4/28/2026
Leveraging Human Feedback for Semantically-Relevant Skill Discovery
Skill discovery in RL using human preference feedback to ensure safe and aligned agent behaviors.
4/28/2026
Meta-Aligner: Bidirectional Preference-Policy Optimization for Multi-Objective LLMs Alignment
Meta-Aligner optimizes multi-objective LLM alignment by dynamically adjusting preference weights during training.
4/28/2026
BitRL: Reinforcement Learning with 1-bit Quantized Language Models for Resource-Constrained Edge Deployment
BitRL enables RL agents on edge devices by quantizing LLMs to 1-bit, reducing memory and computational requirements.
4/28/2026
Self-Abstraction Learning for Effective and Stable Training of Deep Neural Networks
Self-Abstraction Learning framework addressing gradient vanishing, overfitting, and training stability in deep neural networks.
4/28/2026
Mitigating Error Amplification in Fast Adversarial Training
Methods to mitigate catastrophic overfitting and error amplification in Fast Adversarial Training.
4/28/2026
Unveiling the Backdoor Mechanism Hidden Behind Catastrophic Overfitting in Fast Adversarial Training
Research on catastrophic overfitting in Fast Adversarial Training, analyzing backdoor mechanisms in adversarial robustness.
4/28/2026
Diffusion Templates: A Unified Plugin Framework for Controllable Diffusion
Unified plugin framework for composable controllable diffusion methods across different backbones and tasks with compatible training pipelines.
4/28/2026
An Aircraft Upset Recovery System with Reinforcement Learning
Pilot-Activated Recovery System using soft-actor critic reinforcement learning with hyperparameter optimization for aircraft upset recovery.
4/28/2026
DPRM: A Plug-in Doob h transform-induced Token-Ordering Module for Diffusion Language Models
DPRM: Token-ordering module for diffusion language models using Doob h-transform to improve generation efficiency and exploration.
4/28/2026
Complexity of Linear Regions in Self-supervised Deep ReLU Networks
Analysis of linear region complexity in self-supervised ReLU networks during training, extending prior supervised learning research.
4/28/2026
An Automatic Ground Collision Avoidance System with Reinforcement Learning
AI-based Automatic Ground Collision Avoidance System using reinforcement learning for advanced jet trainers.
4/28/2026
Advancing Ligand-based Virtual Screening and Molecular Generation with Pretrained Molecular Embedding Distance
Pretrained molecular embedding distance for ligand-based virtual screening and goal-directed molecular generation in drug discovery.
4/28/2026
SceneSelect: Selective Learning for Trajectory Scene Classification and Expert Scheduling
SceneSelect: Selective learning framework with expert scheduling for trajectory prediction across heterogeneous scene types.
4/28/2026
A Reward-Free Viewpoint on Multi-Objective Reinforcement Learning
Multi-objective reinforcement learning using reward-free RL perspective to train policies adapting to different user preference weightings.
4/28/2026
Stochastic simultaneous optimistic optimization
Global optimization algorithm for noisy function evaluation requiring only weak local smoothness assumptions without prior knowledge.
4/28/2026
GradMAP: Gradient-Based Multi-Agent Proximal Learning for Grid-Edge Flexibility
GradMAP: Decentralized multi-agent proximal learning for coordinating grid-edge devices while respecting AC distribution network constraints.
4/28/2026
Efficient learning by implicit exploration in bandit problems with side observations
Online learning algorithm for bandit problems with partial observability, where learners observe losses of other actions beyond their own.
4/28/2026
Fraud Detection in Cryptocurrency Markets with Spatio-Temporal Graph Neural Networks
Spatio-temporal graph neural networks for detecting market manipulation and fraud in cryptocurrency transactions.
4/28/2026
Uncovering Latent Patterns in Social Media Usage and Mental Health: A Clustering-Based Approach Using Unsupervised Machine Learning
Unsupervised machine learning clustering analysis of social media usage patterns and mental health correlations.
4/28/2026
Cortex-Inspired Continual Learning: Unsupervised Instantiation and Recovery of Functional Task Networks
Functional Task Networks (FTN) for continual learning using parameter isolation inspired by mammalian neocortex structure, avoiding catastrophic forgetting.
4/28/2026
The Last Human-Written Paper: Agent-Native Research Artifacts
Proposes agent-native research artifacts to address limitations of traditional scientific papers, enabling AI agents to access experimental branching and engineering details.
4/28/2026
Diffusion-Guided Feature Selection via Nishimori Temperature: Noise-Based Spectral Embedding
Physics-informed framework for feature selection in high-dimensional data using diffusion and spectral embedding without greedy search.
4/28/2026
Scalable Hyperparameter-Divergent Ensemble Training with Automatic Learning Rate Exploration for Large Models
Method for training large neural networks by using GPU replicas to explore multiple learning rates simultaneously with minimal communication overhead.
4/28/2026
SpecRLBench: A Benchmark for Generalization in Specification-Guided Reinforcement Learning
Benchmark for evaluating generalization of specification-guided reinforcement learning agents across diverse environments and unseen formal specifications.
4/28/2026
Learning to Think from Multiple Thinkers
Research on learning from Chain-of-Thought supervision across multiple different solution approaches to improve model generalization on tasks like math problems.
