4/7/2026
Causal Bandit Over Unknown Graphs: Upper Confidence Bounds With Backdoor Adjustment
Paper on causal bandit algorithms for unknown DAGs using confidence bounds and backdoor adjustment for intervention discovery.
4/7/2026
From Restless to Contextual: A Thresholding Bandit Reformulation For Finite-horizon Improvement
Research on finite-horizon restless bandit problems reformulated as thresholding with improved sample complexity and policy convergence.
4/7/2026
RESIST: Resilient Decentralized Learning Using Consensus Gradient Descent
arXiv paper on decentralized learning using consensus gradient descent with privacy and communication constraints across networked devices.
4/7/2026
Model Privacy: A Unified Framework for Understanding Model Stealing Attacks and Defenses
Research paper on model stealing attacks and defenses, analyzing vulnerabilities of ML services to adversarial extraction through query access.
4/7/2026
An Analytical Theory of Spectral Bias in the Learning Dynamics of Diffusion Models
Analytical framework explaining spectral bias in diffusion model training dynamics using Gaussian equivalence and probability-flow ODEs.
4/7/2026
RaPA: Enhancing Transferable Targeted Attacks via Random Parameter Pruning
RaPA improves transferable targeted adversarial attacks by random parameter pruning to reduce reliance on surrogate model subsets.
4/7/2026
From Set Convergence to Pointwise Convergence: Finite-Time Guarantees for Average-Reward Q-Learning with Adaptive Stepsizes
Finite-time convergence analysis for average-reward Q-learning with adaptive stepsizes, showing O(1/k) convergence rate.
4/7/2026
A Multi-Level Causal Intervention Framework for Mechanistic Interpretability in Variational Autoencoders
First mechanistic interpretability framework for VAEs using multi-level causal interventions to understand generative model representations.
4/7/2026
Understanding Task Representations in Neural Networks via Bayesian Ablation
Introduces Bayesian ablation framework for interpreting learned task representations in neural networks through probabilistic inference.
4/7/2026
MSDformer: Multi-scale Discrete Transformer For Time Series Generation
MSDformer extends discrete token modeling for time series generation using multi-scale transformer architecture to capture temporal patterns.
4/7/2026
SoSBench: Benchmarking Safety Alignment on Six Scientific Domains
SoSBench benchmarks safety alignment of LLMs across six scientific domains with sophisticated risks beyond basic misuse scenarios.
4/7/2026
LLMs Judging LLMs: A Simplex Perspective
Studies the problem of using LLMs as judges for evaluating LLM outputs, addressing epistemic uncertainty in judge quality beyond sampling variability.
4/7/2026
Beyond Linear Steering: Unified Multi-Attribute Control for Language Models
K-Steering enables unified multi-attribute control of LLMs at inference time using non-linear classifiers on hidden activations to handle attribute interference.
4/7/2026
MLorc: Momentum Low-rank Compression for Memory Efficient Large Language Model Adaptation
MLorc proposes momentum low-rank compression for memory-efficient LLM fine-tuning, reducing memory demands compared to LoRA while maintaining performance.
4/7/2026
SFBD Flow: A Continuous-Optimization Framework for Training Diffusion Models with Noisy Samples
SFBD Flow framework trains diffusion models on corrupted/noisy data with clean samples to reduce privacy risks and improve convergence in generative modeling.
4/7/2026
Not All Tokens Matter: Towards Efficient LLM Reasoning via Token Significance in Reinforcement Learning
Token significance approach in RL for efficient LLM reasoning by identifying and prioritizing important tokens over length optimization.
4/7/2026
Federated Item Response Models: A Gradient-driven Privacy-preserving Framework for Distributed Psychometric Estimation
Federated Item Response Theory (FedIRT) framework enabling distributed psychometric estimation without centralizing raw response data.
4/7/2026
Causal Process Models: Reframing Dynamic Causal Graph Discovery as a Reinforcement Learning Problem
Causal Process Models for learning sparse time-varying causal graphs from visual observations using reinforcement learning.
4/7/2026
Evaluating and Learning Robust Bandit Policies Under Uncertain Causal Mechanisms
Causal multi-armed bandit algorithm reasoning under uncertain causal mechanisms from graphical models.
4/7/2026
xRFM: Accurate, scalable, and interpretable feature learning models for tabular data
xRFM feature learning models for tabular data providing accurate, scalable, and interpretable alternatives to gradient boosted trees.
4/7/2026
LoFT: Parameter-Efficient Fine-Tuning for Long-tailed Semi-Supervised Learning in Open-World Scenarios
LoFT parameter-efficient fine-tuning approach for long-tailed semi-supervised learning leveraging foundation models.
4/7/2026
ACT: Agentic Classification Tree
Agentic Classification Tree (ACT) combining LLMs with decision trees for transparent, interpretable decisions on unstructured data.
4/7/2026
Fewer Weights, More Problems: A Practical Attack on LLM Pruning
Security analysis exposing vulnerabilities in LLM weight pruning methods used by inference engines like vLLM.
4/7/2026
A Minimal-Assumption Analysis of Q-Learning with Time-Varying Policies
Theoretical finite-time analysis of Q-learning with time-varying policies under minimal assumptions for Markov decision processes.
4/7/2026
A Model Can Help Itself: Reward-Free Self-Training for LLM Reasoning
Self-evolving Post-Training (SePT) method enabling LLMs to improve reasoning without external rewards through self-generated training data.
