2/17/2026
Detecting Jailbreak Attempts in Clinical Training LLMs Through Automated Linguistic Feature Extraction
Automated linguistic feature extraction for detecting jailbreak attempts in clinical training LLMs.
2/17/2026
Diagnostic Benchmarks for Invariant Learning Dynamics: Empirical Validation of the Eidos Architecture
PolyShapes-Ideal benchmark dataset tests topological invariance in vision models under affine transformations.
2/17/2026
Ask the Expert: Collaborative Inference for Vision Transformers with Near-Edge Accelerators
Collaborative inference framework routing vision transformer queries between edge and near-edge accelerators.
2/17/2026
Nonparametric Distribution Regression Re-calibration
Distribution regression re-calibration for ensuring predictive uncertainty reflects empirical accuracy.
2/17/2026
MoralityGym: A Benchmark for Evaluating Hierarchical Moral Alignment in Sequential Decision-Making Agents
MoralityGym: benchmark for evaluating hierarchical moral alignment in sequential decision-making agents.
2/17/2026
LAF-YOLOv10 with Partial Convolution Backbone, Attention-Guided Feature Pyramid, Auxiliary P2 Head, and Wise-IoU Loss for Small Object Detection in Drone Aerial Imagery
LAF-YOLOv10 detector for small object detection in aerial drone imagery using modified YOLOv10.
2/17/2026
Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents
Multi-turn safety benchmark for tool-using LLM agents evaluating hierarchical risks in sequential interactions.
2/17/2026
FUTON: Fourier Tensor Network for Implicit Neural Representations
FUTON: Fourier tensor network for implicit neural representations using low-rank decomposition.
2/17/2026
Protect$^*$: Steerable Retrosynthesis through Neuro-Symbolic State Encoding
Protect*: neuro-symbolic method for steerable retrosynthesis controlling LLM chemical pathway generation.
2/17/2026
Language Model Memory and Memory Models for Language
Analysis of information storage in language model embeddings versus autoencoders for memory characterization.
2/17/2026
AsyncVLA: An Asynchronous VLA for Fast and Robust Navigation on the Edge
AsyncVLA: asynchronous framework for fast robotic navigation decoupling semantic reasoning from reactive control.
2/17/2026
Stochastic variance reduced extragradient methods for solving hierarchical variational inequalities
Stochastic variance reduced methods for solving hierarchical variational inequalities and optimization.
2/17/2026
Learning Gradient Flow: Using Equation Discovery to Accelerate Engineering Optimization
Data-driven equation discovery for modeling gradient descent dynamics to accelerate optimization.
2/17/2026
SpargeAttention2: Trainable Sparse Attention via Hybrid Top-k+Top-p Masking and Distillation Fine-Tuning
Trainable sparse attention via hybrid Top-k+Top-p masking for accelerating diffusion model inference.
2/17/2026
On Calibration of Large Language Models: From Response To Capability
LLM calibration from response-level to capability-level confidence estimation for reliable deployment.
2/17/2026
LiveNewsBench: Evaluating LLM Web Search Capabilities with Freshly Curated News
LiveNewsBench: benchmark for evaluating LLM web search and agentic capabilities with fresh news data.
2/17/2026
Differentiable Rule Induction from Raw Sequence Inputs
Differentiable inductive logic programming for rule learning from raw sequence data.
2/17/2026
Towards Sparse Video Understanding and Reasoning
ReViS: multi-round agent for video question answering with selective frame sampling and early stopping.
2/17/2026
Multi-Modal Sensing and Fusion in mmWave Beamforming for Connected Vehicles: A Transformer Based Framework
Transformer-based mmWave beamforming for vehicular communication using multi-modal sensing.
2/17/2026
DiffusionRollout: Uncertainty-Aware Rollout Planning in Long-Horizon PDE Solving
Uncertainty-aware rollout planning for diffusion models in long-horizon PDE solving.
2/17/2026
Locally Private Parametric Methods for Change-Point Detection
Parametric change-point detection in time series under local differential privacy constraints.
2/17/2026
LEAD-Drift: Real-time and Explainable Intent Drift Detection by Learning a Data-Driven Risk Score
Intent drift detection in network management using risk scoring and machine learning.
2/17/2026
No Need to Train Your RDB Foundation Model
Foundation model using in-context learning for relational databases that avoids retraining across different prediction targets.
2/17/2026
Fine-tuned Vision Language Model for Localization of Parasitic Eggs in Microscopic Images
Fine-tunes vision-language model to localize parasitic eggs in microscopic images for soil-transmitted helminth diagnostic support.
2/17/2026
NeuroMambaLLM: Dynamic Graph Learning of fMRI Functional Connectivity in Autistic Brains Using Mamba and Language Model Reasoning
Combines Mamba state-space models with LLM reasoning to analyze dynamic fMRI functional connectivity in autistic brains.
2/17/2026
Attention in Constant Time: Vashista Sparse Attention for Long-Context Decoding with Exponential Guarantees
Sparse attention mechanism with constant-time complexity for long-context LLM decoding using projection onto convex hull of keys.
2/17/2026
A Unified Physics-Informed Neural Network for Modeling Coupled Electro- and Elastodynamic Wave Propagation Using Three-Stage Loss Optimization
Physics-Informed Neural Networks for modeling coupled electro-elastodynamic wave propagation with three-stage loss optimization.
