Ax Ian Li, Zilei Shao, Benjie Wang, Rose Yu, Guy Van den Broeck, Anji Liu 3/11/2026

Breaking the Factorization Barrier in Diffusion Language Models

Research on diffusion language models addressing the factorization barrier to enable efficient parallel token generation.

Ax Edouard Lansiaux, Margaux Leman, Mehdi Ammi 3/11/2026

OrthoAI: A Neurosymbolic Framework for Evidence-Grounded Biomechanical Reasoning in Clear Aligner Orthodontics

OrthoAI combines 3D tooth segmentation with biomechanical reasoning for clear aligner orthodontics using sparse-supervision learning.

Ax Abhinav Munagala 3/11/2026

Zero-Shot and Supervised Bird Image Segmentation Using Foundation Models: A Dual-Pipeline Approach with Grounding DINO~1.5, YOLOv11, and SAM~2.1

Dual-pipeline bird image segmentation framework combining Grounding DINO 1.5, YOLOv11, and SAM 2.1 for zero-shot and supervised segmentation.

Ax Jisoo Kim, Jungbin Cho, Sanghyeok Chu, Ananya Bal, Jinhyung Kim, Gunhee Lee, Sihaeng Lee, Seung Hwan Kim, Bohyung Han, Hyunmin Lee, Laszlo A. Jeni, Seungryong Kim 3/11/2026

Pri4R: Learning World Dynamics for Vision-Language-Action Models with Privileged 4D Representation

Pri4R approach equipping Vision-Language-Action models with implicit world dynamics understanding through privileged 4D representation learning.

Ax Yifei Zhang, Xu Yang, Xiao Yang, Bowen Xian, Qizheng Li, Shikai Fang, Jingyuan Li, Jian Wang, Mingrui Xu, Weiqing Liu, Jiang Bian 3/11/2026

Reasoning as Gradient: Scaling MLE Agents Beyond Tree Search

GOME: MLE agent framework replacing tree search with gradient-based optimization for machine learning engineering tasks using LLM reasoning.

Ax Nhat D. A. Nguyen, Duong D. Nguyen, Gianluca Rizzo, Hung X. Nguyen 3/11/2026

Boltzmann-based Exploration for Robust Decentralized Multi-Agent Planning (Extended Version)

Coordinated Boltzmann MCTS for decentralized multi-agent planning, replacing deterministic UCT with stochastic Boltzmann policy for sparse reward environments.

Ax Sayang Mu, Xiangyu Wu, Bo An 3/11/2026

SPARC: Spatial-Aware Path Planning via Attentive Robot Communication

Communication mechanism (RMHA) for decentralized multi-robot path planning using attention over Manhattan distances for spatial-aware coordination.

Ax Yinghong Yu, Guangyuan Li, Jiancheng Yang 3/11/2026

PlaneCycle: Training-Free 2D-to-3D Lifting of Foundation Models Without Adapters

Training-free method (PlaneCycle) for lifting 2D foundation models to 3D volumetric data without adapters or retraining via cyclic spatial aggregation.

Ax Alper Y{\i}ld{\i}r{\i}m 3/11/2026

The Geometric Inductive Bias of Grokking: Bypassing Phase Transitions via Architectural Topology

Study of grokking phenomenon through architectural modifications, identifying topology-based degrees of freedom that influence memorization vs. generalization phases.

Ax Siddharth Boppana, Annabel Ma, Max Loeffler, Raphael Sarfati, Eric Bigelow, Atticus Geiger, Owen Lewis, Jack Merullo 3/11/2026

Reasoning Theater: Disentangling Model Beliefs from Chain-of-Thought

Analysis of performative chain-of-thought in reasoning models, showing hidden beliefs diverge from generated reasoning tokens at task-specific difficulty levels.

Ax Helena Casademunt, Bartosz Cywi\'nski, Khoi Tran, Arya Jakkli, Samuel Marks, Neel Nanda 3/11/2026

Censored LLMs as a Natural Testbed for Secret Knowledge Elicitation

Methods for eliciting truthful outputs from censored LLMs using honesty elicitation and lie detection, tested on models trained to conceal information.

Ax Stamatis Mastromichalakis 3/11/2026

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

OptiRoulette meta-optimizer that dynamically selects update rules during training for faster convergence, packaged as PyTorch-compatible component.

Ax Kuan Zhang, Dongchen Liu, Qiyue Zhao, Jinkun Hou, Xinran Zhang, Qinlei Xie, Miao Liu, Yiming Li 3/11/2026

GameVerse: Can Vision-Language Models Learn from Video-based Reflection?

