Isolater - Feed

Ax Laha Ale, Ning Zhang, Scott A. King, Pingzhi Fan 3/10/2026

Switchable Activation Networks

Switchable Activation Networks that dynamically select activation functions for computational efficiency in LLMs and vision-action models during inference.

Ax Xie Xiaohu, Liu Xiaohu, Yao Benjamin 3/10/2026

Know When You're Wrong: Aligning Confidence with Correctness for LLM Error Detection

Method to align LLM confidence scores with correctness using output token probabilities for reliable error detection and hallucination identification.

Ax Joseph Bingham, Noah Green, Saman Zonouz 3/10/2026

LegoNet: Memory Footprint Reduction Through Block Weight Clustering

LegoNet compression technique for neural networks using block weight clustering to reduce memory footprint for embedded device deployment.

Ax Lukas Thede, Stefan Winzeck, Zeynep Akata, Jonathan Richard Schwarz 3/10/2026

CapTrack: Multifaceted Evaluation of Forgetting in LLM Post-Training

CapTrack benchmark for evaluating multi-faceted forgetting in LLM post-training beyond parametric knowledge loss, addressing domain adaptation challenges.

Ax Yegor Denisov-Blanch, Joshua Kazdan, Jessica Chudnovsky, Rylan Schaeffer, Sheng Guan, Soji Adeshina, Sanmi Koyejo 3/10/2026

Consensus is Not Verification: Why Crowd Wisdom Strategies Fail for LLM Truthfulness

Research showing majority-voting and ensemble inference methods fail to improve LLM truthfulness without external verification, unlike in math/code domains.

Ax Stamatis Mastromichalakis 3/10/2026

OptiRoulette Optimizer: A New Stochastic Meta-Optimizer for up to 5.3x Faster Convergence

OptiRoulette meta-optimizer that dynamically selects update rules during training via warmup locking and random sampling. Torch-compatible drop-in component achieving 5.3x faster convergence.

Ax Sai Hao, Hao Zeng, Hongxin Wei, Bingyi Jing 3/10/2026

RACER: Risk-Aware Calibrated Efficient Routing for Large Language Models

RACER system for efficient multi-model LLM routing formulated as risk-aware optimization problem. Extends base routers to minimize cost-performance trade-off.

Ax Junde Wu, Minhao Hu, Jiayuan Zhu, Yuyuan Liu, Tianyi Zhang, Kang Li, Jingkun Chen, Jiazhen Pan, Min Xu, Yueming Jin 3/10/2026

Evo: Autoregressive-Diffusion Large Language Models with Evolving Balance

Novel language model combining autoregressive and diffusion-based generation through latent trajectory modeling with evolving balance parameter.

Ax Hejian Sang, Yuanda Xu, Zhengze Zhou, Ran He, Zhipeng Wang 3/10/2026

Not all tokens are needed(NAT): token efficient reinforcement learning

Framework for token-efficient reinforcement learning in LLMs. Proposes NAT to reduce computational cost of backpropagation over long chain-of-thought trajectories during training.

Ax Rishabh Tiwari, Aditya Tomar, Udbhav Bamba, Monishwaran Maheswaran, Heng Yang, Michael W. Mahoney, Kurt Keutzer, Amir Gholami 3/10/2026

Reward Under Attack: Analyzing the Robustness and Hackability of Process Reward Models

Research on vulnerabilities in Process Reward Models used in LLM reasoning pipelines. Introduces diagnostic framework to quantify adversarial exploitability and fluency-logic dissociation.

Ax Suhasnadh Reddy Veluru, Sai Teja Erukude, Viswa Chaitanya Marella 3/10/2026

From ARIMA to Attention: Power Load Forecasting Using Temporal Deep Learning

Empirical comparison of ARIMA, LSTM, BiLSTM, and Transformer for short-term power load forecasting.

Ax Zexiang Liu, Xianglong He, Yangguang Li 3/10/2026

Advances in GRPO for Generation Models: A Survey

Survey of Group Relative Policy Optimization for aligning generative models with human preferences.

