3/18/2026
A Family of LLMs Liberated from Static Vocabularies
arXiv: LLM family with dynamic tokenizers eliminating fixed vocabulary constraints, up to 70B parameters, improved domain/language adaptation.
3/18/2026
MobileLLM-Flash: Latency-Guided On-Device LLM Design for Industry Scale
MobileLLM-Flash methodology designs on-device LLMs optimized for latency constraints using hardware-in-the-loop architecture search.
3/18/2026
ExpertGen: Scalable Sim-to-Real Expert Policy Learning from Imperfect Behavior Priors
ExpertGen automates expert policy generation in simulation for scalable sim-to-real robotic behavior cloning transfer.
3/18/2026
MoLoRA: Composable Specialization via Per-Token Adapter Routing
MoLoRA enables per-token adapter routing for multimodal generation and mixed-capability requests in multi-adapter serving.
3/18/2026
100x Cost & Latency Reduction: Performance Analysis of AI Query Approximation using Lightweight Proxy Models
Lightweight proxy models reduce LLM query costs and latency 100x for AI-augmented SQL operations.
3/18/2026
Standardizing Medical Images at Scale for AI
Physics-based preprocessing framework standardizes heterogeneous medical images at scale for improved model generalization.
3/18/2026
Aligning Paralinguistic Understanding and Generation in Speech LLMs via Multi-Task Reinforcement Learning
Multi-task RL with chain-of-thought prompting aligns paralinguistic understanding and generation in speech LLMs.
3/18/2026
RadAnnotate: Large Language Models for Efficient and Reliable Radiology Report Annotation
RadAnnotate uses LLMs with retrieval augmentation and selective automation for efficient radiology report annotation.
3/18/2026
Evaluating Agentic Optimization on Large Codebases
FormulaCode benchmark evaluates LLM coding agents on repository-level codebase optimization with realistic multi-objective constraints.
3/18/2026
Understanding Moral Reasoning Trajectories in Large Language Models: Toward Probing-Based Explainability
Probing-based analysis of moral reasoning trajectories in LLMs across six models showing systematic multi-framework deliberation.
3/18/2026
Collaborative Temporal Feature Generation via Critic-Free Reinforcement Learning for Cross-User Sensor-Based Activity Recognition
Critic-free RL approach for cross-user activity recognition from wearable sensors with temporal feature generation.
3/18/2026
Large Reward Models: Generalizable Online Robot Reward Generation with Vision-Language Models
Framework adapts vision-language models as online reward generators for robotic reinforcement learning policy refinement.
3/18/2026
Resource Consumption Threats in Large Language Models
Survey of resource consumption threats in LLMs including excessive generation, covering efficiency challenges for providers and users.
3/18/2026
Towards the Vision-Sound-Language-Action Paradigm: The HEAR Framework for Sound-Centric Manipulation
HEAR framework extends vision-language-action models to incorporate real-time sound for robotic manipulation tasks.
3/18/2026
RecBundle: A Next-Generation Geometric Paradigm for Explainable Recommender Systems
RecBundle proposes geometric framework for recommender systems addressing information cocoons through topological representation learning.
3/18/2026
CounterRefine: Answer-Conditioned Counterevidence Retrieval for Inference-Time Knowledge Repair in Factual Question Answering
Inference-time repair layer for retrieval-grounded QA using answer-conditioned counterevidence retrieval to fix commitment errors.
3/18/2026
Parallel In-context Learning for Large Vision Language Models
Parallel in-context learning method reducing latency in vision-language models by decoupling demonstration processing from query encoding.
3/18/2026
Efficient LLM Serving for Agentic Workflows: A Data Systems Perspective
LLM serving system optimizing agentic workflows by handling cross-call dependencies and redundancy from speculative execution.
3/18/2026
Frequency Matters: Fast Model-Agnostic Data Curation for Pruning and Quantization
Data curation method for calibration in LLM compression via frequency-based selection for pruning and quantization.
3/18/2026
RepoReviewer: A Local-First Multi-Agent Architecture for Repository-Level Code Review
Local-first multi-agent architecture for automated repository code review with LangGraph orchestration and structured analysis.
3/18/2026
ASDA: Automated Skill Distillation and Adaptation for Financial Reasoning
Automated skill distillation and adaptation method for financial reasoning in LLMs without fine-tuning.
3/18/2026
PathGLS: Evaluating Pathology Vision-Language Models without Ground Truth through Multi-Dimensional Consistency
Reference-free evaluation framework for pathology vision-language models to detect hallucinations without ground truth.
