Isolater - Feed

Ax Cao Yuan, Junjun Wang 14h ago

ECTO: Exogenous-Conditioned Temporal Operator for Ultra-Short-Term Wind Power Forecasting

Deep learning method for ultra-short-term wind power forecasting using exogenous meteorological variables and time-series prediction.

Ax Yuanda Xu, Hejian Sang, Zhengze Zhou, Ran He, Zhipeng Wang, Alborz Geramifard 14h ago

Beyond GRPO and On-Policy Distillation: An Empirical Sparse-to-Dense Reward Principle for Language-Model Post-Training

Research on reward allocation principles for LLM post-training, comparing sparse sequence-level vs dense token-level rewards for different model stages.

Ax Jack Young 14h ago

WriteSAE: Sparse Autoencoders for Recurrent State

WriteSAE introduces sparse autoencoders for matrix updates in recurrent language models like Mamba-2 and RWKV-7, learning rank-1 matrix atoms to directly replace model writes.

Ax Jannis L\"ubsen, Annika Eichler 14h ago

Safe Bayesian Optimization for Uncertain Correlation Matrices in Linear Models of Co-Regionalization

Safety guarantees for multi-task Bayesian optimization with uncertain co-regionalization matrices in linear models.

Ax Jiaqi Sun, Boyang Sun, Rasmy M. H., Xiangchen Song, Kun Zhang 14h ago

MoRe: Modular Representations for Principled Continual Representation Learning on Sequential Data

MoRe: modular representation learning approach for continual learning on sequential data with minimal catastrophic forgetting.

Ax Shukai Du, Junzhe Zhang, Yiming Li 14h ago

Lagrangian Flow Matching: A Least-Action Framework for Principled Path Design

Lagrangian flow matching framework for designing probability paths in generative models beyond standard straight-line transport.

Ax Yuan Zhang, Lifeng Guo, Junwen Pan, Wenzhao Zheng, Wen Zhou, Kuan Cheng, Kurt Keutzer, Shanghang Zhang 14h ago

SEED: Targeted Data Selection by Weighted Independent Set

SEED: data selection method formulated as weighted independent set problem to balance quality and diversity in training datasets.

Ax Yangyou Liu, Zezhi Shao, Xinyu Chen, Hu Chen, Fei Wang, Yuankai Wu 14h ago

PULSE: Generative Phase Evolution for Non-Stationary Time Series Forecasting

PULSE: generative model for non-stationary time series forecasting using phase evolution to handle distribution shifts.

Ax Youngmok Ha, Viktor Schlegel, Yidan Sun, Anil Anthony Bharath 14h ago

Jacobian-Guided Anisotropic Noise Reshaping for Enhancing Representation Utility under Local Differential Privacy

Proposes Jacobian-guided noise reshaping for local differential privacy to improve data utility by task-aware noise injection.

Ax Zhitong Xu, Qiwei Yuan, Yinghao Chen, Shandian Zhe, Bin Shen 14h ago

Structured Neural Marked Point Processes for Interpretable Event Interaction Modeling

Structured neural marked point process model for interpretable multi-class event stream analysis and inter-event dependency discovery.

Ax Yuhan Wang, Haopeng Zhang, Yibo Ding, Jiaqi Yu, Xinyu Zhao, Yuhang Liu, Ziwei Zhang, Xiao Wang, Ruijie Wang 14h ago

S2Aligner: Pair-Efficient and Transferable Pre-Training for Sparse Text-Attributed Graphs

Pre-training method for sparse text-attributed graphs using LLM alignment to improve transferability in sparse node text scenarios.

Ax Muhammad Umer, Muhammad Ahmed Mohsin, Ahsan Bilal, Arslan Chaudhry, Andreas Haupt, Sanmi Koyejo, Emily Fox, John M. Cioffi 14h ago

General Preference Reinforcement Learning

General Preference RL framework unifying online RL and preference optimization for LLM alignment across both structured and open-ended tasks.

Ax Weinuo Ou 14h ago

Exact Linear Attention

Exact Linear Attention mechanism achieving linear computational complexity for Transformers through kernel decomposition without approximation error.

Ax Yongyu Wang 14h ago

Spectral structural distortion reveals redundant neurons in neural networks

Analysis of neural network redundancy using spectral structural distortion to identify prunable neurons beyond weight/activation magnitude.

Ax Yujie Lin, Chengyi Yang, Zhishang Xiang, Yiping Song, Jinsong Su 14h ago

ZeroUnlearn: Few-Shot Knowledge Unlearning in Large Language Models

Few-shot machine unlearning method for removing sensitive information from LLMs without retraining or degrading model utility.

Ax Han Guo, Jack Zhang, Arjun Menon, Driss Guessous, Vijay Thakkar, Yoon Kim, Tri Dao 14h ago

CODA: Rewriting Transformer Blocks as GEMM-Epilogue Programs

CODA optimization technique rewriting Transformer blocks as GEMM operations to reduce memory-bound operator bottlenecks in training.

Ax Lihong Lin, Haidong Kang 14h ago

Unlocking the Potential of Continual Model Merging: An ODE Perspective

Continual Model Merging framework using ODE perspective to control capacity allocation across sequential tasks, reducing catastrophic forgetting.

Ax David Pape, Jonathan Evertz, Lea Sch\"onherr 14h ago

The Silent Hyperparameter: Quantifying the Impact of Inference Backends on LLM Reproducibility

Study of how inference backend software systems impact LLM reproducibility and benchmark results at fractions of percentage point differences.