Ax Yuqi Xu, Rizhen Hu, Zihan Liu, Mou Sun, Kun Yuan 3/10/2026

Grouter: Decoupling Routing from Representation for Accelerated MoE Training

Grouter: decoupled routing method for accelerating Mixture-of-Experts training with structural priors.

Ax Hamideh Khaleghpour, Brett McKinney 3/10/2026

Leakage Safe Graph Features for Interpretable Fraud Detection in Temporal Transaction Networks

Leakage-safe graph feature extraction for fraud detection in temporal transaction networks.

Ax Michal Podstawski 3/10/2026

Graph Property Inference in Small Language Models: Effects of Representation and Inference Strategy

Graph property inference in small language models: effects of representation on structured reasoning.

Ax Qingsong Zou, Zhi Yan, Zhiyao Xu, Kuofeng Gao, Jingyu Xiao, Yong Jiang 3/10/2026

SmartBench: Evaluating LLMs in Smart Homes with Anomalous Device States and Behavioral Contexts

SmartBench: evaluation benchmark for LLMs in smart home environments with anomalous device detection.

Ax Sirui Li, Shuhan Xiao, Mihir Joshi, Ahmed Metwally, Daniel McDuff, Wei Wang, Yuzhe Yang 3/10/2026

HEARTS: Benchmarking LLM Reasoning on Health Time Series

HEARTS: benchmark for evaluating LLM reasoning on diverse health time series tasks and modalities.

Ax Swamynathan V P 3/10/2026

SR-TTT: Surprisal-Aware Residual Test-Time Training

SR-TTT: test-time training for LLMs with infinite context via fast weights, improving long-context reasoning.

Ax Paul Shepherd, Tasos Dagiuklas, Bugra Alkan, Joaquim Bastos, Jonathan Rodriguez 3/10/2026

Trust Aware Federated Learning for Secure Bone Healing Stage Interpretation in e-Health

Trust-aware federated learning framework for bone healing classification in distributed medical environments.

Ax Sanjay Mishra 3/10/2026

ERP-RiskBench: Leakage-Safe Ensemble Learning for Financial Risk

Ensemble learning framework for financial risk detection in ERP systems with leakage-safe evaluation.

Ax Karan Gupta, Pranav Vajreshwari, Yash Pandya, Raghav Magazine, Akshay Nambi, Ahmed Awadallah 3/10/2026

Scaling Agentic Capabilities, Not Context: Efficient Reinforcement Finetuning for Large Toolspaces

ATLAS: reinforcement finetuning framework for scaling agent capabilities with large toolspaces using small language models.

Ax Guanglin Zhou, Armin Catic, Motahare Shabestari, Matthew Young, Chaiquan Li, Katrina Poppe, Sebastiano Barbieri 3/10/2026

From Statistical Fidelity to Clinical Consistency: Scalable Generation and Auditing of Synthetic Patient Trajectories

Synthetic EHR generation pipeline with clinical consistency validation for privacy-preserving health data sharing.

Ax Aditya Ranganath, Hasin Us Sami, Kowshik Thopalli, Bhavya Kailkhura, Wesam Sakla 3/10/2026

ProtAlign: Contrastive learning paradigm for Sequence and structure alignment

ProtAlign: contrastive learning framework for protein sequence-structure alignment using language models.

Ax Yunzhong Qiu, Binzhu Li, Hao Wei, Shenglin Weng, Chen Wang, Zhongyi Pei, Mingsheng Long, Jianmin Wang 3/10/2026

Regression Models Meet Foundation Models: A Hybrid-AI Approach to Practical Electricity Price Forecasting

Hybrid approach combining time series foundation models with regression models for electricity price forecasting capturing temporal and cross-variate patterns.

Ax Jingyuan Feng, Andrew Gambardella, Gouki Minegishi, Takeshi Kojima, Yusuke Iwasawa, Yutaka Matsuo 3/10/2026

Safe Transformer: An Explicit Safety Bit For Interpretable And Controllable Alignment

Safe Transformer: Modular approach adding explicit safety bit as interpretable information bottleneck for controllable alignment in language models.