GameVerse benchmark evaluating Vision-Language Models' ability to learn from video-based reflection and self-correction in game environments.

Ax Xiaoyang Hou, Junqi Liu, Chence Shi, Xin Liu, Zhi Yang, Jian Tang 3/11/2026

Property-driven Protein Inverse Folding With Multi-Objective Preference Alignment

Multi-objective preference alignment approach for protein sequence design balancing designability with competing properties like solubility and thermostability.

Ax Shuai Lu, Meng Wang, Jia Guo, Jiawei Du, Bo Liu, Shengzhu Yang, Weihang Zhang, Huazhu Fu, Huiqi Li 3/11/2026

Deep Expert Injection for Anchoring Retinal VLMs with Domain-Specific Knowledge

Method for improving Vision-Language Models for ophthalmic diagnosis by injecting domain-specific knowledge to address perception and reasoning gaps.

Ax Yuxu Ge 3/11/2026

Governance Architecture for Autonomous Agent Systems: Threats, Framework, and Engineering Practice

Layered Governance Architecture (LGA) framework addressing execution-layer vulnerabilities in LLM-based autonomous agents: prompt injection, retrieval poisoning, tool invocation.

Ax Angad Singh Ahuja 3/11/2026

Adversarial Latent-State Training for Robust Policies in Partially Observable Domains

Adversarial training approach for robust reinforcement learning policies in partially observable domains with latent distribution shift.

Ax Sean Gunn, Jorio Cocola, Oliver De Candido, Vaggos Chatziafratis, Paul Hand 3/11/2026

Latent Generative Models with Tunable Complexity for Compressed Sensing and other Inverse Problems

Tunable-complexity priors for diffusion models, normalizing flows, and VAEs to solve inverse problems with flexible representation capacity.

Ax Colin Aitken, Rajat Masiwal, Adam Marchakitus, Katherine Kowal, Mayank Gupta, Tyler Yang, Amir Jina, Pedram Hassanzadeh, William R. Boos, Michael Kremer 3/11/2026

Designing probabilistic AI monsoon forecasts to inform agricultural decision-making

Decision-theory framework for designing probabilistic weather forecasts tailored to heterogeneous farmers' circumstances for agricultural decision-making.

Ax Sunghyun Baek, Jaemyung Yu, Seunghee Koh, Minsu Kim, Hyeonseong Jeon, Junmo Kim 3/11/2026

IMSE: Intrinsic Mixture of Spectral Experts Fine-tuning for Test-Time Adaptation

Test-time adaptation method using spectral experts in Vision Transformers to handle distribution shift with minimal parameter updates.

Ax Peter Brodeur, Jacob M. Koshy, Anil Palepu, Khaled Saab, Ava Homiar, Roma Ruparel, Charles Wu, Ryutaro Tanno, Joseph Xu, Amy Wang, David Stutz, Hannah M. Ferrera, David Barrett, Lindsey Crowley, Jihyeon Lee, Spencer E. Rittner, Ellery Wulczyn, Selena K. Zhang, Elahe Vedadi, Christine G. Kohn, Kavita Kulkarni, Vinay Kadiyala, Sara Mahdavi, Wendy Du, Jessica Williams, David Feinbloom, Renee Wong, Tao Tu, Petar Sirkovic, Alessio Orlandi, Christopher Semturs, Yun Liu, Juraj Gottweis, Dale R. Webster, Jo\"elle Barral, Katherine Chou, Pushmeet Kohli, Avinatan Hassidim, Yossi Matias, James Manyika, Rob Fields, Jonathan X. Li, Marc L. Cohen, Vivek Natarajan, Mike Schaekermann, Alan Karthikesalingam, Adam Rodman 3/11/2026

A prospective clinical feasibility study of a conversational diagnostic AI in an ambulatory primary care clinic

Clinical feasibility study of LLM-based conversational AI (AMIE) for patient diagnostic history-taking in real-world primary care workflows with safety assessment.

Ax Ben Rank, Hardik Bhatnagar, Ameya Prabhu, Shira Eisenberg, Karina Nguyen, Matthias Bethge, Maksym Andriushchenko 3/11/2026

PostTrainBench: Can LLM Agents Automate LLM Post-Training?

PostTrainBench benchmark evaluating whether LLM agents can automate post-training of base LLMs into assistants.

Ax Panayiotis Raptis, Fatih Aslan, George Iosifidis 3/11/2026

Equitable Multi-Task Learning for AI-RANs

OWO-FMTL framework for fair multi-task learning in AI-enabled radio access networks with equitable user performance.

