Isolater - Feed

Ax Efstratios Zaradoukas, Bardh Prenkaj, Gjergji Kasneci 2/19/2026

Reinforcement Unlearning via Group Relative Policy Optimization

Method for removing memorized sensitive data from LLMs via group relative policy optimization, addressing GDPR/EU AI Act compliance without full retraining.

Ax Yadang Alexis Rouzoumka, Jean Pinsolle, Eug\'enie Terreaux, Christ\`ele Morisseau, Jean-Philippe Ovarlez, Chengfang Ren 2/19/2026

GEPC: Group-Equivariant Posterior Consistency for Out-of-Distribution Detection in Diffusion Models

Out-of-distribution detection for diffusion models using group equivariance properties learned from in-distribution convolutional data.

Ax Abhijit Gupta 2/19/2026

Cardinality-Preserving Attention Channels for Graph Transformers in Molecular Property Prediction

Graph transformer with cardinality-preserving attention channels for molecular property prediction with limited labeled data.

Ax Haocheng Xi, Shuo Yang, Yilong Zhao, Muyang Li, Han Cai, Xingyang Li, Yujun Lin, Zhuoyang Zhang, Jintao Zhang, Xiuyu Li, Zhiying Xu, Jun Wu, Chenfeng Xu, Ion Stoica, Song Han, Kurt Keutzer 2/19/2026

Quant VideoGen: Auto-Regressive Long Video Generation via 2-Bit KV-Cache Quantization

System optimization for autoregressive video generation using 2-bit KV-cache quantization to reduce memory from 30GB and enable hardware deployment.

Ax Jikai Jin, Kenneth Hung, Sanath Kumar Krishnamurthy, Baoyi Shi, Congshan Zhang 2/19/2026

Adaptive Exploration for Latent-State Bandits

Multi-armed bandit algorithms for sequential decision-making with hidden, time-varying states and unobserved confounders.

Ax Duc-Anh Nguyen, Nhien-An Le-Khac 2/19/2026

Align and Adapt: Multimodal Multiview Human Activity Recognition under Arbitrary View Combinations

Multimodal multiview human activity recognition model supporting arbitrary view combinations using contrastive learning and mixture-of-experts.

Ax Benedict Clark, Marta Oliveira, Rick Wilming, Stefan Haufe 2/19/2026

Feature salience -- not task-informativeness -- drives machine learning model explanations

Study showing feature salience rather than task-informativeness drives XAI model explanations, questioning assumption about importance attribution.

Ax Aaditya Vikram Prasad, Connor Watts, Jack Merullo, Dhruvil Gala, Owen Lewis, Thomas McGrath, Ekdeep Singh Lubana 2/19/2026

Features as Rewards: Scalable Supervision for Open-Ended Tasks via Interpretability

Using interpretable LLM features as supervision signal for RL-based open-ended tasks, demonstrated on hallucination reduction.

Ax Xander Davies, Giorgi Giglemiani, Edmund Lau, Eric Winsor, Geoffrey Irving, Yarin Gal 2/19/2026

Boundary Point Jailbreaking of Black-Box LLMs

Novel automated jailbreak attack method that evades classifier-based safeguards in frontier LLMs without requiring white/grey-box access.

Ax Chengzhi Hu, Jonas Dornbusch, David L\"udke, Stephan G\"unnemann, Leo Schwinn 2/19/2026

Closing the Distribution Gap in Adversarial Training for LLMs

Research on fundamental limitations of adversarial training for LLMs, showing models remain vulnerable to distribution-shifted attacks like prompt rewriting.

Ax S. Rijal, R. Neupane, S. P. Mainali, S. K. Regmi, S. Maharjan 2/19/2026

Monaural Multi-Speaker Speech Separation Using Efficient Transformer Model

Efficient transformer-based model for separating multiple speakers from single-channel audio, addressing the cocktail party problem.