Ax Ramchand Kumaresan 3/10/2026

Orion: Characterizing and Programming Apple's Neural Engine for LLM Training and Inference

Orion: First open system enabling direct LLM training and inference on Apple Neural Engine (ANE) with compiler pipeline and on-device training support.

Ax Jiefu Zhang, Yang Xu, Vaneet Aggarwal 3/10/2026

Don't Freeze, Don't Crash: Extending the Safe Operating Range of Neural Navigation in Dense Crowds

Reinforcement learning approach for safe neural navigation in variable-density crowds that generalizes beyond training conditions.

Ax Dongheon Lee, Seokju Yun, Jaegyun Im, Youngmin Ro 3/10/2026

Rank-Factorized Implicit Neural Bias: Scaling Super-Resolution Transformer with FlashAttention

Super-resolution transformer optimization using FlashAttention with rank-factorized implicit neural bias to reduce computational burden.

Ax Zhiying Jiang, Raihan Seraj, Marcos Villagra, Bidhan Roy 3/10/2026

Heterogeneous Decentralized Diffusion Models

Efficient decentralized framework for training diffusion models with heterogeneous objectives across isolated experts with reduced computational requirements.

Ax Xiaoxuan Liang, Saeid Naderiparizi, Yunpeng Liu, Berend Zwartsenberg, Frank Wood 3/10/2026

Improved Constrained Generation by Bridging Pretrained Generative Models

Constrained generation framework for diffusion models handling complex feasible regions in robotics and autonomous driving applications.

Ax Jianyuan Zhong, Kaibo Wang, Ding Ding, Zijin Feng, Haoli Bai, Yang Xiang, Jiacheng Sun, Qiang Xu 3/10/2026

Stabilizing Reinforcement Learning for Diffusion Language Models

Method to stabilize Group Relative Policy Optimization (GRPO) for diffusion language models by addressing reward collapse issues in post-training.

Ax Minjae Kang, Jaehyung Kim 3/10/2026

Enhancing Instruction Following of LLMs via Activation Steering with Dynamic Rejection

DIRECTER: Dynamic rejection steering technique for LLMs to improve instruction following while avoiding oversteering that degrades output quality.

Ax Gregor Baer 3/10/2026

xaitimesynth: A Python Package for Evaluating Attribution Methods for Time Series with Synthetic Ground Truth

xaitimesynth: Reusable Python package for evaluating time series attribution methods using synthetic ground truth data with class-discriminating features.

Ax Mai Pham, Vikrant Vaze, Peter Chin 3/10/2026

Optimistic Policy Regularization

OPR: Lightweight mechanism preventing premature policy convergence in deep RL by maintaining buffer of high-performing episodes during optimization.

Ax Irene Wang, Vishnu Varma Venkata, Arvind Krishnamurthy, Divya Mahajan 3/10/2026

NEST: Network- and Memory-Aware Device Placement For Distributed Deep Learning

NEST: Device placement optimization for distributed deep learning that jointly considers parallelism, memory, and network topology without post-hoc feasibility fixes.

Ax Wenjing Chen, Chengyuan Qian, Shuo Xing, Yi Zhou, Victoria Crawford 3/10/2026

Multi-Agent Reinforcement Learning with Submodular Reward

Framework for cooperative multi-agent RL with submodular rewards modeling overlapping agent contributions. First formal analysis of diminishing returns in team coordination.

Ax Yanjun Chen, Yirong Sun, Hanlin Wang, Xinming Zhang, Xiaoyu Shen, Wenjie Li, Wei Zhang 3/10/2026

Contextual Counterfactual Credit Assignment for Multi-Agent Reinforcement Learning in LLM Collaboration

C3 algorithm for multi-agent reinforcement learning with LLMs. Addresses credit assignment problem in sparse reward scenarios through contextual counterfactual analysis.

