Isolater - Feed

Ax Denizalp Goktas, Gerardo Ria\~no-Brice\~no, Alif Abdullah, Aryan Nair, Chenkai Shen, Beatriz de Lucio, Alexandra Magnusson, Farhan Mashrur, Ahmed Abdulla, Shawrna Sen, Mahitha Thippireddy, Gregory Schwartz, Amy Greenwald 4/14/2026

TempusBench: An Evaluation Framework for Time-Series Forecasting

Evaluation framework and benchmark for assessing time-series foundation models and forecasting approaches.

Ax Haolin Li, Shuyang Jiang, Ruipeng Zhang, Jiangchao Yao, Ya Zhang, Yanfeng Wang 4/14/2026

Eliciting Medical Reasoning with Knowledge-enhanced Data Synthesis: A Semi-Supervised Reinforcement Learning Approach

Semi-supervised reinforcement learning approach using knowledge-enhanced data synthesis to improve medical reasoning in LLMs.

Ax Vincent S. Kather, Sylvain Haupert, Burooj Ghani, Dan Stowell 4/14/2026

bacpipe: a Python package to make bioacoustic deep learning models accessible

Python package providing bioacoustic deep learning models and evaluation tools for analyzing passive acoustic monitoring data.

Ax Patrick Lutz, Themistoklis Haris, Arjun Chandra, Aditya Gangrade, Venkatesh Saligrama 4/14/2026

Layerwise Dynamics for In-Context Classification in Transformers

Study of multi-class linear classification in transformers using feature and label permutation equivariance to extract interpretable algorithms.

Ax Maxim Bolshim (ITMO University, Saint Petersburg, Russia), Alexander Kugaevskikh (ITMO University, Saint Petersburg, Russia) 4/14/2026

Inter-Layer Hessian Analysis of Neural Networks with DAG Architectures

Analytical formalism decomposing Hessian matrices in neural networks with DAG architectures into inter-layer blocks.

Ax Yunhui Jang, Lu Zhu, Jake Fawkes, Alisandra Kaye Denton, Dominique Beaini, Emmanuel Noutahi 4/14/2026

Towards Autonomous Mechanistic Reasoning in Virtual Cells

Framework for autonomous mechanistic reasoning in virtual cells using LLMs, representing biological reasoning as mechanistic action graphs.

Ax Nicolas Rodriguez-Alvarez (Instituto de Educacion Secundaria Parquesol, Valladolid, Spain), Fernando Rodriguez-Merino (University of Valladolid, Valladolid, Spain) 4/14/2026

Fairness is Not Flat: Geometric Phase Transitions Against Shortcut Learning

Methodology to mitigate shortcut learning and demographic bias in deep neural networks using geometric a priori approaches.

Ax J. Oppliger, M. Stifter, A. R\"uegg, I. Bia{\l}o, L. Martinelli, P. G. Freeman, D. Prabhakaran, J. Zhao, Q. Wang, J. Chang 4/14/2026

Autonomous Diffractometry Enabled by Visual Reinforcement Learning

Model-free reinforcement learning system for autonomous crystal alignment using visual information without domain knowledge of crystallography.

Ax Fei Tang, Zhiqiong Lu, Boxuan Zhang, Weiming Lu, Jun Xiao, Yueting Zhuang, Yongliang Shen 4/14/2026

ClawGUI: A Unified Framework for Training, Evaluating, and Deploying GUI Agents

ClawGUI framework for training, evaluating, and deploying GUI agents that interact with software through visual interfaces, with online RL and evaluation infrastructure.

Ax Hugh Blayney, \'Alvaro Arroyo, Johan Obando-Ceron, Pablo Samuel Castro, Aaron Courville, Michael M. Bronstein, Xiaowen Dong 4/14/2026

A Mechanistic Analysis of Looped Reasoning Language Models

Mechanistic analysis of looped reasoning language models examining internal dynamics and latent state evolution compared to standard feedforward models.