Ax Taijie Chen, Rui Su, Siyuan Feng, Laoming Zhang, Hongyang Zhang, Haijiao Wang, Zhaofeng Ma, Jintao Ke 14h ago

D$^3$-Subsidy: Online and Sequential Driver Subsidy Decision-Making for Large-Scale Ride-Hailing Market

Online sequential optimization for driver subsidies in ride-hailing platforms balancing supply-demand dynamics with multiple constraints.

Ax Junsol Kim, Byungkyu Lee 14h ago

AI-Augmented Surveys: Leveraging Large Language Models and Surveys for Opinion Prediction

LLM-based framework for predicting missing survey responses in repeated cross-sectional studies using question/respondent embeddings.

Ax Ikjun Choi, Ilmun Kim 14h ago

Computational-Statistical Trade-off in Kernel Two-Sample Testing with Random Fourier Features

Computational-statistical trade-offs in kernel two-sample testing using random Fourier features to reduce MMD test complexity.

Ax Yannis Montreuil, Shu Heng Yeo, Axel Carlier, Lai Xing Ng, Wei Tsang Ooi 14h ago

Optimal Query Allocation in Extractive QA with LLMs: A Learning-to-Defer Framework with Theoretical Guarantees

Learning-to-Defer framework allocating extractive QA queries to specialized experts with theoretical guarantees for resource-constrained LLM deployment.

Ax Renato Ferreira Pinto Jr., Nathaniel Harms 14h ago

Testing Support Size More Efficiently Than Learning Histograms

Statistical theory on testing probability distribution support size with minimal samples.

Ax Antoine Boutet, Lucas Magnana, Juliette S\'en\'echal 14h ago

Towards the Anonymization of the Language Modeling

Privacy-preserving language modeling approach to prevent memorization and exposure of sensitive personal information in fine-tuned models.

Ax Tansheng Zhu, Hongyu Zhou, Ke Jin, Xusheng Xu, Qiufan Yuan, Lijie Ji 14h ago

Bayesian Optimization by Kernel Regression and Density-based Exploration

Bayesian optimization algorithm using kernel regression and density-based exploration to reduce Gaussian process computational complexity.

Ax Chengpiao Huang, Yuhang Wu, Kaizheng Wang 14h ago

How Many Human Survey Respondents is a Large Language Model Worth? An Uncertainty Quantification Perspective

Framework converting LLM-simulated survey responses into confidence sets with uncertainty quantification for population parameters.

Ax Minh-Nhat Phung, Minh-Binh Tran 14h ago

Control, Optimal Transport and Neural Differential Equations in Supervised Learning

Research on approximating optimal transport equations using neural differential equations (Neural ODEs) for unbalanced transport problems.

Ax Jaemin Kim, Hangeol Chang, Hyunmin Hwang, Choonghan Kim, Jong Chul Ye 14h ago

Universal Reasoner: A Single, Composable Plug-and-Play Reasoner for Frozen LLMs

Plugin reasoner module for frozen LLMs that enables reasoning without retraining, compatible across different model architectures.

Ax Subba Reddy Oota, Khushbu Pahwa, Prachi Jindal, Satya Sai Srinath Namburi, Maneesh Singh, Tanmoy Chakraborty, Bapi S. Raju, Manish Gupta 14h ago

Task-conditioned probing of instruction-tuned multimodal LLMs: Region-specific brain alignment patterns under naturalistic stimuli

Study of how instruction-tuned multimodal LLMs align with brain activity patterns under naturalistic stimuli.

Ax Fanfei Li, Thomas Klein, Wieland Brendel, Robert Geirhos, Roland S. Zimmermann 14h ago

LAION-C: An Out-of-Distribution Benchmark for Web-Scale Vision Models

Benchmark dataset for evaluating out-of-distribution robustness of vision models on web-scale corruptions.

Ax Alain Bensoussan, Minh-Binh Tran, Bangjie Wang 14h ago

Control and optimization for Neural Partial Differential Equations in Supervised Learning

arXiv paper on control and optimization methods for neural PDEs in supervised learning contexts.

Ax Sita Sch\"onbauer, Johanna P. Carbone, Fredrik V. Eriksson, Florian Libisch, Andreas Gr\"uneis 14h ago

Machine-Learned Force Fields for Lattice Dynamics at Coupled-Cluster Level Accuracy

arXiv paper on machine-learned force fields for lattice dynamics at coupled-cluster accuracy levels.

Ax Seonghwan Choi, Beomseok Kang, Dongwon Jo, Jae-Joon Kim 14h ago

Retrospective Sparse Attention for Efficient Long-Context Generation

arXiv paper on efficient long-context LLM inference using retrospective sparse attention to optimize KV cache.

Ax Tianliang Xu, Ariel Hasell, Sabina Tomkins 14h ago

Anti-establishment sentiment on TikTok: Implications for understanding influence(rs) and expertise on social media

Analyzes anti-establishment sentiment on TikTok and its relationship to institutional distrust through social media influence dynamics.

Ax Rui-Yang Zhang, Lachlan Astfalck, Edward Cripps, David S. Leslie, Henry B. Moss 14h ago

BALLAST: Bayesian Active Learning with Look-ahead Amendment for Sea-drifter Trajectories under Spatio-Temporal Vector Fields

Proposes BALLAST, an active learning methodology for placing ocean drifters to infer spatio-temporal vector fields using physics-informed Gaussian processes.

Ax Yuyang Liu, Chuan Wen, Yihang Hu, Dinesh Jayaraman, Yang Gao 14h ago