Ax Francesco Daghero, Mahyar Tourchi Moghaddam, Mikkel Baun Kj{\ae}rgaard 4/30/2026

Hierarchical adaptive control for real-time dynamic inference at the edge

Framework for deploying dynamic ML models on resource-constrained edge devices with hierarchical adaptive control for latency, energy, and memory optimization.

Ax Jiancheng Wang, Mingjia Yin, Hao Wang, Enhong Chen 4/30/2026

Understanding DNNs in Feature Interaction Models: A Dimensional Collapse Perspective

Analysis of how DNNs capture feature interactions in recommendation systems from a dimensional collapse perspective, addressing theoretical limitations in learning high-order interactions.

Ax Jinjiang Guo 4/30/2026

Do Larger Models Really Win in Drug Discovery? A Benchmark Assessment of Model Scaling in AI-Driven Molecular Property and Activity Prediction

Benchmark study comparing molecular foundation models and LLMs against specialized cheminformatics models for drug discovery tasks across 22 property/activity endpoints.

Ax Cyril Shih-Huan Hsu, Wig Yuan-Cheng Cheng, Chrysa Papagianni 4/30/2026

Progressive Semantic Communication for Efficient Edge-Cloud Vision-Language Models

Edge-cloud collaborative architecture for deploying vision-language models with progressive semantic communication to balance computational demands and bandwidth constraints.

Ax Seungyub Han, Hyungjin Kim, Jungwoo Lee 4/30/2026

Lyapunov-Guided Self-Alignment: Test-Time Adaptation for Offline Safe Reinforcement Learning

Transformer framework for test-time adaptation in offline safe reinforcement learning using self-alignment to generate and select safe trajectories without retraining.

Ax Daniele Calandriello, Alessandro Lazaric, Michal Valko 4/30/2026

Large-scale semi-supervised learning with online spectral graph sparsification

arXiv Labs framework description for collaborative feature development on the arXiv platform.

Ax Zhiquan Tan, Yinrong Hong 4/30/2026

PAINT: Partial-Solution Adaptive Interpolated Training for Self-Distilled Reasoners

Self-distillation training method for LLM reasoning with adaptive partial solutions using on-policy exploration and token-level supervision.

Ax Marcus Haywood-Alexander, Gregory Duth\'e, Eleni Chatzi 4/30/2026

PiGGO: Physics-Guided Learnable Graph Kalman Filters for Virtual Sensing of Nonlinear Dynamic Structures under Uncertainty

Physics-guided graph Kalman filter combining physics-based models and neural networks for nonlinear structure state estimation.

Ax Gota Morishita 4/30/2026

Who Trains Matters: Federated Learning under Enrollment and Participation Selection Biases

Analysis of selection bias in federated learning when client enrollment and participation are non-representative of target population.

Ax Zhangzhi Xiong, Haoyi Wu, You Wu, Shuqi Gu, Kan Ren, Kewei Tu 4/30/2026

Exploring the Potential of Probabilistic Transformer for Time Series Modeling: A Report on the ST-PT Framework

Probabilistic Transformer framework showing equivalence to mean-field variational inference on conditional random fields for time series.

Ax Hayate Iso, Tiyasa Mitra, Sudipta Mondal, Rasoul Shafipour, Venmugil Elango, Terry Kong, Yuki Huang, Seonjin Na, Izzy Putterman, Benjamin Chislett, Maor Ashkenazi, Joseph Guman, Gerald Shen, Tugrul Konuk, Ashwath Aithal, Ritika Borkar, Ran Zilberstein, Bita Rouhani 4/30/2026

Accelerating RL Post-Training Rollouts via System-Integrated Speculative Decoding

Speculative decoding method accelerating RL post-training rollouts for frontier language models without changing optimization regime.

Ax Ethan Harvey, Dennis Johan Loevlie, Amir Ali Satani, Wansu Chen, David M. Kent, Michael C. Hughes 4/30/2026

A Multi-Dataset Benchmark of Multiple Instance Learning for 3D Neuroimage Classification

Benchmark comparing multiple instance learning versus 3D CNNs for brain scan classification from CT and MRI data.

Ax Zhaoyuan Cai, Xinglin Zhang 4/30/2026

Asynchronous Federated Unlearning with Invariance Calibration for Medical Imaging

Asynchronous federated unlearning framework with invariance calibration enabling data removal from trained models in medical imaging.

