3/26/2026
StateLinFormer: Stateful Training Enhancing Long-term Memory in Navigation
StateLinFormer: linear-attention navigation model with persistent memory for long-term navigation tasks, combining flexibility with efficiency.
3/26/2026
Dual-Criterion Curriculum Learning: Application to Temporal Data
Dual-Criterion Curriculum Learning proposes a meta-learning approach using dual criteria for difficulty assessment in temporal data training.
3/26/2026
APreQEL: Adaptive Mixed Precision Quantization For Edge LLMs
APreQEL proposes adaptive mixed precision quantization to reduce memory and computational costs of LLMs for edge device deployment while maintaining performance.
3/26/2026
Wafer-Level Etch Spatial Profiling for Process Monitoring from Time-Series with Time-LLM
Time-LLM model for predicting wafer-level spatial etch depth distributions in plasma etching process monitoring.
3/26/2026
LLMORPH: Automated Metamorphic Testing of Large Language Models
LLMORPH automated testing tool for LLMs using metamorphic testing to detect NLP task failures without human-labeled oracles.
3/26/2026
LLMLOOP: Improving LLM-Generated Code and Tests through Automated Iterative Feedback Loops
LLMLOOP framework automating iterative refinement of LLM-generated code and test cases through automated feedback loops.
3/26/2026
A Theory of LLM Information Susceptibility
Theory of LLM information susceptibility analyzing fundamental limits of LLM-mediated optimization in agentic systems.
3/26/2026
Swiss-Bench SBP-002: A Frontier Model Comparison on Swiss Legal and Regulatory Tasks
Swiss-Bench SBP-002: trilingual benchmark of 395 expert-crafted regulatory compliance tasks across FINMA, Legal-CH, and EFK domains.
3/26/2026
Probing Ethical Framework Representations in Large Language Models: Structure, Entanglement, and Methodological Challenges
Probing study revealing how LLMs internally represent different ethical frameworks with asymmetric transfer patterns across model sizes.
3/26/2026
Prototype Fusion: A Training-Free Multi-Layer Approach to OOD Detection
Training-free out-of-distribution detection using multi-layer prototype fusion approach for robust deep learning deployment.
3/26/2026
PLACID: Privacy-preserving Large language models for Acronym Clinical Inference and Disambiguation
Privacy-preserving LLM system for disambiguating clinical acronyms in healthcare without transmitting data to external servers.
3/26/2026
Assessment Design in the AI Era: A Method for Identifying Items Functioning Differentially for Humans and Chatbots
Measurement methodology for identifying assessment items where LLMs perform differently than humans using theory-grounded evaluation.
3/26/2026
The Diminishing Returns of Early-Exit Decoding in Modern LLMs
Analysis of early-exit decoding in modern LLMs showing reduced efficiency gains due to improved architectures with lower layer redundancy.
3/26/2026
An In-Depth Study of Filter-Agnostic Vector Search on a PostgreSQL Database System: [Experiments and Analysis]
Study of filtered vector search algorithms in PostgreSQL for semantic search and GenAI applications, evaluating real-world database performance.
3/26/2026
Self Paced Gaussian Contextual Reinforcement Learning
Self-paced curriculum learning for RL using closed-form Gaussian updates to improve efficiency in high-dimensional contexts.
3/26/2026
AI-driven Intent-Based Networking Approach for Self-configuration of Next Generation Networks
Intent-Based Networking using AI to translate high-level natural language intents into network policies with automated compliance assurance.
3/26/2026
Probabilistic Geometric Alignment via Bayesian Latent Transport for Domain-Adaptive Foundation Models
Bayesian latent transport framework for domain-adaptive foundation models addressing distribution mismatch and uncertainty propagation in limited-supervision scenarios.
3/26/2026
The Cognitive Firewall:Securing Browser Based AI Agents Against Indirect Prompt Injection Via Hybrid Edge Cloud Defense
Cognitive Firewall: hybrid edge-cloud architecture for securing browser-based LLM agents against indirect prompt injection attacks using split-compute security checks.
3/26/2026
Object Search in Partially-Known Environments via LLM-informed Model-based Planning and Prompt Selection
LLM-informed model-based planning for object search using LLM likelihood estimates and environment costs.
3/26/2026
Deep Neural Regression Collapse
Neural Regression Collapse phenomenon across network layers showing feature sparsity and low rank structures.
3/26/2026
Willful Disobedience: Automatically Detecting Failures in Agentic Traces
AgentPex: Framework for detecting procedural failures in agentic traces including workflow routing and tool usage violations.
3/26/2026
Perturbation: A simple and efficient adversarial tracer for representation learning in language models
Method for finding representations in language models via adversarial perturbation without implausible constraints.
3/26/2026
PoliticsBench: Benchmarking Political Values in Large Language Models with Multi-Turn Roleplay
Benchmark (PoliticsBench) measuring political bias in eight LLMs using multi-turn roleplay evaluation.
3/26/2026
Why the Maximum Second Derivative of Activations Matters for Adversarial Robustness
Research on activation function curvature role in adversarial robustness using Recursive Curvature-Tunable Activation Family.
3/26/2026
Generative AI User Experience: Developing Human--AI Epistemic Partnership
Discussion of user experience design for generative AI in education emphasizing human-AI epistemic partnership.
3/26/2026
Can VLMs Reason Robustly? A Neuro-Symbolic Investigation
Investigation of vision-language model robustness under distribution shifts using visual deductive reasoning tasks.
