Ax Leonard B\"armann, Joana Plewnia, Alex Waibel, Tamim Asfour 5/7/2026

Learning to Forget -- Hierarchical Episodic Memory for Lifelong Robot Deployment

Hierarchical episodic memory framework enabling humanoid robots to learn selective memory retention based on user-defined relevance.

Ax Yubai Wei, Chen Wu, Hashem Haghbayan 5/7/2026

Can Explicit Physical Feasibility Benefit VLA Learning? An Empirical Study

Empirical study evaluating whether explicit physical feasibility constraints improve vision-language-action robot policy learning.

Ax Hongyao Liu, Liuqun Zhai, Junyi Wang, Zhengru Fang 5/7/2026

SparKV: Overhead-Aware KV Cache Loading for Efficient On-Device LLM Inference

Framework for efficient on-device LLM inference combining cloud KV streaming with local computation to reduce memory and latency overhead.

Ax Griffin Pitts, Muntasir Hoq, Peter Brusilovsky, Narges Norouzi, Arto Hellas, Juho Leinonen, Bita Akram 5/7/2026

Personalized Worked Example Generation from Student Code Submissions Using Pattern-based Knowledge Components

LLM-based system generating personalized worked examples from student code submissions to address specific logical errors.

Ax Xue Qin, Simin Luan, John See, Zeyd Boukhers, Cong Yang, Zhijun Li 5/7/2026

Atomic-Probe Governance for Skill Updates in Compositional Robot Policies

Framework for analyzing and updating skill libraries in robotic systems during deployment through paired-sampling protocol.

Ax Guang Yang, Xing Hu, Xiang Chen, Xin Xia 5/7/2026

From Mirage to Grounding: Towards Reliable Multimodal Circuit-to-Verilog Code Generation

Multimodal LLM approach for translating circuit diagrams into Verilog code, addressing reliability and safety-critical hardware design translation.

Ax Daniel Zhu, Zihan Wang, Xuchan Bao, Jerry Wei 5/7/2026

Jailbroken Frontier Models Retain Their Capabilities

Analysis showing advanced jailbreaks on frontier LLMs scale with model capability and increasingly impose minimal performance degradation ('jailbreak tax').

Ax Sheng Ran 5/7/2026

Endogenous Regime Switching Driven by Scalar-Irreducible Learning Dynamics

Framework for endogenous regime switching in machine learning using scalar-irreducible dynamics to enable autonomous intelligence without external transitions.

Ax JiangBo Zhao, ZhaoXin Liu 5/7/2026

A Self-Attentive Meta-Optimizer with Group-Adaptive Learning Rates and Weight Decay

MetaAdamW: Adaptive optimizer with self-attention mechanism to dynamically modulate per-group learning rates and weight decay across heterogeneous parameter groups.

Ax Takayuki Komatsu, Yoshiyuki Ohmura, Yasuo Kuniyoshi 5/7/2026

Transformation Categorization Based on Group Decomposition Theory Using Parameter Division

Study of unsupervised transformation categorization using group decomposition theory to define 'good' representations beyond classical independent factor disentanglement.

Ax Zhen Liu, Yuhan Liu, Jingwen Fu 5/7/2026

Structured Progressive Knowledge Activation for LLM-Driven Neural Architecture Search

LLM-driven Neural Architecture Search framework using structured knowledge activation to guide architecture exploration while avoiding non-local behavioral shifts.

Ax Yutong Zhang, Zimeng Wu, Shangcai Liao, Shujiang Wu, Jiaxin Chen 5/7/2026

MP-ISMoE: Mixed-Precision Interactive Side Mixture-of-Experts for Efficient Transfer Learning

MP-ISMoE: Memory-efficient transfer learning approach using mixed-precision mixture-of-experts side networks for parameter-efficient foundation model adaptation.

Ax Nicolas Michel, Maorong Wang, Jiangpeng He, Toshihiko Yamasaki 5/7/2026

Continual Distillation of Teachers from Different Domains

Continual Distillation paradigm where students learn sequentially from teacher models without access to training data using external unlabeled data for knowledge transfer.

