2/17/2026
QuaRK: A Quantum Reservoir Kernel for Time Series Learning
QuaRK framework combines quantum reservoir computing with classical readout for time series learning with theoretical guarantees.
2/17/2026
Fast Swap-Based Element Selection for Multiplication-Free Dimension Reduction
Fast algorithm for multiplication-free dimension reduction via element selection as alternative to PCA for reducing model parameters and accelerating inference.
2/17/2026
Out-of-Support Generalisation via Weight Space Sequence Modelling
Research on out-of-support generalization reformulates extrapolation beyond training data distribution as sequence modeling in weight space to prevent catastrophic failure.
2/17/2026
Interpretable clustering via optimal multiway-split decision trees
Optimal multiway-split decision trees for interpretable clustering that balance accuracy and explainability while reducing computational costs.
2/17/2026
Benchmark Leakage Trap: Can We Trust LLM-based Recommendation?
Investigates benchmark data leakage in LLM-based recommender systems where models memorize training data, inflating evaluation metrics.
2/17/2026
Optimization-Free Graph Embedding via Distributional Kernel for Community Detection
Graph embedding method using distributional kernels for community detection without optimization, addressing over-smoothing in GNNs.
2/17/2026
Joint Time Series Chain: Detecting Unusual Evolving Trend across Time Series
Detects unusual evolving trends in time series using concept of time series chains capturing ordered temporal pattern changes.
2/17/2026
Cumulative Utility Parity for Fair Federated Learning under Intermittent Client Participation
Fairness approach for federated learning addressing cumulative utility parity under intermittent and heterogeneous client participation.
2/17/2026
Zero-Order Optimization for LLM Fine-Tuning via Learnable Direction Sampling
Zero-order optimization method for LLM fine-tuning via learnable direction sampling, reducing memory requirements without backpropagation.
2/17/2026
Optimized Certainty Equivalent Risk-Controlling Prediction Sets
Optimized certainty equivalent risk-controlling prediction sets for safety-critical applications with tail behavior and worst-case guarantees.
2/17/2026
Advancing Analytic Class-Incremental Learning through Vision-Language Calibration
Vision-language model calibration approach for class-incremental learning that addresses representation rigidity and accumulated errors in analytic CIL.
2/17/2026
Attention Head Entropy of LLMs Predicts Answer Correctness
Uses attention head entropy to predict answer correctness in LLMs, enabling detection of hallucinations without external evaluators.
2/17/2026
Optimal Regret for Policy Optimization in Contextual Bandits
First optimal regret bound for policy optimization in contextual multi-armed bandits with general offline function approximation.
2/17/2026
Near-Optimal Regret for Policy Optimization in Contextual MDPs with General Offline Function Approximation
OPO-CMDP presents first policy optimization algorithm for contextual MDPs with general offline function approximation achieving near-optimal regret bounds.
2/17/2026
HBVLA: Pushing 1-Bit Post-Training Quantization for Vision-Language-Action Models
HBVLA applies 1-bit post-training quantization to vision-language-action models for efficient deployment on resource-constrained robots and edge devices.
2/17/2026
Data-driven Bi-level Optimization of Thermal Power Systems with embedded Artificial Neural Networks
Bi-level optimization framework using neural networks for operational optimization of thermal power systems with hierarchical variables.
2/17/2026
On Representation Redundancy in Large-Scale Instruction Tuning Data Selection
Study of instruction-tuning data selection for LLMs using semantic representation similarity to identify redundancy in large-scale datasets.
2/17/2026
MEMTS: Internalizing Domain Knowledge via Parameterized Memory for Retrieval-Free Domain Adaptation of Time Series Foundation Models
MEMTS introduces parameterized memory for domain adaptation of time series foundation models to handle temporal distribution shifts and domain-specific patterns.
2/17/2026
MechPert: Mechanistic Consensus as an Inductive Bias for Unseen Perturbation Prediction
MechPert predicts transcriptional responses to unseen genetic perturbations using mechanistic consensus as inductive bias, combining knowledge graphs with LLM reasoning.
2/17/2026
Cast-R1: Learning Tool-Augmented Sequential Decision Policies for Time Series Forecasting
Cast-R1 applies tool-augmented sequential decision policies and iterative reasoning to time series forecasting, enabling autonomous evidence acquisition and prediction revision.
2/17/2026
AnomaMind: Agentic Time Series Anomaly Detection with Tool-Augmented Reasoning
AnomaMind uses agentic reasoning with tool augmentation for time series anomaly detection, framing it as evidence-driven diagnosis rather than fixed discriminative prediction.
2/17/2026
Pawsterior: Variational Flow Matching for Structured Simulation-Based Inference
Variational flow-matching framework for simulation-based inference respecting structured domains with discrete-continuous variables.
2/17/2026
GREPO: A Benchmark for Graph Neural Networks on Repository-Level Bug Localization
Graph neural network benchmark for repository-level bug localization using code graph structure beyond standard LLM context windows.
2/17/2026
Why Code, Why Now: Learnability, Computability, and the Real Limits of Machine Learning
Analysis comparing code generation learnability to reinforcement learning, proposing hierarchy of feedback quality as ML progress ceiling.
2/17/2026
A Multi-Agent Framework for Code-Guided, Modular, and Verifiable Automated Machine Learning
Multi-agent AutoML framework using LLM-based code generation with modular architecture to reduce hallucinations and improve verifiability.
2/17/2026
You Can Learn Tokenization End-to-End with Reinforcement Learning
End-to-end learnable tokenization for LLMs using reinforcement learning instead of hardcoded compression steps.