Ax Mihir Prabhudesai, Aryan Satpathy, Yangmin Li, Zheyang Qin, Nikash Bhardwaj, Amir Zadeh, Chuan Li, Katerina Fragkiadaki, Deepak Pathak 4/14/2026

Solving Physics Olympiad via Reinforcement Learning on Physics Simulators

Uses reinforcement learning on physics simulators to train models solving Physics Olympiad problems, addressing lack of large-scale physics QA datasets for reasoning models.

Ax Yaxin Yu, Long Chen, Minfu Feng 4/14/2026

SHANG++: Robust Stochastic Acceleration under Multiplicative Noise

SHANG++: Accelerated stochastic gradient descent methods robust to multiplicative noise in gradient updates.

Ax Jon M Laurent, Albert Bou, Michael Pieler, Conor Igoe, Alex Andonian, Siddharth Narayanan, James Braza, Alexandros Sanchez Vassopoulos, Jacob L Steenwyk, Blake Lash, Andrew D White, Samuel G Rodriques 4/14/2026

LABBench2: An Improved Benchmark for AI Systems Performing Biology Research

LABBench2: Improved benchmark for evaluating AI systems and agents on biology research tasks with real-world capabilities.

Ax Muyan Hu, Ahan Gupta, Jiachen Yuan, Vima Gupta, Taeksang Kim, Xin Xu, Janardhan Kulkarni, Ofer Dekel, Vikram Adve, Charith Mendis 4/14/2026

VTC: DNN Compilation with Virtual Tensors for Data Movement Elimination

VTC: DNN compilation method using virtual tensors to eliminate data movement in neural network workloads including LLMs.

Ax Magda Dubois, Ekin Zorer, Maia Hamin, Joe Skinner, Alexandra Souly, Jerome Wynne, Harry Coppock, Lucas Satos, Sayash Kapoor, Sunischal Dev, Keno Juchems, Kimberly Mai, Timo Flesch, Lennart Luettgau, Charles Teague, Eric Patey, JJ Allaire, Lorenzo Pacchiardi, Jose Hernandez-Orallo, Cozmin Ududec 4/14/2026

Seven simple steps for log analysis in AI systems

Pipeline and best practices for log analysis in AI systems to understand model behaviors, with code examples in Inspect framework.

Ax Jiachen Zhu, Lingyu Yang, Rong Shan, Congmin Zheng, Zeyu Zheng, Weiwen Liu, Yong Yu, Weinan Zhang, Jianghao Lin 4/14/2026

Turing Test on Screen: A Benchmark for Mobile GUI Agent Humanization

Benchmark measuring humanization and anti-detection capabilities of mobile GUI agents against platform countermeasures.

Ax Yaniv Leviathan (Cheenu), Dani Valevski (Cheenu), Matan Kalman (Cheenu), Danny Lumen (Cheenu), Eyal Segalis (Cheenu), Eyal Molad (Cheenu), Shlomi Pasternak (Cheenu), Vishnu Natchu (Cheenu), Valerie Nygaard (Cheenu), Srinivasan (Cheenu), Venkatachary, James Manyika, Yossi Matias 4/14/2026

Generative UI: LLMs are Effective UI Generators

Demonstrating LLMs can generate UI interfaces and content together with proper prompting and tool integration.

Ax Hongyu Chen, Liang Lin, Guangrun Wang 4/14/2026

OOWM: Structuring Embodied Reasoning and Planning via Object-Oriented Programmatic World Modeling

Object-Oriented Programmatic World Modeling (OOWM) for embodied reasoning and planning in robotic tasks using LLMs.

Ax Yunfei Feng, Xi Zhao, Cheng Zhang, Dahu Feng, Daolin Cheng, Jianqi Yu, Yubin Xia, Erhu Feng 4/14/2026

MobiFlow: Real-World Mobile Agent Benchmarking through Trajectory Fusion

MobiFlow: Benchmark for mobile agents using trajectory fusion for real-world GUI task evaluation without system-level APIs.

