Ax Zachary Bogorad, Ibrahim Elsharkawy, Yonatan Kahn, Andrew J. Larkoski, Noam Levi 4/30/2026

Generative models on phase space

Applies deep generative models (diffusion, flow matching) to high-energy physics data on constrained manifolds.

Ax Hengshuai Yao, Xing Chen, Ahmed Murtadha, Jin Li, Yasin Abbasi Yadkori, Shuai Shao, Changling Liu, Guan Wang, Mingli Yuan, William Chen, Sen Song 4/30/2026

Why Attend to Everything? Focus is the Key

Introduces Focus, an efficient attention mechanism using learnable centroids as gates to reduce quadratic scaling while maintaining model performance.

Ax Ziyi Chen, Mengxian Lyu, Cheng Peng, Yonghui Wu 4/30/2026

Retrieval-Augmented LLMs for Evidence Localization in Clinical Trial Recruitment from Longitudinal EHR Narratives

Applies retrieval-augmented LLMs for evidence localization in clinical trial patient screening from EHR narratives.

Ax Antonio J. Bujana, Aydin I. Karsilayan 4/30/2026

A Self-Calibrating Framework for Analog Circuit Sizing Using LLM-Derived Analytical Equations

Framework using LLMs to derive interpretable, calibrated analytical equations for analog circuit sizing with device-dimension traceability.

Ax Xingyan Liu, Xiyue Luo, Linyu Li, Ganghong Huang, Jianfeng Liu, Honglin Qiao 4/30/2026

SkillForge: Forging Domain-Specific, Self-Evolving Agent Skills in Cloud Technical Support

SkillForge creates domain-specific, self-evolving LLM agent skills for cloud technical support with systematic failure tracing and refinement mechanisms.

Ax Shreyansh Pathak, Jyotishman Das 4/30/2026

Graph Propagated Projection Unlearning: A Unified Framework for Vision and Audio Discriminative Models

Proposes Graph-Propagated Projection Unlearning algorithm for selective information erasure from vision and audio neural networks.

Ax Davyd Naveriani, Albert Zeyer, Ralf Schl\"uter, Hermann Ney 4/30/2026

Diffusion Language Models for Speech Recognition

Explores masked diffusion language models and uniform-state diffusion models for rescoring ASR hypotheses in speech recognition tasks.

Ax Wentao Hu, Yanbo Zhai, Xiaohui Hu, Mingkuan Zhao, Shanhong yu, Xue Liu, Kaidong Yu, Shuangyong Song, Xuelong Li 4/30/2026

Awakening Dormant Experts:Counterfactual Routing to Mitigate MoE Hallucinations

Addresses hallucinations in Mixture-of-Experts models by using counterfactual routing to activate dormant expert specialists handling long-tail knowledge.

Ax Yiqun Sun, Pengfei Wei, Lawrence B. Hsieh 4/30/2026

Don't Retrieve, Navigate: Distilling Enterprise Knowledge into Navigable Agent Skills for QA and RAG

Corpus2Skill distills document corpora into hierarchical skill directories for LLM agents to navigate at inference time, enabling better evidence combination for RAG and QA tasks.

Ax Benedikt Bollig, Matthias F\"ugger, Thomas Nowak 4/30/2026

Provable Coordination for LLM Agents via Message Sequence Charts

Domain-specific language for specifying LLM agent coordination using message sequence charts, enabling formal verification of multi-agent systems.

Ax Vin Bhaskara, Haicheng Wang 4/30/2026

Curiosity-Critic: Cumulative Prediction Error Improvement as a Tractable Intrinsic Reward for World Model Training

Intrinsic reward mechanism for world model training based on cumulative prediction error improvement with tractable per-step surrogate.