2/17/2026
Experiential Reinforcement Learning
Training paradigm embedding experience replay in reinforcement learning for LMs to learn from sparse, delayed environmental feedback.
2/17/2026
QuRL: Efficient Reinforcement Learning with Quantized Rollout
Quantized reinforcement learning for LLM training that accelerates rollout efficiency by 30% using quantized actor networks.
2/17/2026
Chemical Language Models for Natural Products: A State-Space Model Approach
State-space models (Mamba) applied to natural product chemistry for molecular property prediction and generation.
2/17/2026
EIDOS: Latent-Space Predictive Learning for Time Series Foundation Models
Foundation model for time series that uses latent-space predictive learning instead of direct future value prediction.
2/17/2026
Position Encoding with Random Float Sampling Enhances Length Generalization of Transformers
Position encoding technique using random float sampling to improve transformer length generalization beyond pretraining sequence lengths.
2/17/2026
Policy Gradient with Adaptive Entropy Annealing for Continual Fine-Tuning
Parameter-efficient fine-tuning for vision models using policy gradient with adaptive entropy annealing to prevent catastrophic forgetting in class-incremental learning.
2/17/2026
Sanity Checks for Sparse Autoencoders: Do SAEs Beat Random Baselines?
Sanity checks validating whether sparse autoencoders recover meaningful features beyond random baselines for neural network interpretability.
2/17/2026
ROAST: Rollout-based On-distribution Activation Steering Technique
ROAST uses on-distribution rollouts for parameter-efficient LLM activation steering at inference time, replacing off-distribution supervision with continuous soft scaling.
2/17/2026
Synergistic Intra- and Cross-Layer Regularization Losses for MoE Expert Specialization
Plug-and-play regularization losses for Mixture-of-Experts models promoting expert specialization across intra- and cross-layers without structural modifications.
2/17/2026
When Benchmarks Lie: Evaluating Malicious Prompt Classifiers Under True Distribution Shift
Comprehensive analysis of malicious prompt classifier robustness under distribution shift with 18 datasets spanning jailbreaks and prompt injections for LLM agents.
2/17/2026
Deep Dense Exploration for LLM Reinforcement Learning via Pivot-Driven Resampling
Pivot-driven resampling technique for deep dense exploration in LLM RL, discovering high-quality trajectories within limited sampling budget from language space.
2/17/2026
TS-Haystack: A Multi-Scale Retrieval Benchmark for Time Series Language Models
TS-Haystack benchmark evaluates time series language models on long-context retrieval with millions of datapoints, requiring precise temporal localization.
2/17/2026
Fast Catch-Up, Late Switching: Optimal Batch Size Scheduling via Functional Scaling Laws
Characterizes optimal batch size scheduling for large-scale deep learning under fixed data budget using functional scaling law framework.
2/17/2026
MAGE: All-[MASK] Block Already Knows Where to Look in Diffusion LLM
MAGE optimizes KV cache memory access in block diffusion LLMs for long-context settings using dynamic sparse attention adapted to block diffusion uniqueness.
2/17/2026
Robust multi-task boosting using clustering and local ensembling
RMB-CLE framework for multi-task learning integrating error-based task clustering with local ensembling to mitigate negative transfer from unrelated tasks.
2/17/2026
Evaluating LLMs in Finance Requires Explicit Bias Consideration
Analysis identifying five recurring biases in financial LLM applications: look-ahead, survivorship, narrative, objective, and cost bias that invalidate deployment claims.
2/17/2026
Multi-Agent Debate: A Unified Agentic Framework for Tabular Anomaly Detection
MAD framework treats tabular anomaly detection as multi-agent debate, leveraging disagreement from heterogeneous model families under distribution shift and rare-anomaly regimes.
2/17/2026
Cross-household Transfer Learning Approach with LSTM-based Demand Forecasting
Transfer learning approach using LSTM for cross-household hot water demand forecasting to optimize heat pump operation and reduce energy waste.
2/17/2026
Radial-VCReg: More Informative Representation Learning Through Radial Gaussianization
Radial-VCReg augments VCReg with radial Gaussianization loss for improved self-supervised representation learning by aligning feature norms with Chi distribution.
2/17/2026
Integrating Unstructured Text into Causal Inference: Empirical Evidence from Real Data
Framework leveraging transformer-based language models for causal inference from unstructured text, comparing estimates against structured data baselines.
2/17/2026
Reverse N-Wise Output-Oriented Testing for AI/ML and Quantum Computing Systems
Testing methodology for AI/ML and quantum systems addressing high-dimensional inputs, probabilistic outputs, and evaluation of trustworthiness, fairness, and robustness.
2/17/2026
Whom to Query for What: Adaptive Group Elicitation via Multi-Turn LLM Interactions
Framework for adaptive multi-turn LLM interactions to efficiently elicit group-level information from surveys, optimizing respondent selection and questioning strategy.
2/17/2026
KernelBlaster: Continual Cross-Task CUDA Optimization via Memory-Augmented In-Context Reinforcement Learning
KernelBlaster uses agentic workflows with in-context RL to optimize CUDA code across GPU architectures, aggregating knowledge from prior optimizations without expensive finetuning.
2/17/2026
Machine Learning as a Tool (MLAT): A Framework for Integrating Statistical ML Models as Callable Tools within LLM Agent Workflows
MLAT framework exposes pre-trained ML models as callable tools within LLM agent workflows, enabling agents to invoke quantitative predictions and reason about outputs contextually.