Ax Tim R. Davidson, Veniamin Veselovsky, Martin Josifoski, Maxime Peyrard, Antoine Bosselut, Michal Kosinski, Robert West 2/19/2026

Evaluating Language Model Agency through Negotiations

Framework evaluating language model agency through multi-turn negotiation games, testing six public LLMs against real-world interactive scenarios.

Ax Xinrui Zhou, Yuhao Huang, Haoran Dou, Shijing Chen, Ao Chang, Jia Liu, Weiran Long, Jian Zheng, Erjiao Xu, Jie Ren, Alejandro F. Frangi, Ruobing Huang, Jun Cheng, Xiaomeng Li, Wufeng Xue, Dong Ni 2/19/2026

Ctrl-GenAug: Controllable Generative Augmentation for Medical Sequence Classification

Diffusion-based data augmentation method for medical sequence classification with controllable semantic and sequential generation.

Ax Zuzanna Buchnajzer, Kacper Dobek, Stanis{\l}aw Hapke, Daniel Jankowski, Krzysztof Krawiec 2/19/2026

Autoassociative Learning of Structural Representations for Modeling and Classification in Medical Imaging

Deep learning approach for medical imaging using autoassociative learning to capture discrete object categories instead of continuous features.

Ax Huadong Tang, Youpeng Zhao, Yan Huang, Min Xu, Jun Wang, Qiang Wu 2/19/2026

LMSeg: Unleashing the Power of Large-Scale Models for Open-Vocabulary Semantic Segmentation

Open-vocabulary semantic segmentation using large-scale vision-language models like CLIP to recognize unseen objects in images.

Ax Jinnan Guo, Kapil Vaswani, Andrew Paverd, Peter Pietzuch 2/19/2026

VerifiableFL: Verifiable Claims for Federated Learning using Exclaves

Federated learning system using secure enclaves to provide verifiable claims about model training data and algorithms, preventing malicious deviation.

Ax Fei Wu, Jia Hu, Geyong Min, Shiqiang Wang 2/19/2026

Adaptive Rank Allocation for Federated Parameter-Efficient Fine-Tuning of Language Models

Research on parameter-efficient fine-tuning of language models in federated settings, addressing privacy and efficiency for distributed training on resource-constrained devices.

Ax Jowaria Khan, Alexa Friedman, Sydney Evans, Rachel Klein, Runzi Wang, Katherine E. Manz, Kaley Beins, David Q. Andrews, Elizabeth Bondi-Kelly 2/19/2026

FOCUS on Contamination: Hydrology-Informed Noise-Aware Learning for Geospatial PFAS Mapping

Machine learning approach for geospatial PFAS contamination mapping using satellite and geospatial data.

Ax Jackie Baek, Hamsa Bastani, Shihan Chen 2/19/2026

Strategic Hiring under Algorithmic Monoculture

Game-theoretic analysis of strategic hiring behavior in labor markets using common algorithmic evaluation.

Ax Yinon Goldshtein, Gal Perelman, Assaf Schuster, Avi Ostfeld 2/19/2026

Large Language Models for Water Distribution Systems Modeling and Decision-Making

LLM agent framework enabling natural language interaction with water distribution system simulator EPANET.

Ax Giovanni Compiani, Ilya Morozov, Stephan Seiler 2/19/2026

Demand Estimation with Text and Image Data

Demand estimation using pre-trained deep learning models to extract embeddings from product images and text.

Ax Yipeng Sun, Linda-Sophie Schneider, Siyuan Mei, Jinhua Wang, Ge Hu, Mingxuan Gu, Chengze Ye, Fabian Wagner, Lan Song, Siming Bayer, Andreas Maier 2/19/2026

Filter2Noise: A Framework for Interpretable and Zero-Shot Low-Dose CT Image Denoising

Self-supervised framework for interpretable denoising of low-dose CT images from single images.