Ax Guoqiang Zhang, Kenta Niwa, W. Bastiaan Kleijn 5/7/2026

Lookahead Drifting Model

Lookahead drifting model extending drifting model paradigm for image generation with improved optimization trajectory and convergence.

Ax Bryan Cheng, Jasper Zhang 5/7/2026

Single-Position Intervention Fails: Distributed Output Templates Drive In-Context Learning

Mechanistic interpretability study showing distributed output templates drive in-context learning; single-position intervention fails to predict causal importance.

Ax Shu-Hao Zhang, Le-Tong Huang, Xiang-Sheng Deng, Xin-Yi Zou, Chen Wu, Nan Li, Shao-Qun Zhang 5/7/2026

EdgeRazor: A Lightweight Framework for Large Language Models via Mixed-Precision Quantization-Aware Distillation

EdgeRazor: Lightweight framework for deploying LLMs on resource-constrained devices via mixed-precision quantization-aware distillation combining PTQ and QAT techniques.

Ax Jacob Thrasher, Kaitlyn Heintzelman, Peter Martone, David Kotlowski, Binod Bhattarai, Donald Adjeroh, Prashnna Gyawali 5/7/2026

Investigating Trustworthiness of Nonparametric Deep Survival Models for Alzheimer's Disease Progression Analysis

Evaluation of deep learning survival models for Alzheimer's disease progression analysis with focus on trustworthiness and reliability for clinical applications.

Ax Halil Ibrahim Gulluk, Olivier Gevaert 5/7/2026

Improving Medical VQA through Trajectory-Aware Process Supervision

Medical VQA framework using COMCTS algorithm to generate reasoning trajectories with open-source vision-language models for improved reasoning and interpretability.

Ax Bilel Abderrahmane Benziane, Benoit Lardeux, Ayoub Mcharek, Maher Jridi 5/7/2026

Designing a double deep reinforcement learning selection tool for resilient demand prediction

Double deep reinforcement learning approach for automated selection of forecasting solutions in supply chain demand prediction with dataset-specific features.

Ax Gregory Magarshak 5/7/2026

LAWS: Learning from Actual Workloads Symbolically -- A Self-Certifying Parametrized Cache Architecture for Neural Inference, Robotics, and Edge Deployment

LAWS: Self-certifying inference caching architecture with formal error bounds for neural networks on edge devices and robotics using Probabilistic Language Tries.

Ax Chris Sainsbury, Feng Dong, Andreas Karwath 5/7/2026

FlatASCEND: Autoregressive Clinical Sequence Generation with Continuous Time Prediction and Association-Based Pharmacological Testing

FlatASCEND: 14.5M autoregressive clinical sequence model with continuous time prediction and pharmacological association testing for patient trajectory generation.

Ax Chris Sainsbury, Feng Dong, Andreas Karwath 5/7/2026

Sparse Autoencoder Decomposition of Clinical Sequence Model Representations: Feature Complexity, Task Specialisation, and Mortality Prediction

Sparse Autoencoder decomposition applied to clinical sequence models (FlatASCEND) to interpret learned representations and their task specialization for mortality prediction.

Ax Cor Steging, Tadeusz Zbiegie\'n 5/7/2026

Confronting Label Indeterminacy in Automated Bail Decisions

Analysis of label indeterminacy challenges in automated bail decision systems where counterfactual outcomes remain unobserved, risking bias propagation.

Ax Mohammad AlShaikh Saleh, Sanjay Chawla, Sertac Bayhan, Haitham Abu-Rub, Ali Ghrayeb 5/7/2026

A Physics-Aware Framework for Short-Term GPU Power Forecasting of AI Data Centers

Physics-informed DLinear model for short-term GPU power forecasting in AI data centers handling heterogeneous inference and training workloads.

Ax Sihao Liu, YuFan Xiong, Zhonghua Jiang, Zhaode Wang, chengfei lv Shengyu Zhang 5/7/2026

RetentiveKV: State-Space Memory for Uncertainty-Aware Multimodal KV Cache Eviction

RetentiveKV: KV cache eviction method for multimodal LLMs using uncertainty-aware state-space memory to handle deferred importance of visual tokens in long contexts.