Ax Open-H-Embodiment Consortium, :, Nigel Nelson, Juo-Tung Chen, Jesse Haworth, Xinhao Chen, Lukas Zbinden, Dianye Huang, Alaa Eldin Abdelaal, Alberto Arezzo, Ayberk Acar, Farshid Alambeigi, Carlo Alberto Ammirati, Yunke Ao, Pablo David Aranda Rodriguez, Soofiyan Atar, Mattia Ballo, Noah Barnes, Federica Barontini, Filip Binkiewicz, Peter Black, Sebastian Bodenstedt, Leonardo Borgioli, Nikola Budjak, Benjamin Calm\'e, Fabio Carrillo, Nicola Cavalcanti, Changwei Chen, Haoxin Chen, Sihang Chen, Qihan Chen, Zhongyu Chen, Ziyang Chen, Shing Shin Cheng, Meiqing Cheng, Min Cheng, Zih-Yun Sarah Chiu, Xiangyu Chu, Camilo Correa-Gallego, Giulio Dagnino, Anton Deguet, Jacob Delgado, Jonathan C. DeLong, Kaizhong Deng, Alexander Dimitrakakis, Qingpeng Ding, Hao Ding, Giovanni Distefano, Daniel Donoho, Anqing Duan, Marco Esposito, Shane Farritor, Jad Fayad, Zahi Fayad, Mario Ferradosa, Filippo Filicori, Chelsea Finn, Philipp F\"urnstahl, Jiawei Ge, Stamatia Giannarou, Xavier Giralt Ludevid, Frederic Giraud, Aditya Amit Godbole, Ken Goldberg, Antony Goldenberg, Diego Granero Marana, Xiaoqing Guo, Tam\'as Haidegger, Evan Hailey, Pascal Hansen, Ziyi Hao, Kush Hari, Kengo Hayashi, Jonathon Hawkins, Shelby Haworth, Ortrun Hellig, S. Duke Herrell, Zhouyang Hong, Andrew Howe, Junlei Hu, Zhaoyang Jacopo Hu, Ria Jain, Mohammad Rafiee Javazm, Howard Ji, Rui Ji, Jianmin Ji, Zhongliang Jiang, Dominic Jones, Jeffrey Jopling, Britton Jordan, Ran Ju, Michael Kam, Luoyao Kang, Fausto Kang, Siddhartha Kapuria, Peter Kazanzides, Sonika Kiehler, Ethan Kilmer, Ji Woong Kim, Przemys{\l}aw Korzeniowski, Chandra Kuchi, Nithesh Kumar, Alan Kuntz, Federico Lavagno, Yu Chung Lee, Hao-Chih Lee, Hang Li, Zhen Li, Xiao Liang, Xinxin Lin, Jinsong Lin, Chang Liu, Fei Liu, Pei Liu, Yun-hui Liu, Wanli Liuchen, Eszter Luk\'acs, Sareena Mann, Miles Mannas, Brett Marinelli, Sabina Martyniak, Francesco Marzola, Lorenzo Mazza, Xueyan Mei, Maria Clara Morais, Luigi Muratore, Chetan Reddy Narayanaswamy, Micha{\l} Naskr\k{e}t, David Navarro-Alarcon, Cyrus Neary, Chi Kit Ng, Christopher Nguan, David Noonan, Ki Hwan Oh, Tom Christian Olesch, Allison M. Okamura, Justin Opfermann, Matteo Pescio, Doan Xuan Viet Pham, Tito Porras, Hongliang Ren, Ariel Rodriguez Jimenez, Ferdinando Rodriguez y Baena, Septimiu E. Salcudean, Asmitha Sathya, Preethi Satish, Lalithkumar Seenivasan, Jiaqi Shao, Yiqing Shen, Yu Sheng, Lucy XiaoYang Shi, Zoe Soul\'e, Stefanie Speidel, Mingwu Su, Jianhao Su, Idris Sunmola, Krist\'of Tak\'acs, Yunxi Tang, Patrick Thornycroft, Yu Tian, Jordan Thompson, Mehmet K. Turkcan, Mathias Unberath, Pietro Valdastri, Carlos Vives, Quan Vuong, Martin Wagner, Farong Wang, Wei Wang, Lidian Wang, Chung-Pang Wang, Guankun Wang, Junyi Wang, Erqi Wang, Ziyi Wang, Tanner Watts, Wolfgang Wein, Yimeng Wu, Zijian Wu, Hongjun Wu, Luohong Wu, Jie Ying Wu, Junlin Wu, Victoria Wu, Kaixuan Wu, Mateusz W\'ojcikowski, Yunye Xiao, Nan Xiao, Wenxuan Xie, Hao Yang, Tianqi Yang, Yinuo Yang, Menglong Ye, Ryan S. Yeung, Nural Yilmaz, Chim Ho Yin, Michael Yip, Rayan Younis, Chenhao Yu, Sayem Nazmuz Zaman, Milos Zefran, Han Zhang, Yuelin Zhang, Yidong Zhang, Yanyong Zhang, Xuyang Zhang, Yameng Zhang, Joyce Zhang, Ning Zhong, Peng Zhou, Haoying Zhou, Xiuli Zuo, Nassir Navab, Mahdi Azizian, Sean D. Huver, Axel Krieger 4/30/2026

Open-H-Embodiment: A Large-Scale Dataset for Enabling Foundation Models in Medical Robotics

Large-scale open medical robotics dataset enabling foundation models for autonomous medical robots with multi-embodiment coverage.