3/26/2026
HDPO: Hybrid Distillation Policy Optimization via Privileged Self-Distillation
HDPO method augmenting RL with privileged self-distillation for LLM mathematical reasoning on unsolvable cliff prompts.
3/26/2026
The Luna Bound Propagator for Formal Analysis of Neural Networks
Luna: C++ implementation of alpha-CROWN bound propagation for neural network formal verification.
3/26/2026
AgentChemist: A Multi-Agent Experimental Robotic Platform Integrating Chemical Perception and Precise Control
Multi-agent robotic platform using AI agents for adaptive chemical laboratory automation handling diverse experimental tasks.
3/26/2026
Self-Distillation for Multi-Token Prediction
Self-distillation method for multi-token prediction in LLMs to improve inference efficiency and MTP head acceptance rates.
3/26/2026
DecepGPT: Schema-Driven Deception Detection with Multicultural Datasets and Robust Multimodal Learning
Multimodal deception detection system using schema-driven approach with multicultural datasets and explainable reasoning.
3/26/2026
Revealing Multi-View Hallucination in Large Vision-Language Models
MVH-Bench dataset and analysis of multi-view hallucination in vision-language models processing diverse viewpoint images.
3/26/2026
Policy-Guided Threat Hunting: An LLM enabled Framework with Splunk SOC Triage
LLM-enabled framework for automated threat hunting using Splunk SOC logs to assist security analysts with APT detection.
3/26/2026
The Price Reversal Phenomenon: When Cheaper Reasoning Models End Up Costing More
Systematic study of reasoning LLM inference costs revealing pricing reversal phenomenon where cheaper models cost more across 9 diverse tasks.
3/26/2026
SafeFlow: Real-Time Text-Driven Humanoid Whole-Body Control via Physics-Guided Rectified Flow and Selective Safety Gating
Physics-guided text-to-motion framework for humanoid control using rectified flow and safety gating to prevent kinematic hallucinations.
3/26/2026
From Untamed Black Box to Interpretable Pedagogical Orchestration: The Ensemble of Specialized LLMs Architecture for Adaptive Tutoring
Ensemble of specialized LLMs architecture for adaptive tutoring that separates pedagogical decision-making from response generation.
3/26/2026
Understanding the Challenges in Iterative Generative Optimization with LLMs
Analysis of challenges in iterative generative optimization with LLMs for self-improving agents, identifying hidden design choices limiting adoption.
3/26/2026
Schema on the Inside: A Two-Phase Fine-Tuning Method for High-Efficiency Text-to-SQL at Scale
Fine-tuned 8B model for text-to-SQL at scale, reducing API costs and latency for production deployment in conversational applications.
3/26/2026
From Oracle to Noisy Context: Mitigating Contextual Exposure Bias in Speech-LLMs
Training framework addressing contextual exposure bias in speech-LLMs using teacher error knowledge and contrastive learning.
3/26/2026
Mitigating Object Hallucinations in LVLMs via Attention Imbalance Rectification
Method to reduce object hallucinations in LVLMs by rectifying attention imbalance across and within vision-language modalities.
3/26/2026
When Understanding Becomes a Risk: Authenticity and Safety Risks in the Emerging Image Generation Paradigm
Safety analysis of MLLMs for image generation, identifying semantic understanding capabilities that may introduce new risks compared to diffusion models.
3/26/2026
Knowledge-Guided Manipulation Using Multi-Task Reinforcement Learning
Multi-task robotic manipulation framework using knowledge graphs and dynamic relation mechanisms for vision-grounded policy learning.
3/26/2026
Towards Effective Experiential Learning: Dual Guidance for Utilization and Internalization
Dual-guidance RL framework for LLMs that combines external execution feedback with internal experience for improved reasoning task learning.
3/26/2026
Comparative analysis of dual-form networks for live land monitoring using multi-modal satellite image time series
Comparative study of dual-form attention networks for multi-modal satellite time series analysis in land monitoring applications.
3/26/2026
The Alignment Tax: Response Homogenization in Aligned LLMs and Its Implications for Uncertainty Estimation
Analysis of response homogenization in RLHF-aligned LLMs and its impact on uncertainty estimation methods, identifying alignment-robustness tradeoffs.
3/26/2026
MedAidDialog: A Multilingual Multi-Turn Medical Dialogue Dataset for Accessible Healthcare
Multilingual multi-turn medical dialogue dataset for training conversational AI systems in healthcare with improved realism and accessibility.
3/26/2026
A Deep Dive into Scaling RL for Code Generation with Synthetic Data and Curricula
Scaling RL for LLM code generation using synthetic data pipelines and curriculum learning, addressing data diversity over volume.
3/26/2026
Invisible Threats from Model Context Protocol: Generating Stealthy Injection Payload via Tree-based Adaptive Search
Security analysis of Model Context Protocol (MCP) tool-augmented LLM agents, demonstrating stealthy injection attacks on tool responses.
3/26/2026
Powerful Teachers Matter: Text-Guided Multi-view Knowledge Distillation with Visual Prior Enhancement
Knowledge distillation method using dual-modality (vision + text/CLIP) teacher models to improve student model efficiency and quality.
3/26/2026
Uncovering Memorization in Timeseries Imputation models: LBRM Membership Inference and its link to attribute Leakage
Privacy analysis of time series imputation models, demonstrating membership inference and attribute leakage vulnerabilities in black-box settings.