Ax Mingi Kang, Zai Yang, Jeova Farias Sales Rocha Neto 3/10/2026

IGLU: The Integrated Gaussian Linear Unit Activation Function

Gaussian Linear Unit activation function analyzing mathematical relationships among modern transformer alternatives to ReLU.

Ax Abdulrahman Alswaidan, Jeffrey D. Varner 3/10/2026

Stochastic Attention via Langevin Dynamics on the Modern Hopfield Energy

Stochastic attention mechanism via Langevin sampling on Hopfield energy enabling temperature-controlled retrieval and generation.

Ax Gyujun Jeong (School of Electrical and Computer Engineering, Georgia Institute of Technology, GA, USA), Sungwon Cho (School of Electrical and Computer Engineering, Georgia Institute of Technology, GA, USA), Minji Shon (School of Electrical and Computer Engineering, Georgia Institute of Technology, GA, USA), Namhoon Kim (School of Electrical and Computer Engineering, Georgia Institute of Technology, GA, USA), Woohyun Hwang (Semiconductor Research and Development, Samsung Electronics Co., Ltd, South Korea), Kwangyou Seo (Semiconductor Research and Development, Samsung Electronics Co., Ltd, South Korea), Suhwan Lim (Semiconductor Research and Development, Samsung Electronics Co., Ltd, South Korea), Wanki Kim (Semiconductor Research and Development, Samsung Electronics Co., Ltd, South Korea), Daewon Ha (Semiconductor Research and Development, Samsung Electronics Co., Ltd, South Korea), Prasanna Venkatesan (NVIDIA, Santa Clara, CA, USA), Kihang Youn (NVIDIA, Santa Clara, CA, USA), Ram Cherukuri (NVIDIA, Santa Clara, CA, USA), Yiyi Wang (NVIDIA, Santa Clara, CA, USA), Suman Datta (School of Electrical and Computer Engineering, Georgia Institute of Technology, GA, USA), Asif Khan (School of Electrical and Computer Engineering, Georgia Institute of Technology, GA, USA), Shimeng Yu (School of Electrical and Computer Engineering, Georgia Institute of Technology, GA, USA) 3/10/2026

Physics-informed AI Accelerated Retention Analysis of Ferroelectric Vertical NAND: From Day-Scale TCAD to Second-Scale Surrogate Model

Physics-informed surrogate model for ferroelectric NAND retention analysis replacing expensive TCAD simulations.

Ax Yan-Ying Chen, Dule Shu, Matthew Hong, Andrew Taber, Jonathan Li, Matthew Klenk 3/10/2026

Learning From Design Procedure To Generate CAD Programs for Data Augmentation

LLM-based CAD program generation using design procedures and geometric constraints for parametric model synthesis.

Ax Jim Achterberg, Marcel Haas, Bram van Dijk, Marco Spruit 3/10/2026

XGenBoost: Synthesizing Small and Large Tabular Datasets with XGBoost

Generative models for tabular data using XGBoost as score estimator with denoising diffusion for small and large datasets.

Ax Nandan Kumar Jha, Brandon Reagen 3/10/2026

NerVE: Nonlinear Eigenspectrum Dynamics in LLM Feed-Forward Networks

Eigenspectral framework analyzing information flow in LLM feed-forward networks through lightweight spectral metrics.

Ax Zhixu Du, Krishna Teja Chitty-Venkata, Murali Emani, Venkatram Vishwanath, Hai Helen Li, Yiran Chen 3/10/2026

Swimba: Switch Mamba Model Scales State Space Models

Mixture-of-experts approach for state space models with expert specialization while maintaining computational efficiency.

Ax Milad Shirani, Pete H. Gueldner, Murat Khidoyatov, Jeremy L. Warren, Federica Ninno 3/10/2026

Physics-Consistent Neural Networks for Learning Deformation and Director Fields in Microstructured Media with Loss-Based Validation Criteria

Physics-consistent neural networks for Cosserat elasticity modeling deformation and director fields in microstructured materials.

Ax Ege C. Kaya, Mahsa Ghasemi, Abolfazl Hashemi 3/10/2026