Ax Alexander Baumann, Leonardo Ayala, Silvia Seidlitz, Jan Sellner, Alexander Studier-Fischer, Berkin \"Ozdemir, Lena Maier-Hein, Slobodan Ilic 2/19/2026

CARL: Camera-Agnostic Representation Learning for Spectral Image Analysis

Representation learning approach for spectral image analysis that generalizes across different camera types.

Ax Kenichi Fujita, Shota Horiguchi, Yusuke Ijima 2/19/2026

Voice Impression Control in Zero-Shot TTS

Technique for controlling voice impression characteristics in zero-shot text-to-speech synthesis.

Ax Josh Qixuan Sun, Xiaoying Xing, Huaiyuan Weng, Chul Min Yeum, Mark Crowley 2/19/2026

View Invariant Learning for Vision-Language Navigation in Continuous Environments

Method for view-invariant learning in vision-language navigation for embodied AI agents.

Ax Zhan Qu, Shuzhou Yuan, Michael F\"arber 2/19/2026

PoeTone: A Framework for Constrained Generation of Structured Chinese Songci with LLMs

Framework for constrained generation of Chinese Songci poetry with LLMs using structural and tonal constraints.

Ax C\'edric Colas, Tracey Mills, Ben Prystawski, Michael Henry Tessler, Noah Goodman, Jacob Andreas, Joshua Tenenbaum 2/19/2026

Language and Experience: A Computational Model of Social Learning in Complex Tasks

Computational model of social learning combining linguistic guidance with sensorimotor experience for AI systems.

Ax Meshi Bashari, Yonghoon Lee, Roy Maor Lotan, Edgar Dobriban, Yaniv Romano 2/19/2026

Statistical Inference Leveraging Synthetic Data with Distribution-Free Guarantees

Framework for combining synthetic and real data in statistical inference with distribution-free guarantees.

Ax Yanfeng Yang, Siwei Chen, Pingping Hu, Zhaotong Shen, Yingjie Zhang, Zhuoran Sun, Shuai Li, Ziqi Chen, Kenji Fukumizu 2/19/2026

Conditionally Whitened Generative Models for Probabilistic Time Series Forecasting

Generative framework for probabilistic multivariate time series forecasting using conditional whitening.

Ax Lucas Bandarkar, Chenyuan Yang, Mohsen Fayyaz, Junlin Hu, Nanyun Peng 2/19/2026

Multilingual Routing in Mixture-of-Experts

Analysis of expert routing patterns in multilingual Mixture-of-Experts language models across languages.

Ax Daiki Chijiwa, Taku Hasegawa, Kyosuke Nishida, Shin'ya Yamaguchi, Tomoya Ohba, Tamao Sakao, Susumu Takeuchi 2/19/2026

Lossless Vocabulary Reduction for Auto-Regressive Language Models

Method for reducing vocabulary size in auto-regressive language models while maintaining lossless compression.

Ax Rongzhi Zhang, Liqin Ye, Yuzhao Heng, Xiang Chen, Tong Yu, Lingkai Kong, Sudheer Chava, Chao Zhang 2/19/2026

Precise Attribute Intensity Control in Large Language Models via Targeted Representation Editing

Technique for precise control over attribute intensity in LLM outputs via targeted representation editing.