Ax Shevya Panda, Shinjini Bose, Ananya Joshi 4/30/2026

Reliability Auditing for Downstream LLM tasks in Psychiatry: LLM-Generated Hospitalization Risk Scores

Reliability auditing methodology for LLM outputs in psychiatry risk assessment, addressing algorithmic biases and prompt sensitivity.

Ax Longju Bai, Zhemin Huang, Xingyao Wang, Jiao Sun, Rada Mihalcea, Erik Brynjolfsson, Alex Pentland, Jiaxin Pei 4/30/2026

How Do AI Agents Spend Your Money? Analyzing and Predicting Token Consumption in Agentic Coding Tasks

Systematic study of token consumption in AI coding agents, analyzing where tokens are spent, model efficiency comparisons, and token usage prediction.

Ax Niels Bracher, Xavier Intes, Stefan T. Radev 4/30/2026

Inverting Foundation Models of Brain Function with Simulation-Based Inference

Using foundation models of brain activity with LLMs for stimulus reconstruction via simulation-based inference in neuroscience applications.

Ax Jiaqi Wang, Wenhao Zhang, Weijie Shi, Yaliang Li, James Cheng 4/30/2026

TCOD: Exploring Temporal Curriculum in On-Policy Distillation for Multi-turn Autonomous Agents

Temporal curriculum approach for on-policy distillation in multi-turn autonomous agents, addressing trajectory-level KL instability in reasoning transfer.

Ax Grigorios Papanikolaou, Ioannis Kontopoulos, Konstantinos Tserpes 4/30/2026

A Comparative Analysis on the Performance of Upper Confidence Bound Algorithms in Adaptive Deep Neural Networks

Comparison of upper confidence bound algorithms in adaptive deep neural networks for edge computing with latency and energy constraints.

Ax Orhan Demirci, Sezer Aptourachman, Ayd{\i}n Kaya 4/30/2026

ADE: Adaptive Dictionary Embeddings -- Scaling Multi-Anchor Representations to Large Language Models

Multi-anchor word embeddings scaled to large language models with computational efficiency, improving semantic expressiveness for polysemous words.

Ax Ming Li, Jie Wu, Justin Cui, Xiaojie Li, Rui Wang, Chen Chen 4/30/2026

ViPO: Visual Preference Optimization at Scale

Visual preference optimization framework for scaling generative models, addressing noise in conflicting preference datasets for robust training.

Ax Joshua Sherwood, Ben Aybar, Benjamin Kaplan 4/30/2026

Frontier Coding Agents Can Now Implement an AlphaZero Self-Play Machine Learning Pipeline For Connect Four That Performs Comparably to an External Solver

Benchmark measuring frontier AI coding agents' capability to autonomously implement end-to-end ML pipelines from minimal task descriptions, using AlphaZero Connect Four as test case.

Ax Li Ju, Junzhe Wang, Qi Zhang 4/30/2026

Faithfulness-QA: A Counterfactual Entity Substitution Dataset for Training Context-Faithful RAG Models

Dataset of 99,094 QA samples for training RAG models to prioritize retrieved context over internal knowledge, addressing faithfulness issues in retrieval-augmented generation.

Ax Ma Zirui, Fan Zhihua, Li Wenxing, Wu Haibin, Zhang Fulin, Ye Xiaochun, Li Wenming 4/30/2026

AHASD: Asynchronous Heterogeneous Architecture for LLM Adaptive Drafting Speculative Decoding on Mobile Devices

Asynchronous heterogeneous architecture for adaptive speculative decoding of LLMs on mobile single-NPU devices with reduced idle and wasted computation.

Ax Catherine Ning, Yu Ma, Cindy Beini Wang, Sean McMahon, Joseph Radojevic, Steven Zweibel, Dimitris Bertsimas 4/30/2026

A Multimodal and Explainable Machine Learning Approach to Diagnosing Multi-Class Ejection Fraction from Electrocardiograms

Multimodal ML framework combining ECG and EHR data to classify left ventricular ejection fraction with explainability for clinical use.

Ax Valentin Cuzin-Rambaud (LIRIS, UCBL), Laetitia Matignon (LIRIS, UCBL), Maxime Morge (LIRIS, UCBL) 4/30/2026

A Survey of Multi-Agent Deep Reinforcement Learning with Graph Neural Network-Based Communication

Survey of multi-agent deep reinforcement learning methods using GNN-based communication for agent coordination and information sharing.

