3/18/2026
Steering Frozen LLMs: Adaptive Social Alignment via Online Prompt Routing
Inference-time steering mechanism for frozen LLMs using adaptive prompt routing to enable evolving safety alignment without retraining.
3/18/2026
How to Achieve Prototypical Birth and Death for OOD Detection?
Prototype-based OOD detection method with dynamic prototype count adaptation based on category complexity.
3/18/2026
A federated learning framework with knowledge graph and temporal transformer for early sepsis prediction in multi-center ICUs
Federated learning framework combining knowledge graphs and temporal transformers for early sepsis prediction across multi-center ICUs.
3/18/2026
Discovering the Hidden Role of Gini Index In Prompt-based Classification
Study of Gini Index role in detecting and debiasing class accuracy disparities in prompt-based classification tasks.
3/18/2026
Beyond Reward Suppression: Reshaping Steganographic Communication Protocols in MARL via Dynamic Representational Circuit Breaking
Defense mechanism against steganographic collusion in multi-agent RL using dynamic representational circuit breaking at optimization substrate.
3/18/2026
Attribution-Guided Model Rectification of Unreliable Neural Network Behaviors
Attribution-guided framework using rank-one model editing to rectify unreliable neural network behavior on non-robust features.
3/18/2026
Spectral Edge Dynamics of Training Trajectories: Signal--Noise Geometry Across Scales
Analysis of transformer training dynamics via spectral edge detection showing parameter updates concentrate in few coherent directions.
3/18/2026
Evidential Domain Adaptation for Remaining Useful Life Prediction with Incomplete Degradation
Domain adaptation method for remaining useful life prediction with incomplete degradation trajectories using evidential learning.
3/18/2026
Tackling Over-smoothing on Hypergraphs: A Ricci Flow-guided Neural Diffusion Approach
Hypergraph neural network approach using Ricci flow to address over-smoothing and improve message passing.
3/18/2026
Mastering the Minority: An Uncertainty-guided Multi-Expert Framework for Challenging-tailed Sequence Learning
Multi-expert framework with uncertainty guidance for imbalanced sequence learning and minority class detection.
3/18/2026
Embedding-Aware Feature Discovery: Bridging Latent Representations and Interpretable Features in Event Sequences
Method bridging learned embeddings and interpretable handcrafted features for temporal event sequences in financial systems.
3/18/2026
Meta-TTRL: A Metacognitive Framework for Self-Improving Test-Time Reinforcement Learning in Unified Multimodal Models
Metacognitive test-time reinforcement learning framework for unified multimodal models enabling knowledge accumulation across similar prompts.
3/18/2026
OMNIFLOW: A Physics-Grounded Multimodal Agent for Generalized Scientific Reasoning
Physics-grounded multimodal LLM agent combining language models with PDE solvers for scientific reasoning without domain-specific fine-tuning.
3/18/2026
Time-Aware Prior Fitted Networks for Zero-Shot Forecasting with Exogenous Variables
Zero-shot forecasting method for time series with exogenous variables using prior-fitted networks.
3/18/2026
Mask Is What DLLM Needs: A Masked Data Training Paradigm for Diffusion LLMs
Masked data training paradigm for discrete diffusion language models using information density-driven noise scheduling.
3/18/2026
Hypothesis Class Determines Explanation: Why Accurate Models Disagree on Feature Attribution
Empirical study showing prediction-equivalent ML models produce substantially different feature attributions across 24 datasets, challenging assumptions in explainable AI.
3/18/2026
When Stability Fails: Hidden Failure Modes Of LLMS in Data-Constrained Scientific Decision-Making
Evaluates LLM failure modes in scientific decision-making when stability doesn't guarantee agreement with statistical ground truth.
3/18/2026
Informationally Compressive Anonymization: Non-Degrading Sensitive Input Protection for Privacy-Preserving Supervised Machine Learning
Privacy-preserving machine learning technique using informational compression for anonymization without performance degradation.
3/18/2026
Evaluating Black-Box Vulnerabilities with Wasserstein-Constrained Data Perturbations
Applies optimal transport theory to evaluate ML model vulnerabilities through Wasserstein-constrained data perturbations.
3/18/2026
Counteractive RL: Rethinking Core Principles for Efficient and Scalable Deep Reinforcement Learning
Proposes counteractive reinforcement learning approach addressing computational complexity in high-dimensional MDPs.
3/18/2026
PhasorFlow: A Python Library for Unit Circle Based Computing
Python library for unit circle based computing using complex phasors and unitary wave interference gates.
3/18/2026
Game-Theory-Assisted Reinforcement Learning for Border Defense: Early Termination based on Analytical Solutions
Hybrid approach combining game theory and reinforcement learning for adversarial scenarios using analytical solutions for early termination.
3/18/2026
The Agentic Researcher: A Practical Guide to AI-Assisted Research in Mathematics and Machine Learning
Practical guide for using AI systems and agents in mathematics and machine learning research with discussion of responsible guardrails.
3/18/2026
Auto Researching, not hyperparameter tuning: Convergence Analysis of 10,000 Experiments
Analyzes 10,469 experiments from LLM agents performing architecture search vs hyperparameter tuning using ANOVA decomposition.
3/18/2026
Generative Inverse Design with Abstention via Diagonal Flow Matching
Introduces diagonal flow matching for inverse design problems with better stability than conditional flow matching.