Ax A. Thomsen (DES Collaboration), J. Bucko (DES Collaboration), T. Kacprzak (DES Collaboration), V. Ajani (DES Collaboration), J. Fluri (DES Collaboration), A. Refregier (DES Collaboration), D. Anbajagane (DES Collaboration), F. J. Castander (DES Collaboration), A. Fert\'e (DES Collaboration), M. Gatti (DES Collaboration), N. Jeffrey (DES Collaboration), A. Alarcon (DES Collaboration), A. Amon (DES Collaboration), K. Bechtol (DES Collaboration), M. R. Becker (DES Collaboration), G. M. Bernstein (DES Collaboration), A. Campos (DES Collaboration), A. Carnero Rosell (DES Collaboration), C. Chang (DES Collaboration), R. Chen (DES Collaboration), A. Choi (DES Collaboration), M. Crocce (DES Collaboration), C. Davis (DES Collaboration), J. DeRose (DES Collaboration), S. Dodelson (DES Collaboration), C. Doux (DES Collaboration), K. Eckert (DES Collaboration), J. Elvin-Poole (DES Collaboration), S. Everett (DES Collaboration), P. Fosalba (DES Collaboration), D. Gruen (DES Collaboration), I. Harrison (DES Collaboration), K. Herner (DES Collaboration), E. M. Huff (DES Collaboration), M. Jarvis (DES Collaboration), N. Kuropatkin (DES Collaboration), P. -F. Leget (DES Collaboration), N. MacCrann (DES Collaboration), J. McCullough (DES Collaboration), J. Myles (DES Collaboration), A. Navarro-Alsina (DES Collaboration), S. Pandey (DES Collaboration), A. Porredon (DES Collaboration), J. Prat (DES Collaboration), M. Raveri (DES Collaboration), M. Rodriguez-Monroy (DES Collaboration), R. P. Rollins (DES Collaboration), A. Roodman (DES Collaboration), E. S. Rykoff (DES Collaboration), C. S\'anchez (DES Collaboration), L. F. Secco (DES Collaboration), E. Sheldon (DES Collaboration), T. Shin (DES Collaboration), M. A. Troxel (DES Collaboration), I. Tutusaus (DES Collaboration), T. N. Varga (DES Collaboration), N. Weaverdyck (DES Collaboration), R. H. Wechsler (DES Collaboration), B. Yanny (DES Collaboration), B. Yin (DES Collaboration), Y. Zhang (DES Collaboration), J. Zuntz (DES Collaboration), M. Aguena (DES Collaboration), S. Allam (DES Collaboration), F. Andrade-Oliveira (DES Collaboration), D. Bacon (DES Collaboration), J. Blazek (DES Collaboration), D. Brooks (DES Collaboration), R. Camilleri (DES Collaboration), J. Carretero (DES Collaboration), R. Cawthon (DES Collaboration), L. N. da Costa (DES Collaboration), M. E. da Silva Pereira (DES Collaboration), T. M. Davis (DES Collaboration), J. De Vicente (DES Collaboration), S. Desai (DES Collaboration), P. Doel (DES Collaboration), J. Garc\'ia-Bellido (DES Collaboration), G. Gutierrez (DES Collaboration), S. R. Hinton (DES Collaboration), D. L. Hollowood (DES Collaboration), K. Honscheid (DES Collaboration), D. J. James (DES Collaboration), K. Kuehn (DES Collaboration), O. Lahav (DES Collaboration), S. Lee (DES Collaboration), J. L. Marshall (DES Collaboration), J. Mena-Fern\'andez (DES Collaboration), F. Menanteau (DES Collaboration), R. Miquel (DES Collaboration), J. Muir (DES Collaboration), R. L. C. Ogando (DES Collaboration), A. A. Plazas Malag\'on (DES Collaboration), E. Sanchez (DES Collaboration), D. Sanchez Cid (DES Collaboration), I. Sevilla-Noarbe (DES Collaboration), M. Smith (DES Collaboration), E. Suchyta (DES Collaboration), M. E. C. Swanson (DES Collaboration), D. Thomas (DES Collaboration), C. To (DES Collaboration), D. L. Tucker (DES Collaboration) 2/19/2026

Dark Energy Survey Year 3 results: Simulation-based $w$CDM inference from weak lensing and galaxy clustering maps with deep learning: Analysis design

Deep learning pipeline for cosmological inference combining weak lensing and galaxy clustering data from Dark Energy Survey.