Ax Jiaming Yang, Chenwei Tang, Liangli Zhen, Jiancheng Lv 4/30/2026

Rethinking KV Cache Eviction via a Unified Information-Theoretic Objective

Information-theoretic approach to KV cache eviction in LLM inference using Information Bottleneck principle to optimize long-context generation.

Ax Kieran Maguire, Srinandan Dasmahapatra 4/30/2026

Mini-Batch Class Composition Bias in Link Prediction

Shows GNN link prediction models learn representations inconsistent with node classification due to mini-batch class composition bias.

Ax Marta Ziosi, Miro Plueckebaum, Stephen Casper, Henry Papadatos, Ze Shen Chin, Peter Slattery, James Gealy, Tim G. J. Rudner, Brian Tse, Ariel Gil, Patricia Paskov, Maximilian Negele, Rokas Gipi\v{s}kis, Nada Madkour, Vera Lummis, Rupal Jain, Luise Eder, Kristina Fort, Malou C. van Draanen Glismann, In\`es Belhadj, Amin Oueslati, Anna K. Wisakanto, Richard Mallah, Koen Holtman, Ranj Zuhdi, Daniel S. Schiff, Jessica Newman, Malcolm Murray, Robert Trager 4/30/2026

Open Problems in Frontier AI Risk Management

Survey of open problems in frontier AI risk management, examining misalignments between emerging safety practices and established risk frameworks.

Ax Taylor Maxson, Roberto Corizzo, Yaning Wu, Nathalie Japkowicz, Colin Bellinger 4/30/2026

Correcting Performance Estimation Bias in Imbalanced Classification with Minority Subconcepts

Method for detecting and correcting performance estimation bias in imbalanced classification across minority subconcepts without explicit subconcept labels.

Ax Vyom Sharma, Debajyoti Datta 4/30/2026

RaMP: Runtime-Aware Megakernel Polymorphism for Mixture-of-Experts

Runtime-aware dispatch framework for Mixture-of-Experts inference that dynamically selects optimal kernel configurations based on batch size and routing patterns.

Ax Jennifer Wendland, Nicolas Freitag, Maik Kschischo 4/30/2026

Observable Neural ODEs for Identifiable Causal Forecasting in Continuous Time

Continuous-time causal inference in sequential decision problems using Observable Neural ODEs. Links control-theoretic observability to causal identifiability.

Ax Teetat Pipattaratonchai, Aueaphum Aueawatthanaphisut 4/30/2026

Privacy-Preserving Federated Learning Framework for Distributed Chemical Process Optimization

Federated learning framework for privacy-preserving chemical process optimization across distributed industrial facilities.

Ax Karim Alghoul, Hussein Al Osman, Abdulmotaleb El Saddik 4/30/2026

PPG-Based Affect Recognition with Long-Range Deep Models: A Measurement-Driven Comparison of CNN, Transformer, and Mamba Architectures

Compares CNN, Transformer, and Mamba architectures for PPG-based emotion recognition in wearable devices.

Ax Jiahong Wang, Logan Numerow, Stelian Coros, Christian Theobalt, Vahid Babaei, Bernhard Thomaszewski 4/30/2026

Momentum-Conserving Graph Neural Networks for Deformable Objects

Novel GNN architecture (MomentumGNN) for modeling deformable materials that preserves physical conservation laws.

Ax Mohammed Suhail B Nadaf 4/30/2026

reward-lens: A Mechanistic Interpretability Library for Reward Models

Open-source mechanistic interpretability library for reward models used in RLHF training. Ports logit lens and related tools to scalar reward heads.

Ax Abhirami Pillai 4/30/2026

Budget-Constrained Causal Bandits: Bridging Uplift Modeling and Sequential Decision-Making

Causal bandit framework for budget-constrained treatment allocation in advertising, combining heterogeneous treatment effect estimation with sequential decision-making.

Ax Wenshuo Zhao, Qi Zhu, Xingshan Zeng, Fei Mi, Lifeng Shang, Yiren Feng 4/30/2026

Entropy Centroids as Intrinsic Rewards for Test-Time Scaling

Intrinsic reward signal using entropy centroids for test-time scaling in LLMs, enabling efficient best-of-N sampling without external reward models.

Ax Jason Wu, Shir-Kang Scott Jinn, Yuyang Yuan, Maggie Wigness, Lance M. Kaplan, Hang Qiu, Mani Srivastava 4/30/2026

SWAN: World-Aware Adaptive Multimodal Networks for Runtime Variations

Adaptive multimodal network architecture handling runtime variations in modality quality, input complexity, and computational resource availability.

