2/24/2026
ISO-Bench: Can Coding Agents Optimize Real-World Inference Workloads?
ISO-Bench benchmark evaluates coding agents on real-world LLM inference optimization tasks from vLLM and SGLang frameworks with 54 curated tasks.
2/24/2026
Is Your Diffusion Sampler Actually Correct? A Sampler-Centric Evaluation of Discrete Diffusion Language Models
Framework for evaluating discrete diffusion language models by separating sampler-induced error from denoiser approximation error.
2/24/2026
VecFormer: Towards Efficient and Generalizable Graph Transformer with Graph Token Attention
VecFormer improves graph transformers with token-level attention to reduce computational complexity and improve generalization.
2/24/2026
Compositional Planning with Jumpy World Models
Compositional planning with world models enabling agents to compose pre-trained policies for solving complex tasks.
2/24/2026
Representation Stability in a Minimal Continual Learning Agent
Study of representational dynamics in minimal continual learning agents across multiple task executions.
2/24/2026
PaReGTA: An LLM-based EHR Data Encoding Approach to Capture Temporal Information
PaReGTA encodes temporal electronic health records using LLMs with lightweight fine-tuning for domain adaptation.
2/24/2026
The Confusion is Real: GRAPHIC - A Network Science Approach to Confusion Matrices in Deep Learning
GRAPHIC: network science approach to visualize and analyze class confusion patterns in deep learning models.
2/24/2026
Bayesian Meta-Learning with Expert Feedback for Task-Shift Adaptation through Causal Embeddings
Bayesian meta-learning method using causal embeddings to improve adaptation to out-of-distribution tasks.
2/24/2026
Stop Preaching and Start Practising Data Frugality for Responsible Development of AI
Position paper advocating adoption of data frugal machine learning practices to reduce computational and environmental costs.
2/24/2026
I Dropped a Neural Net
Method to recover the correct ordering of shuffled neural network layers using only the dataset and layer information.
2/24/2026
Generalized Random Direction Newton Algorithms for Stochastic Optimization
Novel Hessian estimators using random direction stochastic approximation for optimization with noisy measurements.
2/24/2026
DSDR: Dual-Scale Diversity Regularization for Exploration in LLM Reasoning
DSDR: dual-scale diversity regularization method to improve exploration in LLM reasoning tasks with reinforcement learning from verifiers.
2/24/2026
Uncertainty-Aware Rank-One MIMO Q Network Framework for Accelerated Offline Reinforcement Learning
Proposes uncertainty-aware rank-one MIMO Q-network for offline reinforcement learning to address extrapolation error from out-of-distribution data.
2/24/2026
Rethinking LoRA for Privacy-Preserving Federated Learning in Large Models
Analysis of LoRA parameter-efficient fine-tuning in differentially private federated learning for large vision and language models addressing privacy-utility trade-off.
2/24/2026
Expanding the Role of Diffusion Models for Robust Classifier Training
Extends diffusion model role in robust classifier training beyond synthetic data generation to leverage internal feature representations for adversarial robustness.
2/24/2026
A Replicate-and-Quantize Strategy for Plug-and-Play Load Balancing of Sparse Mixture-of-Experts LLMs
Load balancing strategy for sparse mixture-of-experts LLMs addressing expert underutilization through replicate-and-quantize approach at inference time.
2/24/2026
DP-FedAdamW: An Efficient Optimizer for Differentially Private Federated Large Models
Efficient optimizer for differentially private federated learning of large models addressing variance amplification and bias from privacy noise.
2/24/2026
Sparse Masked Attention Policies for Reliable Generalization
Information removal method using sparse masked attention for RL policies that generalize reliably to unseen tasks and observations.
2/24/2026
On the Equivalence of Random Network Distillation, Deep Ensembles, and Bayesian Inference
Theoretical analysis establishing equivalence between random network distillation, deep ensembles, and Bayesian inference for uncertainty quantification.
2/24/2026
Unlearning Noise in PINNs: A Selective Pruning Framework for PDE Inverse Problems
Selective pruning framework for physics-informed neural networks to improve robustness against noise in PDE inverse problems.
2/24/2026
Discrete Diffusion Models Exploit Asymmetry to Solve Lookahead Planning Tasks
Analysis of how discrete diffusion models exploit asymmetry mechanisms for multi-step lookahead planning compared to autoregressive transformers.
2/24/2026
A Computationally Efficient Multidimensional Vision Transformer
Tensor-based Vision Transformer architecture exploiting multilinear structures for computational efficiency in image processing tasks.
2/24/2026
A Secure and Private Distributed Bayesian Federated Learning Design
Distributed federated learning framework addressing privacy leakage, slow convergence, and Byzantine robustness using Bayesian methods without central server.
2/24/2026
Learning Discriminative and Generalizable Anomaly Detector for Dynamic Graph with Limited Supervision
Semi-supervised approach for dynamic graph anomaly detection addressing label scarcity and poor generalization to unseen anomalies via discriminative learning.
2/24/2026
A Theory of How Pretraining Shapes Inductive Bias in Fine-Tuning
Theoretical analysis of how pretraining initializations impact feature learning and refinement during fine-tuning in modern deep learning systems.