Ax Javier Gil Bl\'azquez 4/30/2026

Random Cloud: Finding Minimal Neural Architectures Without Training

Training-free neural architecture search method discovering minimal feedforward networks via stochastic exploration without backpropagation.

Ax Bernd Frauenknecht, Lukas Kesper, Daniel Mayfrank, Henrik Hose, Sebastian Trimpe 4/30/2026

Uncertainty-Aware Predictive Safety Filters for Probabilistic Neural Network Dynamics

Probabilistic neural network safety filters for reinforcement learning using ensemble models for constraint satisfaction during exploration.

Ax Zihan Zhao, Baotong Lu, Shengjie Lin, Yizou Chen, Jing Liu, Yanqi Zhang, Ziming Miao, Ming-Chang Yang, Haiying Shen, Qi Chen, Fan Yang 4/30/2026

Unifying Sparse Attention with Hierarchical Memory for Scalable Long-Context LLM Serving

Method unifying sparse attention and hierarchical memory for efficient long-context LLM serving with reduced KV cache costs.

Ax Bao Pham, Mohammed J. Zaki, Luca Ambrogioni, Dmitry Krotov, Matteo Negri 4/30/2026

Language Diffusion Models are Associative Memories Capable of Retrieving Unseen Data

Study showing discrete diffusion models function as associative memories with emergent creative capabilities for data retrieval and generation.

Ax Janis Mohr, J\"org Frochte 4/30/2026

Multiple Additive Neural Networks for Structured and Unstructured Data

Multiple Additive Neural Networks methodology extending gradient boosting with CNNs and capsule networks for structured and unstructured data.

Ax Evangelia Kopadi, Dimitris Kalles 4/30/2026

Causal Learning with Neural Assemblies

Research on neural assemblies learning causal directionality between variables, extending prior work on classification and planning.

Ax Liu Xiao 4/30/2026

Associative-State Universal Transformers: Sparse Retrieval Meets Structured Recurrence

UniMatrix architecture combining structured recurrence with sparse retrieval for language modeling using token-conditioned embeddings and hybrid state updates.

Ax Chao Jiang, Dugang Liu, Cheng Wen, Zhiwu Xu, Hua Zheng, Muhammad Sadiq, Jawwad Ahmed Shamsi, Shengchao Qin, Zhong Ming 4/30/2026

Large Language Models for Multilingual Code Intelligence: A Survey

Survey of large language models for multilingual code intelligence covering code generation and translation across diverse programming languages.

Ax Ronald Deng, Samuel McCauley, Aidin Niaparast, Helia Niaparast, Bennett Ptak, Shirel Quintanilla, Shikha Singh, Nathan Vosburg 4/30/2026

Incremental Strongly Connected Components with Predictions

Learned data structure for incremental strongly connected components problem using machine learning predictions for beyond-worst-case algorithm design.

Ax Zhongkai Yu, Haotian Ye, Chenyang Zhou, Ohm Rishabh Venkatachalam, Zaifeng Pan, Zhengding Hu, Junsung Kim, Won Woo Ro, Po-An Tsai, Shuyi Pei, Yangwook Kang, Yufei Ding 4/30/2026

AMMA: A Multi-Chiplet Memory-Centric Architecture for Low-Latency 1M Context Attention Serving

Hardware architecture for low-latency LLM serving with 1M token context, optimized for memory-bound decode-phase attention operations.

Ax Baturay Saglam, Dionysis Kalogerias 4/30/2026

Test-Time Safety Alignment

Test-time safety alignment method using input word embeddings as control variables to steer aligned LLM outputs toward desired safety properties.

Ax Zhentao Xu, Shangjing Zhang, Emir Poyraz, Yvonne Li, Ye Jin, Xie Lu, Xiaoyang Gu, Karthik Ramgopal, Praveen Kumar Bodigutla, Xiaofeng Wang 4/30/2026

Hierarchical Long-Term Semantic Memory for LinkedIn's Hiring Agent

LinkedIn's hierarchical long-term semantic memory system for LLM agents, extracting and retrieving context from behavioral data for personalized interactions.

Ax Jeremy Nixon, Annika Singh 4/30/2026

OMEGA: Optimizing Machine Learning by Evaluating Generated Algorithms

OMEGA end-to-end framework automates ML research by generating novel ML algorithms via meta-prompt engineering and code generation.