Ax Zhiyuan Zeng, Jiameng Huang, Zhangyue Yin, Jiashuo Liu, Ziniu Li, Bingrui Li, Yuhao Wu, Yining Zheng, Ge Zhang, Wenhao Huang, Xipeng Qiu 5/7/2026

Balanced Aggregation: Understanding and Fixing Aggregation Bias in GRPO

arXiv: Balanced Aggregation fixes aggregation bias in GRPO reinforcement learning for LLM reasoning and code generation.

Ax Khouloud Saadi, Di Wang 5/7/2026

Validity-Calibrated Reasoning Distillation

arXiv: Validity-Calibrated Reasoning Distillation transfers multi-step reasoning from large to smaller LLMs using global correctness constraints.

Ax Wu Wang, Yuang Cheng, Fouzi Harrou, Ying Sun 5/7/2026

Efficient Handwriting-Based Alzheimer,s Disease Diagnosis Using a Low-Rank Mixture of Experts Deep Learning Framework

arXiv: LoRA-MoE deep learning framework for Alzheimer's disease diagnosis using handwriting analysis as digital biomarker.

Ax Mohammad Raeisi Gahrouei, Pedram Ramin, Vincenzo A. Riggio, Carlos Domingo-Felez 5/7/2026

Enhancing the interpretability of spatially variable N2O model predictions with soft sensors during wastewater treatment

arXiv: ML models for predicting N2O emissions in wastewater treatment with interpretability analysis using soft sensors.

Ax Tadhg Looram, Lucas Nuzzi, Kyle Waters, Steven Dillmann 5/7/2026

AsymmetryZero: A Framework for Operationalizing Human Expert Preferences as Semantic Evals

arXiv: AsymmetryZero framework operationalizing human expert preferences as semantic evaluations for RL reward signals.

Ax Ye Qiao, Yian Wang, Zhiheng Chen, Hyoukjun Kwon, Sitao Huang 5/7/2026

FASQ: Flexible Accelerated Subspace Quantization for Calibration-Free LLM Compression

arXiv: FASQ calibration-free LLM compression via product quantization on weight matrices with flexible bit-width tuning.

Ax Matteo Gambella, Fabrizio Pittorino, Manuel Roveri 5/7/2026

HERCULES: Hardware-Efficient, Robust, Continual Learning Neural Architecture Search

arXiv: HERCULES framework for hardware-aware neural architecture search balancing efficiency, robustness, and continual learning.

Ax Yoav Ger, Omri Barak 5/7/2026

Learning reveals invisible structure in low-rank RNNs

arXiv research extending low-rank RNN framework to learning dynamics, deriving gradient descent in reduced overlap space.

Ax Hanfei Zhou, Lei Shi 5/7/2026

Simultaneous CNN Approximation on Manifolds with Applications to Boundary Value Problems

arXiv research on CNN approximation methods for functions on Riemannian manifolds with boundary value problem applications.

Ax Devon Jarvis, Richard Klein, Benjamin Rosman, Steven James, Stefano Sarao Mannelli 5/7/2026

Position: the Stochastic Parrot in the Coal Mine. Model Collapse is a Threat to Low-Resource Communities

Position paper on model collapse threat: generative models trained on AI outputs degrade performance, disproportionately harming low-resource communities.

Ax Sreejata Dey, Guoxiang Grayson Tong, Jonathan F. MacArt, Daniele E. Schiavazzi 5/7/2026

Model synthesis and identifiability analysis of stiff chemical reaction systems with inVAErt networks

inVAErt networks for learning efficient emulators of stiff chemical reaction systems using conditional neural networks.

Ax Qijun Liao, Zhaoxin Yu, Jue Yang 5/7/2026

Constraint-Enhanced Reinforcement Learning Based on Dynamic Decoupled Spherical Radial Squashing

Constraint-enhanced RL method handling heterogeneous per-joint actuator rate limits for robotic control.