2/24/2026
Training-Free Generative Modeling via Kernelized Stochastic Interpolants
Kernel method for generative modeling within stochastic interpolant framework replacing neural network training with linear systems solved from data.
2/24/2026
BarrierSteer: LLM Safety via Learning Barrier Steering
BarrierSteer framework for LLM safety using barrier steering to prevent adversarial attacks and unsafe content generation with theoretical guarantees.
2/24/2026
Reliable Abstention under Adversarial Injections: Tight Lower Bounds and New Upper Bounds
Theoretical study of online learning under adversarial injection with tight bounds on learner's ability to abstain from uncertain predictions in clean-label setting.
2/24/2026
Adaptation to Intrinsic Dependence in Diffusion Language Models
Theoretical analysis of diffusion language models examining how unmasking schedules affect parallel token generation quality compared to autoregressive approaches.
2/24/2026
LAD: Learning Advantage Distribution for Reasoning
New RL objective for LLM reasoning using advantage distribution matching to improve diversity and exploration beyond reward maximization in multi-step reasoning tasks.
2/24/2026
Behavior Learning (BL): Learning Hierarchical Optimization Structures from Data
Novel ML framework learning interpretable hierarchical optimization structures from data, unifying predictive performance with intrinsic interpretability for scientific applications.
2/24/2026
Replication Study: Federated Text-Driven Prompt Generation for Vision-Language Models
Replication study of federated learning approach for vision-language models using text-driven prompt generation to improve zero-shot generalization to unseen classes.
2/24/2026
Prompt Optimization Via Diffusion Language Models
Diffusion language models optimize system prompts via masked denoising without gradient access to base LLM.
2/24/2026
Asymptotic Semantic Collapse in Hierarchical Optimization
Studies failure mode in multi-agent language systems where dominant context absorbs individual agent semantics.
2/24/2026
The Story is Not the Science: Execution-Grounded Evaluation of Mechanistic Interpretability Research
AI agents as evaluators for mechanistic interpretability research, addressing reproducibility via execution-grounded testing.
2/24/2026
The Million-Label NER: Breaking Scale Barriers with GLiNER bi-encoder
GLiNER-bi-Encoder: Named entity recognition architecture decoupling label and context encoding for zero-shot flexibility with industrial-scale efficiency.
2/24/2026
DCInject: Persistent Backdoor Attacks via Frequency Manipulation in Personal Federated Learning
DCInject: Frequency-domain backdoor attack method for personalized federated learning exploiting DC component manipulation.
2/24/2026
RDBLearn: Simple In-Context Prediction Over Relational Databases
RDBLearn extends tabular in-context learning to relational databases, enabling single models to adapt to prediction tasks across linked tables.
2/24/2026
Hierarchical Reward Design from Language: Enhancing Alignment of Agent Behavior with Human Specifications
Framework for translating human language specifications into hierarchical reward functions to align AI agent behavior with human expectations.
2/24/2026
Luna-2: Scalable Single-Token Evaluation with Small Language Models
Luna-2: Single-token evaluation architecture using small language models for fast, deterministic LLM-as-judge metrics at scale.
2/24/2026
Enhancing Goal Inference via Correction Timing
Method for robots to infer task goals from user corrections by analyzing timing and content of behavioral feedback.
2/24/2026
Auto Quantum Machine Learning for Multisource Classification
Automated quantum machine learning pipeline for multisource classification tasks in remote sensing applications.
2/24/2026
Toward AI Autonomous Navigation for Mechanical Thrombectomy using Hierarchical Modular Multi-agent Reinforcement Learning (HM-MARL)
Hierarchical multi-agent reinforcement learning framework for autonomous navigation in endovascular mechanical thrombectomy procedures.
2/24/2026
Many AI Analysts, One Dataset: Navigating the Agentic Data Science Multiverse
Using AI agents to conduct multiple independent analyses on same dataset, demonstrating variability in research conclusions from analytic choices.
2/24/2026
MANATEE: Inference-Time Lightweight Diffusion Based Safety Defense for LLMs
MANATEE: inference-time defense against adversarial jailbreaks using density estimation on benign representation manifolds.
2/24/2026
Carbon-aware decentralized dynamic task offloading in MIMO-MEC networks via multi-agent reinforcement learning
Multi-agent reinforcement learning for dynamic task offloading in edge computing with renewable energy constraints.
2/24/2026
Operational Robustness of LLMs on Code Generation
Study of LLM robustness to natural language variations in code generation tasks, addressing sensitivity in developer applications.
2/24/2026
Habilis-$\beta$: A Fast-Motion and Long-Lasting On-Device Vision-Language-Action Model
Habilis-β: on-device vision-language-action model for real-world robotic deployment with long-duration evaluation metrics.
2/24/2026
Limits of Convergence-Rate Control for Open-Weight Safety
Theoretical analysis of convergence-rate control as defense mechanism against fine-tuning of open-weight models for harmful purposes.
2/24/2026
Could Large Language Models work as Post-hoc Explainability Tools in Credit Risk Models?
Investigation of using LLMs as post-hoc explainability tools for credit risk model predictions via in-context learning.