Ax Sk Tanzir Mehedi, Raja Jurdak, Chadni Islam, Abu Bakar Siddique Mahi, Gowri Ramachandran 4/30/2026

eDySec: A Deep Learning-based Explainable Dynamic Analysis Framework for Detecting Malicious Packages in PyPI Ecosystem

eDySec deep learning framework detects malicious packages in PyPI using explainable dynamic analysis of system behaviors.

Ax Hongyeon Yu, Young-Bum Kim, Yoon Kim 4/30/2026

FlowBot: Inducing LLM Workflows with Bilevel Optimization and Textual Gradients

FlowBot framework automatically induces LLM workflows using bilevel optimization and textual gradients to coordinate multi-step LLM calls.

Ax Chidera Biringa, Ajmal Abbas, Vishnu Selvaraj, Gokhan Kul 4/30/2026

VulStyle: A Multi-Modal Pre-Training for Code Stylometry-Augmented Vulnerability Detection

VulStyle multi-modal model for vulnerability detection combining source code, AST structure, and code stylometry features.

Ax Aditya Ukarande, Deep Shekhar, Marc Blackstein, Ram Rangan 4/30/2026

Efficient, VRAM-Constrained xLM Inference on Clients

Novel technique for efficient LLM/VLM inference on resource-constrained client devices using pipelined sharding and CPU-GPU hybrid scheduling.

Ax Zhirong Shen, Rui Huang, Jiacheng Liu, Chang Zou, Peiliang Cai, Shikang Zheng, Zhengyi Shi, Liang Feng, Linfeng Zhang 4/30/2026

Beyond Fixed Formulas: Data-Driven Linear Predictor for Efficient Diffusion Models

L2P learnable linear predictor framework for efficient feature caching acceleration in Diffusion Transformers.

Ax Qi Li (Key Lab of Smart Prevention and Mitigation of Civil Engineering Disasters of the Ministry of Industry and Information Technology, School of Civil Engineering, Harbin Institute of Technology, Harbin, 150090, China, Key Lab of Structures Dynamic Behavior and Control of the Ministry of Education, Harbin Institute of Technology, Harbin, 150090, China), Yong Huang (Key Lab of Smart Prevention and Mitigation of Civil Engineering Disasters of the Ministry of Industry and Information Technology, School of Civil Engineering, Harbin Institute of Technology, Harbin, 150090, China, Key Lab of Structures Dynamic Behavior and Control of the Ministry of Education, Harbin Institute of Technology, Harbin, 150090, China), Hui Li (Key Lab of Smart Prevention and Mitigation of Civil Engineering Disasters of the Ministry of Industry and Information Technology, School of Civil Engineering, Harbin Institute of Technology, Harbin, 150090, China, Key Lab of Structures Dynamic Behavior and Control of the Ministry of Education, Harbin Institute of Technology, Harbin, 150090, China) 4/30/2026

Probabilistic data quality assessment for structural monitoring data via outlier-resistant conditional diffusion model

Diffusion model approach for structural monitoring data quality assessment using conditional embeddings.

Ax Junwon You, Mihyun Jang, Sangwoo Mo, Jae-Hun Jung 4/30/2026

Topology-Aware Representation Alignment for Semi-Supervised Vision-Language Learning

Research on topology-aware alignment for semi-supervised vision-language model learning in specialized domains.

Ax Gabriel Stefan, Sergiu Nisioi 4/30/2026

SG-UniBuc-NLP at SemEval-2026 Task 6: Multi-Head RoBERTa with Chunking for Long-Context Evasion Detection

NLP system for SemEval task on detecting political question evasion using multi-head RoBERTa with chunking.

Ax Yimeng Shan, Zhaorui Zhang, Sheng Di, Yu Liu, Xiaoyi Lu, Benben Liu 4/30/2026

SplitFT: An Adaptive Federated Split Learning System For LLMs Fine-Tuning

Federated split learning approach for privacy-preserving LLM fine-tuning across distributed clients.

Ax Peter M\"ullner, Dominik Kowald, Markus Schedl, Elisabeth Lex 4/30/2026

Meta-Learning and Targeted Differential Privacy to Improve the Accuracy-Privacy Trade-off in Recommendations

Research on differential privacy and meta-learning to improve accuracy-privacy trade-offs in recommender systems.

Ax Romeo Valentin 4/30/2026

Recipes for Calibration Checks in Safety-Critical Applications

Research on calibration validation methods for probabilistic forecasters in safety-critical systems.