Ax Panyi Dong, Zhiyu Quan 4/30/2026

Efficient and Interpretable Transformer for Counterfactual Fairness

Transformer-based model balancing predictive performance, interpretability, and fairness constraints for high-stakes regulated domains.

Ax Yuhan Wang, Ruobing Yan, Zhe Su, Hejing Chen, Ningjing Sang, Yunfei Nie 4/30/2026

Unsupervised Graph Modeling for Anomaly Detection in Accounting Subject Relationships

Graph neural network framework for unsupervised anomaly detection in accounting subject relationships within general ledger structures.

Ax Tianhao Hu, Xiangcheng Liu, Youshao Xiao, Yang Zheng, Xuan Huang, Jinrui Ding, Yufei Zhang, Tao Liang, Hongyu Zang, Quan Chen, Yueqing Sun, Wenjie Shi, Chao Zhang, Wei Wang, Qi Gu, Yerui Sun, Yucheng Xie, Xunliang Cai 4/30/2026

DORA: A Scalable Asynchronous Reinforcement Learning System for Language Model Training

Asynchronous RL system for LLM post-training that overlaps generation with training to resolve bottlenecks from long-tailed trajectory generation.

Ax Douglas Jiang, Yuechen Wang, Jiayi Wang, Jiaying Geng, Qinglong Wang, Feng Tian 4/30/2026

NeuroPlastic: A Plasticity-Modulated Optimizer for Biologically Inspired Learning Dynamics

Novel optimizer inspired by synaptic plasticity that augments gradient-based updates with adaptive multi-signal modulation for deep learning.

Ax Mengyang Zhao, Longlong Li, Cunquan Qu 4/30/2026

Cheeger--Hodge Contrastive Learning for Structurally Robust Graph Representation Learning

Graph contrastive learning framework using Cheeger-Hodge signatures for robust graph representation learning under structural perturbations.

Ax Bolian Li, Yifan Wang, Yi Ding, Anamika Lochab, Ananth Grama, Ruqi Zhang 4/30/2026

Addressing Performance Saturation for LLM RL via Precise Entropy Curve Control

Method for preventing entropy collapse and performance saturation in LLM reinforcement learning through precise entropy curve control.

Ax Weihang Li, Jianchun Liu, Hongli Xu 4/30/2026

Adaptive and Fine-grained Module-wise Expert Pruning for Efficient LoRA-MoE Fine-Tuning

Parameter-efficient fine-tuning approach combining LoRA with adaptive expert pruning for mixture-of-experts models across heterogeneous transformer modules.

Ax Disha Singha 4/30/2026

Uncertainty-Aware Reward Discounting for Mitigating Reward Hacking

Reinforcement learning method addressing reward hacking through uncertainty-aware reward discounting to handle uncertain and inconsistent human preferences.

Ax Zhe Ding, Su Pan, Duowei Pan 4/30/2026

CoQuant: Joint Weight-Activation Subspace Projection for Mixed-Precision LLMs

Mixed-precision post-training quantization method for LLMs using joint weight-activation subspace projection to preserve critical model components.

Ax Mathieu Dario, Florent Chenevier, K\'evin Delmas, Joris Guerin, J\'er\'emie Guiochet 4/30/2026

Unifying Runtime Monitoring Approaches for Safety-Critical Machine Learning: Application to Vision-Based Landing

Unified framework for runtime monitoring of ML in safety-critical systems, categorizing approaches for ODD monitoring, OOD detection, and prediction confidence.

Ax Yongliang Ding, Qigong Bi, Peng Pu 4/30/2026

STLGT: A Scalable Trace-Based Linear Graph Transformer for Tail Latency Prediction in Microservices

Scalable graph transformer model for predicting tail latency in microservices using trace-based span graph encoding.

Ax Aleksander Tankman 4/30/2026

Layer-wise Lipschitz-Product Control for Deep Kolmogorov--Arnold Network Representations of Compositionally Structured Functions

Theoretical framework proving Kolmogorov-Arnold Networks can represent compositionally structured functions with controlled Lipschitz properties.

Ax Jizhou Huang, Brendan Juba 4/30/2026

Near-Optimal Cryptographic Hardness of Learning With Homogeneous Halfspaces Under Gaussian Marginals

Theoretical analysis of cryptographic hardness for learning homogeneous halfspaces under Gaussian distributions with applications to agnostic learning and fairness.