Isolater - Feed

Ax Jinyang Li, Shuhao Mei, Xiaoyu Xiao, Shuhang Li, Ruoxi Yun, Jinbo Sun 3/4/2026

Rethinking Time Series Domain Generalization via Structure-Stratified Calibration

Domain generalization for time series via structure-stratified calibration addressing heterogeneous dynamical systems with distinct feature distributions.

Ax George Bredis, Nikita Balagansky, Daniil Gavrilov, Ruslan Rakhimov 3/4/2026

Next Embedding Prediction Makes World Models Stronger

NE-Dreamer agent uses temporal transformer to predict next-step embeddings for improved model-based reinforcement learning in high-dimensional domains.

Ax Qi Huang, Furong Ye, Ananta Shahane, Thomas B\"ack, Niki van Stein 3/4/2026

From Heuristic Selection to Automated Algorithm Design: LLMs Benefit from Strong Priors

LLMs for automated algorithm design benefit from strong priors; investigates token-wise attribution of prompts guiding algorithm generation.

Ax Thomas L. Lee, Edoardo M. Ponti, Amos Storkey 3/4/2026

Adapting Time Series Foundation Models through Data Mixtures

Fine-tuning time series foundation models using data mixtures and LoRA for improved zero-shot forecasting on new domains.

Ax Jiaqi Wang, Zhiguang Cao, Peng Zhao, Rui Cao, Yubin Xiao, Yuan Jiang, You Zhou 3/4/2026

Learning Memory-Enhanced Improvement Heuristics for Flexible Job Shop Scheduling

Deep reinforcement learning approach for flexible job-shop scheduling using memory-enhanced improvement heuristics for manufacturing optimization.

Ax Davide Maran, Davide Salaorni, Marcello Restelli 3/4/2026

Learning in Markov Decision Processes with Exogenous Dynamics

RL algorithm design for MDPs with exogenous dynamics where only subset of state variables are affected by agent actions.

Ax Ziruo Hao, Tao Yang, Xiaofeng Wu, Bo Hu 3/4/2026

Distributed Dynamic Invariant Causal Prediction in Environmental Time Series

Framework for extracting invariant causal relationships from environmental time series data using distributed dynamic causal prediction methods.

Ax Bo Liu, Shao-Bo Lin, Changmiao Wang, Xiaotong Liu 3/4/2026

Towards Accurate and Interpretable Time-series Forecasting: A Polynomial Learning Approach

Interpretable time series forecasting approach using polynomial learning to improve trust and debuggability for predictive maintenance applications.

Ax Julianna Piskorz, Katarzyna Kobalczyk, Mihaela van der Schaar 3/4/2026

Eliciting Numerical Predictive Distributions of LLMs Without Autoregression

Method for obtaining numerical predictive distributions from LLMs without autoregression for efficient regression tasks like time series and tabular forecasting.

Ax Pardhu Sri Rushi Varma Konduru 3/4/2026

On the Structural Limitations of Weight-Based Neural Adaptation and the Role of Reversible Behavioral Learning

Study on structural irreversibility in weight-based neural model adaptation, proposing reversible behavioral learning as alternative to parameter fine-tuning.

Ax Mohammadreza Nakheai, Aidan Scannell, Kevin Luck, Joni Pajarinen 3/4/2026

Contextual Latent World Models for Offline Meta Reinforcement Learning

Framework introducing contextual latent world models for offline meta-reinforcement learning with improved task representation learning via self-supervised methods.

Ax Fengzhi Li, Liang Zhang, Yuan Zuo, Ruiqing Zhao, YanSong Liu, Yunfei Ma, Fanyu Meng, Junlan Feng 3/4/2026

Beyond One-Size-Fits-All: Adaptive Subgraph Denoising for Zero-Shot Graph Learning with Large Language Models

Approach combining LLMs with Graph Neural Networks for zero-shot graph learning using adaptive subgraph denoising to handle cross-modal alignment issues.

Ax Alberto Mi\~no Calero, Luis Salamanca, Konstantinos E. Tatsis 3/4/2026

Enhancing Physics-Informed Neural Networks with Domain-aware Fourier Features: Towards Improved Performance and Interpretable Results

Method improving Physics-Informed Neural Networks using Domain-aware Fourier Features for better training and interpretability of PDE-based models.

Ax Zhenquan Yao, Zitong Huang, Yihan Zeng, Jianhua Han, Hang Xu, Chun-Mei Feng, Jianwei Ma, Wangmeng Zuo 3/4/2026

CGL: Advancing Continual GUI Learning via Reinforcement Fine-Tuning

Framework for continual learning in GUI agents using multimodal LLMs with reinforcement fine-tuning to adapt to new tasks without catastrophic forgetting.

Ax Kohki Akiba, Shinnosuke Matsuo, Shota Harada, Ryoma Bise 3/4/2026

Leveraging Label Proportion Prior for Class-Imbalanced Semi-Supervised Learning

Method addressing class imbalance in semi-supervised learning using Proportion Loss regularization to align predictions with global class distribution.

Ax Yenan Wang, Carla Fabiana Chiasserini, Elad Michael Schiller 3/4/2026

Integrating Homomorphic Encryption and Synthetic Data in FL for Privacy and Learning Quality

Research on federated learning combining homomorphic encryption and synthetic data to improve privacy and learning quality while reducing computational costs.

Ax Eliott Van Dieren, Tommaso Vanzan, Fabio Nobile 3/4/2026

LAGO: A Local-Global Optimization Framework Combining Trust Region Methods and Bayesian Optimization

arXiv: Optimization algorithm combining Bayesian optimization with trust region methods through adaptive competition.