Ax Nikita Araslanov, Martin Sundermeyer, Hidenobu Matsuki, David Joseph Tan, Federico Tombari 4/30/2026

Featurising Pixels from Dynamic 3D Scenes with Linear In-Context Learners

Research on pixel-level feature learning in dynamic 3D scenes using in-context learning.

Ax Hanna Foerster, Ilia Shumailov, Cheng Zhang, Yiren Zhao, Jamie Hayes, Robert Mullins 4/30/2026

Quantamination: Dynamic Quantization Leaks Your Data Across the Batch

Research on data privacy vulnerabilities in dynamic quantization for ML inference serving.

Ax Mahnoor Shahid, Hannes Rothe 4/30/2026

AGEL-Comp: A Neuro-Symbolic Framework for Compositional Generalization in Interactive Agents

Neuro-symbolic agent architecture combining causal program graphs with LLMs to improve compositional generalization in interactive environments.

Ax Jinho Choo, JunSeung Lee, Jimyeong Kim, Yeeho Song, S. K. Hong, Yeong-Dae Kwon 4/30/2026

TLPO: Token-Level Policy Optimization for Mitigating Language Confusion in Large Language Models

Token-level policy optimization technique to mitigate language confusion in multilingual LLMs without degrading general capabilities.

Ax Tony Xu, Sarah Klamt, Katherine Turner, Anne Brustle, Felix Marsh-Wakefield, Givanna Putri 4/30/2026

FloatSOM: GPU-Accelerated, Distributed, Topology-Flexible Self-Organizing Maps

GPU-accelerated distributed Self-Organizing Maps framework supporting multi-GPU, disk-backed streaming, and flexible topologies.

Ax Gery Geenens, Pierre Lafaye de Micheaux, Ivan Muyun Zou 4/30/2026

Deep-testing: the case of dependence detection

Investigation of neural networks for hypothesis testing and dependence detection in statistical analysis.

Ax Stavros Orfanoudakis, Ziyan Li, Ruixiao Yang, Nikolay Aristov, Pedro P. Vergara, Chuchu Fan, Elenna Dugundji 4/30/2026

Learning to Route Electric Trucks Under Operational Uncertainty

Learning-based framework for electric truck routing under operational constraints including battery range and charging infrastructure.

Ax Dianyu Liu, Chuan Qin, Xi Chen, Xiaohan Li, Wenxi Xu, Yuyang Wang, Xin Chen, Yuanchun Zhou, Hengshu Zhu 4/30/2026

SciHorizon-DataEVA: An Agentic System for AI-Readiness Evaluation of Heterogeneous Scientific Data

Agentic system for evaluating AI-readiness of heterogeneous scientific data to improve ML model effectiveness in scientific workflows.

Ax Guiyi Zeng, Junqing Yu, Yi-Ping Phoebe Chen, Xu Chen, Wei Yang, Zikai Song 4/30/2026

CurEvo: Curriculum-Guided Self-Evolution for Video Understanding

Curriculum-guided self-evolution framework for autonomous video understanding without human annotations.

Ax Zhixin Han, Yanzhi Zhang, Chuyang Wei, Maohang Gao, Xiawei Yue, Kefei Chen, Yu Zhuang, Haoxiang Guan, Jiyan He, Jian Li, Yitong Duan, Yu Shi, Mengting Hu, Shuxin Zheng 4/30/2026

FutureWorld: A Live Environment for Training Predictive Agents with Real-World Outcome Rewards

Live environment for training LLM-based predictive agents using real-world outcome rewards for future prediction tasks.

Ax Manar Aljohani, Brandon Ho, Kenneth McKinley, Dennis Ren, Xuan Wang 4/30/2026

Domain-Adapted Small Language Models for Reliable Clinical Triage

Evaluation of open-source small language models for clinical triage decision-support with domain adaptation and privacy preservation.

Ax Mahya Ramezani, Holger Voos 4/30/2026

Rule-based High-Level Coaching for Goal-Conditioned Reinforcement Learning in Search-and-Rescue UAV Missions Under Limited-Simulation Training

Hierarchical framework combining rule-based high-level advisor with goal-conditioned reinforcement learning for UAV search-and-rescue missions.

Ax Dimitris Dimakopoulos, Shay B. Cohen, Ioannis Konstas 4/30/2026

MoRFI: Monotonic Sparse Autoencoder Feature Identification

Research on sparse autoencoders to identify mechanisms causing hallucinations in LLMs from supervised fine-tuning on new knowledge.