3/18/2026
Discovery of interaction and diffusion kernels in particle-to-mean-field multi-agent systems
Data-driven framework learning interaction kernels in stochastic multi-agent systems via sparse regression on trajectory data.
3/18/2026
Data-Local Autonomous LLM-Guided Neural Architecture Search for Multiclass Multimodal Time-Series Classification
LLM-guided neural architecture search for time-series classification in privacy-constrained domains using data-local constraints.
3/18/2026
MobileLLM-Flash: Latency-Guided On-Device LLM Design for Industry Scale
Hardware-in-the-loop architecture search methodology for designing efficient on-device LLMs with real-time latency constraints for mobile deployment.
3/18/2026
GASP: Guided Asymmetric Self-Play For Coding LLMs
Proposes guided asymmetric self-play method for post-training coding LLMs with better problem selection to improve model capabilities.
3/18/2026
Deriving Hyperparameter Scaling Laws via Modern Optimization Theory
Derives hyperparameter scaling laws for modern optimizers enabling transfer across model sizes, batch sizes, and training horizons.
3/18/2026
W2T: LoRA Weights Already Know What They Can Do
Analyzes whether LoRA checkpoint weights encode task performance information readable without running the base model, enabling efficient adapter analysis.
3/18/2026
Collaborative Temporal Feature Generation via Critic-Free Reinforcement Learning for Cross-User Sensor-Based Activity Recognition
Reinforcement learning approach for temporal feature generation in cross-user activity recognition from wearable sensor data.
3/18/2026
MDM-Prime-v2: Binary Encoding and Index Shuffling Enable Compute-optimal Scaling of Diffusion Language Models
Masked diffusion model optimization using binary encoding and index shuffling for improved scaling of diffusion language models.
3/18/2026
Noisy Data is Destructive to Reinforcement Learning with Verifiable Rewards
Analysis showing noisy data significantly degrades reinforcement learning with verifiable rewards despite claims of robustness.
3/18/2026
HIPO: Instruction Hierarchy via Constrained Reinforcement Learning
Constrained reinforcement learning approach for hierarchical instruction following in LLMs with priority-ordered system prompts.
3/18/2026
DyJR: Preserving Diversity in Reinforcement Learning with Verifiable Rewards via Dynamic Jensen-Shannon Replay
Experience replay mechanism preserving diversity in on-policy reinforcement learning for LLM reasoning using Jensen-Shannon divergence.
3/18/2026
Execution-Grounded Credit Assignment for GRPO in Code Generation
Credit assignment method using execution traces to improve GRPO performance in code generation tasks with verifiable rewards.
3/18/2026
The Finetuner's Fallacy: When to Pretrain with Your Finetuning Data
Study of specialized pretraining strategy using domain data during pretraining to improve finetuning performance and reduce forgetting.
3/18/2026
Offline Exploration-Aware Fine-Tuning for Long-Chain Mathematical Reasoning
Fine-tuning approach for improving mathematical reasoning in LLMs by optimizing exploration-aware trajectories with verifiable rewards.
3/18/2026
Dual Consensus: Escaping from Spurious Majority in Unsupervised RLVR via Two-Stage Vote Mechanism
Dual consensus mechanism for improving reinforcement learning from verifiable rewards in LLMs, avoiding convergence to spurious answers.
3/18/2026
Decoding the Critique Mechanism in Large Reasoning Models
Study of how large reasoning models use backtracking and self-verification to detect and correct errors in complex logical reasoning tasks.
3/18/2026
Behavioral Steering in a 35B MoE Language Model via SAE-Decoded Probe Vectors: One Agency Axis, Not Five Traits
Research on steering behavior in 35B MoE language models using sparse autoencoders and probe vectors to identify and control agentic traits.
3/18/2026
DynamicGate MLP Conditional Computation via Learned Structural Dropout and Input Dependent Gating for Functional Plasticity
MLP architecture with learned structural dropout and input-dependent gating for conditional computation and regularization.
3/18/2026
FederatedFactory: Generative One-Shot Learning for Extremely Non-IID Distributed Scenarios
Federated learning framework for non-IID distributed scenarios using generative one-shot learning without foundation model dependencies.
3/18/2026
Prior-Informed Neural Network Initialization: A Spectral Approach for Function Parameterizing Architectures
Spectral initialization method for neural networks designed for function parameterization using prior information.
3/18/2026
Trained Persistent Memory for Frozen Encoder--Decoder LLMs: Six Architectural Methods
Methods for adding persistent memory to frozen encoder-decoder LLMs using continuous latent space adapters for multi-session learning.
3/18/2026
DISCOVER: A Solver for Distributional Counterfactual Explanations
Solver for distributional counterfactual explanations using optimal transport with statistical certification for model interpretability.
3/18/2026
Capability-Guided Compression: Toward Interpretability-Aware Budget Allocation for Large Language Models
LLM compression method using capability-guided budget allocation that interprets what model components encode before pruning.
3/18/2026
Bridging the High-Frequency Data Gap: A Millisecond-Resolution Network Dataset for Advancing Time Series Foundation Models
High-frequency time series dataset at millisecond resolution for training and evaluating time series foundation models.
3/18/2026
From the Inside Out: Progressive Distribution Refinement for Confidence Calibration
Test-time scaling and confidence calibration strategy using internal model information for improved reinforcement learning.