4/28/2026
Artificial General Intelligence Forecasting and Scenario Analysis: State of the Field, Methodological Gaps, and Strategic Implications
arXiv report reviewing AGI forecasting methodologies, their limitations, and gaps in current prediction approaches.
4/28/2026
The Randomness Floor: Measuring Intrinsic Non-Randomness in Language Model Token Distributions
Systematic study measuring intrinsic non-randomness in LLM token distributions via Entropic Deviation metric across models and prompts.
4/28/2026
When VLMs 'Fix' Students: Identifying and Penalizing Over-Correction in the Evaluation of Multi-line Handwritten Math OCR
arXiv paper on evaluating VLM performance on multi-line handwritten math OCR with semantic reasoning metrics.
4/28/2026
An Intelligent Fault Diagnosis Method for General Aviation Aircraft Based on Multi-Fidelity Digital Twin and FMEA Knowledge Enhancement
arXiv paper on intelligent fault diagnosis for aircraft using multi-fidelity digital twin and LLM-enhanced interpretable reporting.
4/28/2026
A General Framework for Generative Self-supervised Learning in Non-invasive Estimation of Physiological Parameters Using Photoplethysmography
arXiv paper on generative self-supervised learning framework for physiological parameter estimation from photoplethysmography data.
4/28/2026
Accelerating Reinforcement Learning for Wind Farm Control via Expert Demonstrations
arXiv paper on accelerating RL training for wind farm control using domain knowledge from steady-state wake models.
4/28/2026
Load constrained wind farm flow control through multi-objective multi-agent reinforcement learning
arXiv paper on multi-agent reinforcement learning for wind farm flow control with structural load constraints using I-SAC architecture.
4/28/2026
Hierarchical RL-MPC Control for Dynamic Wake Steering in Wind Farms
arXiv paper combining reinforcement learning with model predictive control for wind farm wake steering optimization.
4/28/2026
Applied AI-Enhanced RF Interference Rejection
arXiv paper on deep learning for RF interference rejection in signal detection and demodulation across varying SINR levels.
4/28/2026
In-Sync: Adaptation of Speech Aware Large Language Models for ASR with Word Level Timestamp Predictions
Speech-aware LLM extended with word-level timestamp prediction for ASR applications like captioning and media synchronization.
4/28/2026
Representation Homogeneity and Systemic Instability in AI-Dominated Financial Markets: A Structural Approach
Structural model analyzing how AI trading agents with similar representations create systemic instability in financial markets.
4/28/2026
Audio2Tool: Bridging Spoken Language Understanding and Function Calling
Audio2Tool dataset for evaluating tool-calling capabilities of speech language models across smart home, finance, and navigation domains.
4/28/2026
WebSerial Vision Training for Microcontrollers: A Browser-Based Companion to On-Device CNN Training
Browser-based tool for training CNN models on microcontrollers with image collection, training, and deployment pipeline for edge devices.
4/28/2026
IntrAgent: An LLM Agent for Content-Grounded Information Retrieval through Literature Review
LLM-based agent system for fine-grained information retrieval grounded in scientific literature for research queries.
4/28/2026
Beyond Single-Agent Alignment: Preventing Context-Fragmented Violations in Multi-Agent Systems
Framework addressing context-fragmented violations in multi-agent systems where distributed agent actions collectively violate policies.
4/28/2026
AutoPyVerifier: Learning Compact Executable Verifiers for Large Language Model Outputs
Framework for learning compact executable verifiers for LLM outputs combining interpretability with expressive capability.
4/28/2026
Large language model-enabled automated data extraction for concrete materials informatics
LLM-powered pipeline for automated extraction and structuring of materials science data from scientific literature.
4/28/2026
VS-DDPM: Efficient Low-Cost Diffusion Model for Medical Modality Translation
Variable-step diffusion model framework for efficient medical image translation with reduced computational cost.
4/28/2026
Rethinking Trust Region Bayesian Optimization in High Dimensions
Improved trust region Bayesian optimization strategy addressing lengthscale design issues in high-dimensional settings.
4/28/2026
AmaraSpatial-10K: A Spatially and Semantically Aligned 3D Dataset for Spatial Computing and Embodied AI
3D asset dataset of 10,000+ spatially and semantically aligned objects for embodied AI and robotics simulation.
4/28/2026
Self-Supervised Learning for Android Malware Detection on a Time-Stamped Dataset
Self-supervised learning framework for Android malware detection addressing temporal bias in training data.
4/28/2026
DeepImagine: Learning Biomedical Reasoning via Successive Counterfactual Imagining
Framework teaching LLMs biomedical reasoning through counterfactual imagining for clinical trial outcome prediction.
4/28/2026
MOCA: A Transformer-based Modular Causal Inference Framework with One-way Cross-attention and Cutting Feedback
Transformer-based framework for causal inference from observational data with improved handling of complex treatment mechanisms.
4/28/2026
Mixture of Heterogeneous Grouped Experts for Language Modeling
Mixture-of-Experts architecture with heterogeneous expert sizes for improved LLM scaling and efficiency.
4/28/2026
Fine-tuning vs. In-context Learning in Large Language Models: A Formal Language Learning Perspective
Formal language learning perspective comparing fine-tuning vs in-context learning in LLMs with controlled experimental setup.