑replacement noise and multi‑task supervision (spectral reconstruction, source separation regularization, and an ASR head for lyric alignment). The encoder is frozen and hard‑routed codebooks for vocals and accompaniment are learned, while a diffusion decoder restores fine acoustic detail from the discrete tokens, achieving a favorable predictability‑fidelity trade‑off at ultra‑low bitrate across public benchmarks.
By Rui Lin, Zhiyue Wu, Jiahe Lei, Kangdi Wang, Weixiong Chen, Junyu Dai, Tao Jiang
The paper introduces Semantic Confusion to assess how consistently large language models refuse similar prompts. It presents ParaGuard, a 10k‑prompt corpus of controlled paraphrase clusters, and proposes three token‑level metrics—Confusion Index, Confusion Rate, and Confusion Depth—to measure contradictory refusal decisions across meaning‑preserving paraphrases. Experiments show that global false rejection rates can mask local inconsistencies, revealing that refusal evaluation must consider both frequency and consistency across nearby paraphrases.
By Riad Ahmed Anonto, Md Labid Al Nahiyan, Md Tanvir Hassan
The paper introduces an enhanced end‑to‑end circuit analysis framework built on Gemini 2.5 Pro, targeting engineering education. It addresses two key failure modes—circuit‑recognition hallucinations and reasoning‑process hallucinations—by adding a YOLO detector for source polarity re‑identification and an ngspice verification loop for iterative refinement. The resulting pipeline achieves 97.59 % accuracy on 83 undergraduate problems, markedly outperforming the baseline Gemini model and demonstrating significant gains across varied diagram styles and textbooks.
By Liangliang Chen, Weiyu Sun, Huiru Xie, Yongnuo Cai, Ying Zhang
The paper introduces Dynamic Semantic Steering (DSS), a training‑free, inference‑time defense for robust concept erasure in text‑to‑image diffusion models. DSS models local semantic neighborhoods geometrically, automatically identifies benign semantic anchors, and applies context‑aware, constrained feature correction using cross‑attention signals. Experiments show DSS achieves an average erasure rate of 91.0%, outperforming prior defenses while reducing semantic drift and preserving generation fidelity.
By Qinghui Gong, Zhengchun Zhou, Hua Meng, Yihuai Liang, Yuxuan Zhang
EventVL introduces the first generative event-based multimodal large language model (MLLM) designed for explicit semantic understanding of event streams. The framework leverages a newly annotated dataset of nearly 1.4 million event–image/video–text pairs and incorporates an Event Spatiotemporal Representation to capture comprehensive event information, along with Dynamic Semantic Alignment to refine sparse semantic spaces. Experiments demonstrate that EventVL outperforms existing MLLM baselines in event captioning and scene description generation tasks, advancing the field of event vision.
By Pengteng Li, Yunfan Lu, Pinghao Song, Wuyang Li, Huizai Yao, Hui Xiong
arXiv:2609. 14959v1 Announce Type: new Abstract: We study decentralized learning of Nash equilibria (NE) in infinite-horizon discounted Markov games under bandit feedback, focusing on Markov $\alpha$-potential games.
By S. Rasoul Etesami
The paper proposes a privacy‑aligned personalized federated learning method that releases a private client context once and limits repeated adaptation to a fixed coefficient space, thereby reducing dimensionality misalignment. A factorized generator creates an adaptive optimization geometry that reshapes noisy updates, and most of the private‑training benefit is preserved by radial evolution. Variable‑length Gaussian quantization is used for coefficient updates, allowing the quantization error to act as the privacy perturbation and cutting protected uplink communication by a factor of 2.67 on CIFAR‑10 at ε=16 while maintaining comparable future‑client accuracy.
By Yilin Xu, Chun Hei Michael Shiu, Chih Wei Ling, Linqi Song
The paper investigates the bias in variable importance scores produced by tree-based methods, noting that continuous predictors are favored over categorical ones. It offers a theoretical explanation for this bias and proposes a straightforward fix: adding a small amount of noise to each categorical predictor. The authors validate the correction on both simulated and real-world datasets and integrate it with integrated path stability selection to achieve variable selection with false discovery control for mixed data.
By Jiahe Li, Omar Melikechi
Dynamic Learning Solutions is an automated pipeline that transforms NCERT textbook PDFs into interactive video explanations. Users upload a PDF and ask a question; the system retrieves relevant content, generates a multi-scene script, creates images with Stable Diffusion, animates them with DynamiCrafter, and adds synchronized narration via Google Text‑to‑Speech. The result is a coherent, textbook‑aligned video that turns static material into an engaging learning experience.