Ax Hui-Ze Tan, Xiao-Wen Yang, Hao Chen, Jie-Jing Shao, Yi Wen, Yuteng Shen, Weihong Luo, Xiku Du, Lan-Zhe Guo, Yu-Feng Li 3/11/2026

Hindsight Credit Assignment for Long-Horizon LLM Agents

HCAPO framework addressing credit assignment in long-horizon LLM agent tasks using hindsight and value baseline alignment.

Ax Kaushik Roy, Giovanni D'urso, Nicholas Lawrance, Brendan Tidd, Peyman Moghadam 3/11/2026

SPREAD: Subspace Representation Distillation for Lifelong Imitation Learning

SPREAD framework for lifelong imitation learning preserving task representation manifolds across sequential skill acquisition without catastrophic forgetting.

Ax Sichen Yang (Johns Hopkins University), Mauro Maggioni (Johns Hopkins University) 3/11/2026

Multi-level meta-reinforcement learning with skill-based curriculum

Multi-level meta-reinforcement learning with skill-based curriculum for hierarchical sequential decision-making through MDP compression.

Ax Michael Leznik 3/11/2026

The Temporal Markov Transition Field

Temporal Markov Transition Field extension handling non-stationary time series by tracking regime changes instead of using global transition matrix.

Ax Anselm Paulus, A. Ren\'e Geist, V\'it Musil, Sebastian Hoffmann, Onur Beker, Georg Martius 3/11/2026

SoftJAX & SoftTorch: Empowering Automatic Differentiation Libraries with Informative Gradients

SoftJAX and SoftTorch extensions enabling informative gradients for hard primitives like thresholding and discrete operations in automatic differentiation frameworks.

Ax Jay Revolinsky, Harry Shomer, Jiliang Tang 3/11/2026

Are Expressive Encoders Necessary for Discrete Graph Generation?

GenGNN modular framework for discrete graph generation achieving competitive validity with graph Transformers at 2-5x faster inference speed.

Ax John Cooper, Ilias Diakonikolas, Mingchen Ma, Frederic Sala 3/11/2026

Expressivity-Efficiency Tradeoffs for Hybrid Sequence Models

Study of hybrid sequence models combining Transformers and state-space models, analyzing expressivity-efficiency tradeoffs and mechanisms for performance benefits.

Ax Mohammad Hossein Safarpour, Seyed Mohammad Alavi, Mohammad Izadikhah, Hossein Dibachi 3/11/2026

A New Modeling to Feature Selection Based on the Fuzzy Rough Set Theory in Normal and Optimistic States on Hybrid Information Systems

Feature selection method for hybrid information systems using fuzzy rough set theory to reduce dimensionality in big data scenarios.

Ax Abhinaba Basu 3/11/2026

Cross-Domain Uncertainty Quantification for Selective Prediction: A Comprehensive Bound Ablation with Transfer-Informed Betting

Comprehensive ablation of confidence bounds for selective prediction with risk control, introducing Transfer-Informed Betting for cross-domain uncertainty quantification.

Ax Alexander Nemecek, Wenbiao Li, Xiaoqian Jiang, Jaideep Vaidya, Erman Ayday 3/11/2026

Quantifying Memorization and Privacy Risks in Genomic Language Models

Study of memorization and privacy risks in genomic language models trained on sensitive DNA sequence data, assessing data leakage concerns.

Ax Itamar Tsayag, Ofir Lindenbaum 3/11/2026

Uncovering a Winning Lottery Ticket with Continuously Relaxed Bernoulli Gates

Strong Lottery Ticket method using continuously relaxed Bernoulli gates to find sparse subnetworks in over-parameterized networks without weight training.

Ax Vignesh Adhinarayanan, Nuwan Jayasena 3/11/2026

The $qs$ Inequality: Quantifying the Double Penalty of Mixture-of-Experts at Inference

Analysis of Mixture-of-Experts inference inefficiency identifying double penalty from microbatch fragmentation and reduced KV cache headroom during decoding.

Ax Edward Izgorodin 3/11/2026

Semantic Level of Detail: Multi-Scale Knowledge Representation via Heat Kernel Diffusion on Hyperbolic Manifolds

Semantic Level of Detail introduces continuous resolution control for knowledge graphs via heat kernel diffusion, enabling agents to navigate abstraction levels.

Ax Federico Giannini, Emanuele Della Valle 3/11/2026

MAcPNN: Mutual Assisted Learning on Data Streams with Temporal Dependence

MAcPNN proposes mutual assisted learning for IoT edge devices handling streaming data with temporal dependencies and concept drift without catastrophic forgetting.