2/17/2026
VAR-3D: View-aware Auto-Regressive Model for Text-to-3D Generation via a 3D Tokenizer
Auto-regressive transformer for text-to-3D generation using discrete 3D tokenizer to address information loss in encoding.
2/17/2026
Causally constrained reduced-order neural models of complex turbulent dynamical systems
Causal constraints framework using response theory and score matching for reduced-order neural emulators of turbulent dynamical systems.
2/17/2026
Evolving Multi-Channel Confidence-Aware Activation Functions for Missing Data with Channel Propagation
Proposes evolved activation functions that account for missing data indicators and confidence scores in neural networks.
2/17/2026
Enabling Option Learning in Sparse Rewards with Hindsight Experience Replay
Integrates Hindsight Experience Replay into Option-Critic hierarchical RL to improve multi-goal learning in sparse reward environments.
2/17/2026
Ensemble-Conditional Gaussian Processes (Ens-CGP): Representation, Geometry, and Inference
Formulation of Ensemble-Conditional Gaussian Processes connecting ensemble methods with conditional Gaussian inference and Kalman filtering.
2/17/2026
Ambient Physics: Training Neural PDE Solvers with Partial Observations
Framework for training neural PDE solvers on partial observations using diffusion models, avoiding need for complete observation datasets.
2/17/2026
Parameter-Efficient Fine-Tuning of DINOv2 for Large-Scale Font Classification
Fine-tunes DINOv2 Vision Transformer with LoRA for font classification, achieving 86% accuracy while training <1% of 87.2M parameters.
2/17/2026
RPGD: RANSAC-P3P Gradient Descent for Extrinsic Calibration in 3D Human Pose Estimation
Calibration framework (RANSAC-P3P Gradient Descent) for aligning MoCap skeletal data with RGB camera views using human motion.
2/17/2026
Quantifying Normality: Convergence Rate to Gaussian Limit for Stochastic Approximation and Unadjusted OU Algorithm
Non-asymptotic convergence rate analysis for stochastic approximation algorithms using Wasserstein distance bounds.
2/17/2026
MamaDino: A Hybrid Vision Model for Breast Cancer 3-Year Risk Prediction
Hybrid vision model combining Mamba and DINO architectures for breast cancer 3-year risk prediction from imaging data.
2/17/2026
Statistical Early Stopping for Reasoning Models
Statistical early stopping methods for LLM reasoning that monitor uncertainty signals to prevent overthinking during generation.
2/17/2026
A Theoretical Framework for LLM Fine-tuning Using Early Stopping for Non-random Initialization
Theoretical framework explaining why LLM fine-tuning requires only few epochs, combining early stopping theory with Neural Tangent Kernel analysis.
2/17/2026
Cognitive Chunking for Soft Prompts: Accelerating Compressor Learning via Block-wise Causal Masking
Method for compressing long LLM contexts into soft prompts via block-wise causal masking to reduce inference latency from quadratic attention costs.
2/17/2026
Named Entity Recognition for Payment Data Using NLP
Research on NER algorithms (CRF, BiLSTM-CRF, transformers) for extracting structured information from payment transaction data.
2/17/2026
Computable Bernstein Certificates for Cross-Fitted Clipped Covariance Estimation
Research on robust covariance estimation from heavy-tailed samples with outliers using clipped Euclidean norm approach and computable Bernstein certificates.
2/17/2026
Why Self-Training Helps and Hurts: Denoising vs. Signal Forgetting
Theoretical analysis of iterative self-training showing tradeoffs between noise denoising and signal forgetting in overparameterized linear regression.
2/17/2026
MC$^2$Mark: Distortion-Free Multi-Bit Watermarking for Long Messages
MC²Mark: distortion-free multi-bit watermarking framework for embedding long provenance identifiers in LLM-generated text.
2/17/2026
Choosing How to Remember: Adaptive Memory Structures for LLM Agents
Adaptive memory structures for LLM agents enabling context-dependent memory selection across heterogeneous interaction patterns.
2/17/2026
Geometry-Preserving Aggregation for Mixture-of-Experts Embedding Models
Geometry-preserving aggregation for mixture-of-experts embedding models accounting for hyperspherical manifold structure of expert outputs.
2/17/2026
GTS: Inference-Time Scaling of Latent Reasoning with a Learnable Gaussian Thought Sampler
GTS: learnable Gaussian thought sampler for efficient inference-time scaling in latent reasoning models with structured exploration.
2/17/2026
GUI-GENESIS: Automated Synthesis of Efficient Environments with Verifiable Rewards for GUI Agent Post-Training
GUI-GENESIS: framework for automated synthesis of efficient GUI training environments with verifiable rewards for agent post-training.
2/17/2026
Algebraic Quantum Intelligence: A New Framework for Reproducible Machine Creativity
Algebraic quantum intelligence framework proposing quantum computing approach to improve creative output generation in LLMs.
2/17/2026
DenseMLLM: Standard Multimodal LLMs are Intrinsic Dense Predictors
DenseMLLM: framework enabling multimodal LLMs for dense prediction tasks like segmentation and depth estimation without task-specific decoders.