By Siddhanth Sridhar, Shreya Chaurasia, Baddela Sai Yaswantha Reddy, Deepak Parmar, Shylaja S S
OrchSLM is a routing framework that unifies non‑interactive orchestration methods for small language models (SLMs). It allows heterogeneous SLMs to independently generate candidate solutions while a router manages their cached outputs without further model interaction. By systematically probing OrchSLM, the study shows how orchestration behavior depends on task structure, model‑pool composition, and multi‑agent consensus.
By Chengxi Zhang, Yu Yao
The paper investigates the trade‑off between the costs of participating in federated learning (privacy, communication, compute) and the potential gains in model performance, framing this as a game‑theoretic problem of individual rationality versus autarky. It shows that clients can remain below their local‑training baseline for many rounds and that simply capping per‑round contributions harms learning. The authors propose a new mechanism that provides short‑term participation guarantees and personalized model evaluation, demonstrating theoretically and empirically that clients can avoid short‑term losses without significantly harming overall performance, even under moderate heterogeneity.
By Amin Meghrazi, Srinivasan Parthasarathy, Andrew Perrault
arXiv:2609.13655v1 Announce Type: new
Abstract: On evolving graphs, node classifiers must satisfy two key requirements: inductive generalization to newly arriving nodes under distribution shift and c...
By Jinwen Xu, Gonzalo Mateos Buckstein, Qin Lu
MedDiME is a latent-space, classifier‑guided diffusion framework designed for medical counterfactual image generation. It introduces a gradient‑driven adaptive masking mechanism that works directly in latent space, enabling spatially precise edits while avoiding the high computational and memory costs of pixel‑space methods. Experiments show MedDiME can produce high‑quality counterfactuals up to 40× faster and using 13× less GPU memory than previous diffusion baselines.
By Yan Zeng, Changlu Guo, Anders Nymark Christensen, Morten Rieger Hannemose, Anders Bjorholm Dahl
The paper proposes AI Deployment Accountability Engineering (ADAE), a new subdiscipline focused on establishing measurable, continuous, and actionable accountability for AI systems once they are deployed. ADAE treats accountability as a deployment-layer property, aiming to ensure systems remain within acceptable risk limits, identify failure contexts, attribute failures across technical and human components, and translate technical failures into downstream consequences. The authors outline a research agenda built around four pillars—structured discovery of context-dependent failure modes, privacy-preserving accountability measurement, system-level risk analysis for agentic AI, and translation of technical failures into operational and institutional risks—to support timely intervention in safety-critical socio-technical environments.
By Murat Kantarcioglu
The paper introduces Contextual Scalarisation Thompson Sampler (CSTS), a multi‑objective contextual bandit algorithm that learns to weight competing objectives based on observed context. It addresses the need for adaptable decision‑making in public media, where goals such as audience reach, cultural values, and operational constraints must be balanced. Experiments on Radio Télévision Suisse data demonstrate that CSTS improves contextual relevance and aligns more closely with expert curation than fixed‑weight or standard bandit methods.
By Th\'eo Ma\"etz, Luc Guillet, Andrea Cavallaro
The study examines how teens interact with companion AI chatbots, revealing that these systems offer comfort, recognition, identity exploration, and relationship rehearsal. However, teens also experience problematic attachment, social substitution, emotional dependence, and disruptions to academic and social life. The findings highlight the need for safety measures that address long‑term relationships, user‑controlled memory, privacy, relational boundaries, and healthy disengagement.
By Mohammad Namvarpour (Matt), Tyler Chang, Afsaneh Razi
VisInteract introduces a new paradigm for Text-to-Visualization that treats imperfect user queries as a dynamic, interaction-driven problem, requiring the system to recover true intent through multi-turn dialogue. The authors present VisInteract-Bench, the first benchmark for interactive Text-to-Vis, featuring controlled imperfection injection, a realistic user agent, and dual-perspective automated evaluation. They also propose Vis-MCTS, an enhanced Monte Carlo Tree Search algorithm that incorporates progressive widening, cross-rollout information sharing, and dimension-aware reward decomposition, achieving significant performance gains over existing baselines.
By Wenxin Xu, Jinwei Lu, Hwanhee Kim, Chen Jason Zhang, Xiao-Yong Wei, Haoyang Li, Yuanfeng Song