4/7/2026
An Information-Theoretic Analysis of OOD Generalization in Meta-Reinforcement Learning
Information-theoretic analysis of out-of-distribution generalization in meta-reinforcement learning with bounds under distribution shift scenarios.
4/7/2026
SPORE: Skeleton Propagation Over Recalibrating Expansions
Classical clustering algorithm handling arbitrary cluster geometry without global density assumptions using skeleton propagation and recalibrating expansions.
4/7/2026
SynQuE: Estimating Synthetic Dataset Quality Without Annotations
Method for ranking synthetic datasets by real-world performance without annotations, establishing benchmarks for synthetic data quality estimation.
4/7/2026
Controllable protein design with particle-based Feynman-Kac steering
Feynman-Kac framework for guiding diffusion-based generative models toward proteins with specified properties and tailored structures.
4/7/2026
A Unified Stability Analysis of SAM vs SGD: Role of Data Coherence and Emergence of Simplicity Bias
Unified stability analysis comparing SAM and SGD optimization algorithms showing role of data coherence and simplicity bias in generalization.
4/7/2026
Semantic Superiority vs. Forensic Efficiency: A Comparative Analysis of Deep Learning and Psycholinguistics for Business Email Compromise Detection
Comparative analysis of transformer models (DistilBERT) versus psycholinguistic features for detecting business email compromise attacks.
4/7/2026
Mitigating Structural Overfitting: A Distribution-Aware Rectification Framework for Missing Feature Imputation
Framework addressing structural overfitting in graph neural networks for missing feature imputation using distribution-aware rectification.
4/7/2026
Personalized Federated Distillation Assisted Vehicle Edge Caching Strategy
Federated learning approach for vehicle edge caching using personalized distillation to predict user content preferences while preserving privacy.
4/7/2026
Random-Bridges as Stochastic Transports for Generative Models
Random-bridges framework for generative models using stochastic processes conditioned on target distributions for flexible transport between distributions.
4/7/2026
GRAFT: Grid-Aware Load Forecasting with Multi-Source Textual Alignment and Fusion
Electric load forecasting model integrating multi-source textual data (news, social media, policies) with temporal grid-aware predictions.
4/7/2026
Kinetic-Mamba: Mamba-Assisted Predictions of Stiff Chemical Kinetics
Mamba-based neural operator framework for accurate chemical kinetics modeling in combustion simulations using efficient temporal modeling.
4/7/2026
SFBD-OMNI: Bridge models for lossy measurement restoration with limited clean samples
Distribution restoration method using noisy samples and optimal transport to recover fully observed data from partial corrupted observations.
4/7/2026
Bridging the Semantic Gap for Categorical Data Clustering via Large Language Models
Method using large language models to measure semantic similarity in categorical data clustering by bridging gap in attribute distance representation.
4/7/2026
ASSS: A Differentiable Adversarial Framework for Task-Aware Data Reduction
Differentiable adversarial framework for task-aware data reduction using learnable selector and minimax optimization to identify informative samples.
4/7/2026
From Bits to Chips: An LLM-based Hardware-Aware Quantization Agent for Streamlined Deployment of LLMs
LLM-based hardware-aware quantization agent automating model quantization for efficient LLM deployment on resource-constrained hardware.
4/7/2026
Understanding and inverse design of implicit bias in stochastic learning: a geometric perspective
Theoretical analysis of implicit bias in stochastic learning using geometric perspective to explain solution selection in overparameterized models.
4/7/2026
HOSL: Hybrid-Order Split Learning for Memory-Constrained Edge Training
Split learning optimization method reducing memory overhead for LLM training on edge devices using hybrid-order optimization instead of first-order approaches.
4/7/2026
Auxiliary-predicted Compress Memory Model(ApCM Model): A Neural Memory Storage Model Based on Invertible Compression and Learnable Prediction
Neural memory storage architecture for LLMs with invertible compression and learnable prediction for runtime memory.
4/7/2026
InfoTok: Information-Theoretic Regularization for Capacity-Constrained Shared Visual Tokenization in Unified MLLMs
Information-theoretic approach for designing shared visual tokenizers in unified multimodal LLMs.
4/7/2026
RASA: Routing-Aware Safety Alignment for Mixture-of-Experts Models
Safety alignment framework addressing unique challenges of sparse routing in Mixture-of-Experts language models.
4/7/2026
ModalImmune: Immunity Driven Unlearning via Self Destructive Training
Training framework for multimodal systems to maintain performance when input channels are lost at deployment.
4/7/2026
Unlearning Noise in PINNs: A Selective Pruning Framework for PDE Inverse Problems
Pruning framework for physics-informed neural networks to improve robustness to noise in PDE inverse problems.
4/7/2026
T1: One-to-One Channel-Head Binding for Multivariate Time-Series Imputation
Time series imputation method using channel-head binding for handling diverse missing patterns.
4/7/2026
MOOSE-Star: Unlocking Tractable Training for Scientific Discovery by Breaking the Complexity Barrier
Training method for LLMs to directly model generative reasoning process in scientific discovery applications.
4/7/2026
NePPO: Near-Potential Policy Optimization for General-Sum Multi-Agent Reinforcement Learning
Multi-agent reinforcement learning algorithm for general-sum games with convergence guarantees in heterogeneous agent settings.