Ax Maria Bou-Sakr-El-Tayar, Jason J. Bramburger, Matthew J. Colbrook 2/19/2026

Weighted Birkhoff Averages Accelerate Data-Driven Methods

Mathematical framework for accelerating ergodic averages in dynamical systems using weighted Birkhoff averaging methods.

Ax Ho Jang, Jackson C. Glass, Gia-Wei Chern 2/19/2026

Learning Degenerate Manifolds of Frustrated Magnets with Boltzmann Machines

Restricted Boltzmann Machines model ground-state manifolds of frustrated magnets. Physics simulation, not AI tools.

Ax Kunj Joshi, David A. Smith 2/19/2026

Randomized Masked Finetuning: An Efficient Way to Mitigate Memorization of PIIs in LLMs

Randomized Masked Fine-Tuning reduces PII memorization in LLMs during fine-tuning while maintaining performance. Privacy-preserving technique.

Ax Alessandro Quadrio, Antonio Candelieri 2/19/2026

Transformers for Tabular Data: A Training Perspective of Self-Attention via Optimal Transport

Self-attention training analysis via optimal transport theory for tabular classification. Theoretical perspective on transformers.

Ax Duc Hoang, Aarush Gupta, Philip Harris 2/19/2026

KANEL\'E: Kolmogorov-Arnold Networks for Efficient LUT-based Evaluation

KANELÉ: Kolmogorov-Arnold Networks optimized for FPGA lookup table deployment. Efficient neural network inference framework.

Ax Anshima Singh, David J. Silvester 2/19/2026

Adaptive Sampling for Hydrodynamic Stability

Adaptive sampling for detecting bifurcation boundaries in fluid dynamics simulations. Scientific computing, not AI-focused.

Ax Zhou Fan, Leda Wang 2/19/2026

High-dimensional learning dynamics of multi-pass Stochastic Gradient Descent in multi-index models

Theoretical analysis of SGD learning dynamics in high-dimensional multi-index models. Fundamental ML research.

Ax Fabian P. Kr\"uger, Andrea Hunklinger, Adrian Wolny, Tim J. Adler, Igor Tetko, Santiago David Villalba 2/19/2026

SEISMO: Increasing Sample Efficiency in Molecular Optimization with a Trajectory-Aware LLM Agent

SEISMO: LLM agent for sample-efficient molecular optimization using trajectory awareness. Applies agents to chemistry.

Ax Damien Ferbach, Courtney Paquette, Gauthier Gidel, Katie Everett, Elliot Paquette 2/19/2026

Logarithmic-time Schedules for Scaling Language Models with Momentum

Time-varying AdamW schedules (beta, weight-decay) for language model training exploiting power-law data structure. Improves LLM training efficiency.

Ax Amir H. Ashouri, Shayan Shirahmad Gale Bagi, Kavin Satheeskumar, Tejas Srikanth, Jonathan Zhao, Ibrahim Saidoun, Ziwen Wang, Bryan Chan, Tomasz S. Czajkowski 2/19/2026

Protean Compiler: An Agile Framework to Drive Fine-grain Phase Ordering

Compiler optimization using machine learning for phase ordering decisions. Developer tools, not AI-focused.

Ax Ross Greer, Maitrayee Keskar, Angel Martinez-Sanchez, Parthib Roy, Shashank Shriram, Mohan Trivedi 2/19/2026

Vision and Language: Novel Representations and Artificial intelligence for Driving Scene Safety Assessment and Autonomous Vehicle Planning

Vision-language models for autonomous driving safety assessment and planning. Applies VLMs to scene understanding and decision-making.

Ax Soham Bakshi, Sunrit Chakraborty 2/19/2026

From Collapse to Improvement: Statistical Perspectives on the Evolutionary Dynamics of Iterative Training on Contaminated Sources

Statistical analysis of model collapse in iterative training with synthetic data. Shows conditions for improvement despite contamination.

Ax Zachary Pedram Dadfar 2/19/2026