Ax Prahlad G. Menon 4/14/2026

Persistent Identity in AI Agents: A Multi-Anchor Architecture for Resilient Memory and Continuity

Architecture for maintaining persistent identity in AI agents through multi-anchor memory to prevent catastrophic forgetting.

Ax Jash Vira, Ashley Harris 4/14/2026

Spatial Competence Benchmark

Spatial Competence Benchmark (SCBench) evaluating large models on spatial reasoning, environment representation, and planning tasks.

Ax Xinyi Hu, Yuhao Shen, Baolin Zhang, Hengxin Zhang, Jun Dai, Shuang Ge, Lei Chen, Yue Li, Mingcheng Wan 4/14/2026

ECHO: Elastic Speculative Decoding with Sparse Gating for High-Concurrency Scenarios

ECHO: Speculative decoding optimization for LLM inference in high-concurrency serving with sparse gating.

Ax Stephan Sandfuchs, Maximilian Melchert, J\"org Frochte 4/14/2026

LLMs for Text-Based Exploration and Navigation Under Partial Observability

Benchmark evaluating LLMs as text-only controllers for exploration and navigation in gridworlds under partial observability.

Ax Mohamed Rissal Hedna, Jan Strich, Martin Semmann, Chris Biemann 4/14/2026

Self-Calibrating Language Models via Test-Time Discriminative Distillation

Method for self-calibrating LLMs at test-time through discriminative distillation to reduce overconfidence without labeled data.

Ax Xinyuan An, Tao Luo, Gengyun Peng, Yaobing Wang, Kui Ren, Dongxia Wang 4/14/2026

FlowHijack: A Dynamics-Aware Backdoor Attack on Flow-Matching Vision-Language-Action Models

Study of backdoor security vulnerabilities in flow-matching Vision-Language-Action models used for robotics, exploiting vector field dynamics.

Ax Jiani Cao, Kun Wang, Yang Liu, Zhenjiang Li 4/14/2026

NeuroPath: Practically Adopting Motor Imagery Decoding through EEG Signals

NeuroPath system for motor imagery decoding from EEG signals for brain-computer interfaces in prosthetics and rehabilitation.

Ax Kumar Saurav 4/14/2026

Real-Time Voicemail Detection in Telephony Audio Using Temporal Speech Activity Features

Lightweight speech activity-based approach for real-time voicemail detection in telephony using tree ensemble classification.

Ax Amey Thakur, Sarvesh Talele 4/14/2026

A Modular Zero-Shot Pipeline for Accident Detection, Localization, and Classification in Traffic Surveillance Video

Zero-shot modular pipeline for traffic accident detection, localization, and classification without labeled training data.

Ax S. Afifi, O. Alo, I. Thakkar, S. Pasricha 4/14/2026

Sustainable Transformer Neural Network Acceleration with Stochastic Photonic Computing

ASTRA silicon-photonic accelerator for transformers using stochastic computing to reduce computation and memory demands.

Ax Dhruv Atreja, Julia White, Nikhil Nayak, Kelton Zhang, Henrijs Princis, George Hurn-Maloney, Ash Lewis, Urchade Zaratiana 4/14/2026

Pioneer Agent: Continual Improvement of Small Language Models in Production

Pioneer Agent automates continuous improvement of small language models in production through closed-loop data curation, failure diagnosis, and iteration control.

Ax Kyle Waters, Lucas Nuzzi, Tadhg Looram, Alessandro Tomasiello, Ariel Ghislain Kemogne Kamdoum, Bikun Li, Damien Sileo, Egor Kretov, Francesco Fournier-Facio, Georgios Soloupis, Haile Kassahun, Hew Wolff, Jiaqi Cai, Lianghui Li, Marc Roth, Mohinder Naiya, Naixu Guo, Qicheng Tang, Richard Wheeler, Samuele Sala, Serguei Popov, Steven Dillman, Yuqi Li 4/14/2026