Ax Alexander I. Khrabry, Edward A. Startsev, Andrew T. Powis, Igor D. Kaganovich 5/7/2026

Deep Wave Network for Modeling Multi-Scale Physical Dynamics

Deep Wave Network architecture for modeling multi-scale physical dynamics with systematic width-depth scaling analysis.

Ax Michael Rottoli, Subhankar Roy, Stefano Paraboschi 5/7/2026

Predict-then-Diffuse: Adaptive Response Length for Compute-Budgeted Inference in Diffusion LLMs

Predict-then-Diffuse: adaptive response length method for compute-budgeted inference in diffusion-based LLMs.

Ax Yaobo Zhang 5/7/2026

Jordan-RoPE: Non-Semisimple Relative Positional Encoding via Complex Jordan Blocks

Jordan-RoPE: non-semisimple relative positional encoding for transformers using complex Jordan blocks.

Ax Seifallah Fetni, Thinh Quy Duc Pham, Truong Vinh Hoang, Hoang Son Tran, Laurent Duch\^ene, Xuan-Van Tran, Anne Marie Habraken 5/7/2026

Capabilities of Auto-encoders and Principal Component Analysis of the Reduction of Microstructural Images; Application on the Acceleration of Phase-Field Simulations

Data-driven framework using autoencoders and PCA for dimensionality reduction of microstructural simulation images.

Ax Simon Dufort-Labb\'e, Pierre-Luc Bacon, Razvan Pascanu, Simon Lacoste-Julien, Aristide Baratin 5/7/2026

Layerwise LQR for Geometry-Aware Optimization of Deep Networks

Layerwise LQR framework for geometry-aware optimization of deep networks preserving cross-layer interactions.

Ax Roberto Medina 5/7/2026

Adaptive Consensus in LLM Ensembles via Sequential Evidence Accumulation: Automatic Budget Identification and Calibrated Commit Signals

DASE: adaptive stopping heuristic for LLM ensembles that detects performance boundaries and calibrates commit signals.

Ax Senne Deproost, Mehrdad Asadi, Ann Now\'e 5/7/2026

Hierarchical Support Vector State Partitioning for Distilling Black Box Reinforcement Learning Policies

Method to distill black box RL policies into interpretable subpolicies using support vector partitioning.

Ax Etienne Gauthier, Francis Bach, Michael I. Jordan 5/7/2026

Explaining and Preventing Alignment Collapse in Iterative RLHF

Analysis of alignment collapse in iterative RLHF showing how policy-generated data retraining creates feedback loops that degrade alignment.

Ax Tarun Kathuria, Sachin Kumar 5/7/2026

Leveraging Pretrained Language Models as Energy Functions for Glauber Dynamics Text Diffusion

Discrete diffusion language model using Glauber dynamics with pretrained LM energy functions for text generation.

Ax Hamed Karimi, Vaishali Meyappan, Reza Samavi 5/7/2026

LLMs Uncertainty Quantification via Adaptive Conformal Semantic Entropy

Adaptive conformal semantic entropy method for uncertainty quantification and hallucination detection in LLMs.

Ax Kaustubh Pethkar, Ziyang Xiong, Zuofeng Shang, Yingcong Li 5/7/2026

Memory as a Markov Matrix: Sample Efficient Knowledge Expansion via Token-to-Dictionary Mapping

Token-to-dictionary mapping method for continual knowledge incorporation in LLMs while mitigating catastrophic forgetting.

Ax Nicholas J. Cooper, Fran\c{c}ois G. Meyer, Michael L. Roberts, Carlos Zapata-Carratal\'a, Lijun Chen, Danna Gurari 5/7/2026

On the Architectural Complexity of Neural Networks

Theoretical framework for analyzing and constructing deep neural networks with explicit tensor operation structure modeling.

Ax Mohammed Sabry, Anya Belz 5/7/2026

Budgeted LoRA: Distillation as Structured Compute Allocation for Efficient Inference

Budgeted LoRA distillation framework for efficient LLM inference with structured compute allocation.