Ax Kailong Fan, Anqi Pu, Yichen Wu, Wanhua Li, Yicong Li, Hanspeter Pfister, Huafeng Liu, Xiang Li, Quanzheng Li, Ning Guo 3/11/2026

MAPLE: Elevating Medical Reasoning from Statistical Consensus to Process-Led Alignment

MAPLE framework improves medical LLM reasoning by replacing majority voting with process-led alignment in test-time reinforcement learning for clinical decision-making.

Ax Rahman Taleghani, Maryam Mohammadi, Francesco Marchetti 3/11/2026

An accurate flatness measure to estimate the generalization performance of CNN models

Flatness measure for CNN generalization based on Hessian spectrum accounting for convolutional architecture-specific geometric structure.

Ax Ren Fujiwara, Yasuko Matsubara, Yasushi Sakurai 3/11/2026

When to Retrain after Drift: A Data-Only Test of Post-Drift Data Size Sufficiency

CALIPER: data-only detector estimating post-drift data size sufficiency for stable model retraining after concept drift in data streams.

Ax Yuchen Yuan, Junhuan Yang, Hao Wan, Yipei Liu, Hanhan Wu, Youzuo Lin, Lei Yang 3/11/2026

Two Teachers Better Than One: Hardware-Physics Co-Guided Distributed Scientific Machine Learning

Distributed scientific ML approach co-guiding models with hardware and physics constraints for edge processing without central data aggregation.

Ax Renos Zabounidis, Yue Wu, Simon Stepputtis, Woojun Kim, Yuanzhi Li, Tom Mitchell, Katia Sycara 3/11/2026

SCALAR: Learning and Composing Skills through LLM Guided Symbolic Planning and Deep RL Grounding

SCALAR framework coupling LLM planning with deep RL through learned skill libraries, enabling agents to ground language into low-level control with iterative correction.

Ax Hongyu Cao, Jinghan Zhang, Kunpeng Liu, Dongjie Wang, Feng Xia, Haifeng Chen, Xiaohua Hu, Yanjie Fu 3/11/2026

Sim2Act: Robust Simulation-to-Decision Learning via Adversarial Calibration and Group-Relative Perturbation

Simulation-to-decision learning framework addressing prediction errors in simulators through adversarial calibration for reliable policy training in industrial domains.

Ax Seungha Hong, Sukang Chae, Suyeon Kim, Sanghwan Jang, Hwanjo Yu 3/11/2026

Dynamic Multi-period Experts for Online Time Series Forecasting

Dynamic multi-period experts framework for online time series forecasting handling recurring and emergent concept drift.

Ax Chloe H. Su, Zhe Ye, Samuel Tenka, Aidan Yang, Soonho Kong, Udaya Ghai 3/11/2026

Learning Adaptive LLM Decoding

Learning adaptive decoding policies for LLMs that dynamically select sampling strategies based on prompt difficulty and compute.

Ax Shuangfei Zhai 3/11/2026

Exclusive Self Attention

Exclusive self-attention mechanism constraining attention to orthogonal information for improved Transformer language modeling.

Ax Wei Feng, Jingbo Zhang, Qiong Wu, Pingyi Fan, Qiang Fan 3/11/2026

PPO-Based Hybrid Optimization for RIS-Assisted Semantic Vehicular Edge Computing

PPO-based optimization for vehicular edge computing with reconfigurable intelligent surfaces and semantic communication.

Ax Alvaro Paredes Amorin, Andre Python, Christoph Weisser 3/11/2026

Not All News Is Equal: Topic- and Event-Conditional Sentiment from Finetuned LLMs for Aluminum Price Forecasting

Finetuned LLM sentiment analysis from news for aluminum price forecasting under different market conditions.

Ax Runyao Yu, Viviana Kleine, Philipp Gromotka, Thomas Rudolf, Adrian Eisenmann, Gautham Ram Chandra Mouli, Peter Palensky, Jochen L. Cremer 3/11/2026

Probabilistic Hysteresis Factor Prediction for Electric Vehicle Batteries with Graphite Anodes Containing Silicon

Machine learning model for predicting voltage hysteresis in silicon-graphite EV batteries using uncertainty quantification.

Ax Zhengzhao Ma, Xueru Wen, Boxi Cao, Yaojie Lu, Hongyu Lin, Jinglin Yang, Min He, Xianpei Han, Le Sun 3/11/2026

Decoupling Reasoning and Confidence: Resurrecting Calibration in Reinforcement Learning from Verifiable Rewards

Method to decouple reasoning from confidence in LLM RL training, addressing over-confidence calibration issues.