Ax Robin Young 3/4/2026

Why Does RLAIF Work At All?

arXiv: Theoretical explanation for reinforcement learning from AI feedback through latent value hypothesis.

Ax Tian-Shuang Wu, Shen-Huan Lyu, Ning Chen, Yi-Xiao He, Bing Tang, Baoliu Ye, Qingfu Zhang 3/4/2026

Breaking the Prototype Bias Loop: Confidence-Aware Federated Contrastive Learning for Highly Imbalanced Clients

arXiv: Federated contrastive learning framework addressing prototype bias in imbalanced distributed data.

Ax Cheng Peng, Yonghao Li, Wanfu Gao, Jie Wen, Weiping Ding 3/4/2026

SEHFS: Structural Entropy-Guided High-Order Correlation Learning for Multi-View Multi-Label Feature Selection

arXiv: Information-theoretic feature selection for multi-view multi-label learning using structural entropy.

Ax Xuan Yang, Jiayu Liu, Yuhang Lai, Hao Xu, Zhenya Huang, Ning Miao 3/4/2026

Step-Level Sparse Autoencoder for Reasoning Process Interpretation

arXiv: Step-level sparse autoencoders for interpreting LLM reasoning processes in chain-of-thought outputs.

Ax Federico Giannini, Giacomo Ziffer, Emanuele Della Valle 3/4/2026

cPNN: Continuous Progressive Neural Networks for Evolving Streaming Time Series

arXiv: Continuous progressive neural networks handling streaming time series with concept drift and temporal dependencies.

Ax Marko Pranji\'c, Boshko Koloski, Nada Lavra\v{c}, Senja Pollak, Marko Robnik-\v{S}ikonja 3/4/2026

Incremental Graph Construction Enables Robust Spectral Clustering of Texts

arXiv: Incremental k-NN graph construction method improving robustness of spectral clustering on text embeddings.

Ax Thomas Krug, Daniel Neider 3/4/2026

Reinforcement Learning with Symbolic Reward Machines

arXiv: Reinforcement learning approach using symbolic reward machines to eliminate manual labeling functions.

Ax Linyan Gu, Lihua Yang, Feng Zhou 3/4/2026

On the Expressive Power of Transformers for Maxout Networks and Continuous Piecewise Linear Functions

arXiv: Study of Transformer expressive power proving universal approximation for maxout networks and piecewise linear functions.

Ax Ruinan Jin, Yingbin Liang, Shaofeng Zou 3/4/2026

Why Adam Can Beat SGD: Second-Moment Normalization Yields Sharper Tails

arXiv: Theoretical analysis explaining Adam optimizer's empirical advantage over SGD through second-moment normalization.

Ax Jiaqi Lv, Qingfeng Du, Yu Zhang, Yongqi Han, Sheng Li 3/4/2026

Multi-Scale Adaptive Neighborhood Awareness Transformer For Graph Fraud Detection

arXiv: Multi-scale adaptive transformer for graph fraud detection using neighborhood awareness.

Ax Pengyu Lai, Yixiao Chen, Dewu Yang, Rui Wang, Feng Wang, Hui Xu 3/4/2026

From Complex Dynamics to DynFormer: Rethinking Transformers for PDEs

DynFormer: Transformer architecture for solving PDEs that respects scale separation in complex dynamics, replacing expensive classical solvers.

Ax Haotian Wang 3/4/2026

Joint Training Across Multiple Activation Sparsity Regimes

Training strategy applying global top-k activation sparsity constraints cyclically to improve neural network generalization across dense/sparse regimes.

Ax Dan Stowell 3/4/2026

Torus embeddings

Torus embeddings: research on representing deep learning embeddings on toroidal manifolds instead of Euclidean space for efficiency.

Ax Carolin Heinzler, Kasra Malihi, Amartya Sanyal 3/4/2026

Less Noise, Same Certificate: Retain Sensitivity for Unlearning

Certified machine unlearning with reduced noise requirements using sensitivity-aware DP techniques.

Ax Cullen Anderson, Narmeen Oozeer, Foad Namjoo, Remy Ogasawara, Amirali Abdullah, Jeff M. Phillips 3/4/2026

Understanding and Mitigating Dataset Corruption in LLM Steering

Analysis of contrastive steering robustness in LLMs, addressing data corruption and adversarial noise in inference-time steering.

Ax Hirofumi Suzuki, Kentaro Kanamori, Takuya Takagi, Thong Pham, Takashi Nicholas Maeda, Shohei Shimizu 3/4/2026

I-CAM-UV: Integrating Causal Graphs over Non-Identical Variable Sets Using Causal Additive Models with Unobserved Variables

Causal discovery method for integrating multiple datasets with non-identical variables using causal additive models.

Ax Dragan Ma\v{s}ulovi\'c 3/4/2026

Coalgebras for categorical deep learning: Representability and universal approximation

Categorical deep learning framework using coalgebras for equivariant representations and universal approximation theory.

Ax Adam Watts (Los Alamos National Laboratory), Andrew Jeon (Los Alamos National Laboratory), Destry Newton (Los Alamos National Laboratory), Ryan Bowering (University of Rochester) 3/4/2026

Inverse Reconstruction of Shock Time Series from Shock Response Spectrum Curves using Machine Learning

ML approach to reconstruct time-domain shock signals from response spectrum curves via neural networks.

Ax Mertcan Daysalilar, Fuat Uyguroglu, Gabriel Nicolosi, Adam Meyers 3/4/2026

SynthCharge: An Electric Vehicle Routing Instance Generator with Feasibility Screening to Enable Learning-Based Optimization and Benchmarking

Parametric benchmark generator for electric vehicle routing problems with learning-based optimization evaluation.

Ax Patrick Inoue, Florian R\"ohrbein, Andreas Knoblauch 3/4/2026