3/18/2026
SWE-QA-Pro: A Representative Benchmark and Scalable Training Recipe for Repository-Level Code Understanding
Benchmark for repository-level code understanding with executable environments, enabling agentic code automation tasks.
3/18/2026
When Generative Augmentation Hurts: A Benchmark Study of GAN and Diffusion Models for Bias Correction in AI Classification Systems
Benchmark comparing generative augmentation strategies (GANs, diffusion) for bias correction in imbalanced classification under low-data conditions.
3/18/2026
HIPO: Instruction Hierarchy via Constrained Reinforcement Learning
Constrained RL method for enforcing hierarchical instruction priority in LLMs via system prompt compliance.
3/18/2026
GATS: Gaussian Aware Temporal Scaling Transformer for Invariant 4D Spatio-Temporal Point Cloud Representation
Transformer architecture for 4D point cloud video understanding with temporal scale invariance.
3/18/2026
DyJR: Preserving Diversity in Reinforcement Learning with Verifiable Rewards via Dynamic Jensen-Shannon Replay
RL method preserving diversity in LLM reasoning via dynamic Jensen-Shannon replay to improve sample efficiency and avoid mode collapse.
3/18/2026
Open-Source Reproduction and Explainability Analysis of Corrective Retrieval Augmented Generation
Open-source reproduction of Corrective RAG replacing proprietary components with Wikipedia API and open models for improved reproducibility.
3/18/2026
MemX: A Local-First Long-Term Memory System for AI Assistants
Local-first long-term memory system for AI assistants with vector and keyword retrieval, implemented in Rust for conversational agents.
3/18/2026
360{\deg} Image Perception with MLLMs: A Comprehensive Benchmark and a Training-Free Method
Benchmark and method for evaluating 360° image perception in multimodal LLMs, addressing geometric distortion and spatial reasoning challenges.
3/18/2026
Robust Generative Audio Quality Assessment: Disentangling Quality from Spurious Correlations
Domain adversarial training approach for robust AI-generated audio quality assessment without spurious correlations.
3/18/2026
A Scoping Review of AI-Driven Digital Interventions in Mental Health Care: Mapping Applications Across Screening, Support, Monitoring, Prevention, and Clinical Education
Scoping review of AI-driven digital mental health interventions including GenAI and HCAI across screening, support, and monitoring.
3/18/2026
CoMAI: A Collaborative Multi-Agent Framework for Robust and Equitable Interview Evaluation
CoMAI multi-agent framework with task decomposition for robust and fair interview evaluation using coordinated LLM agents.
3/18/2026
Generative AI for Quantum Circuits and Quantum Code: A Technical Review and Taxonomy
Technical review and taxonomy of 13 generative systems for quantum circuit and quantum code generation including agentic approaches.
3/18/2026
Visual Prompt Discovery via Semantic Exploration
Visual prompt discovery method to diagnose and mitigate LVLM perception failures through semantic exploration.
3/18/2026
Grounding the Score: Explicit Visual Premise Verification for Reliable Vision-Language Process Reward Models
Vision-language process reward models with explicit visual premise verification for reliable step scoring in reasoning.
3/18/2026
Surrogate-Assisted Genetic Programming with Rank-Based Phenotypic Characterisation for Dynamic Multi-Mode Project Scheduling
Genetic programming with surrogate models for dynamic multi-mode project scheduling with simulation-based optimization.
3/18/2026
VisBrowse-Bench: Benchmarking Visual-Native Search for Multimodal Browsing Agents
VisBrowse-Bench benchmark for evaluating visual-native search in multimodal browsing agents using MLLMs.
3/18/2026
Attention-guided Evidence Grounding for Spoken Question Answering
End-to-end framework using Speech LLMs for spoken question answering with attention-guided evidence grounding.
3/18/2026
A Human-Centred Architecture for Large Language Models-Cognitive Assistants in Manufacturing within Quality Management Systems
Human-centered architecture for integrating LLM-based cognitive assistants into manufacturing quality management systems.
3/18/2026
Detecting Sentiment Steering Attacks on RAG-enabled Large Language Models
Security research on sentiment steering attacks targeting RAG-enabled large language models and LLM robustness.
3/18/2026
Explainable machine learning workflows for radio astronomical data processing
Machine learning pipelines for radio astronomy data processing with explainability focus on automating configuration.
3/18/2026
Automated identification of Ichneumonoidea wasps via YOLO-based deep learning: Integrating HiresCam for Explainable AI
YOLO-based deep learning for automated wasp identification with explainable AI integration for taxonomic classification.
3/18/2026
DynamicGate MLP Conditional Computation via Learned Structural Dropout and Input Dependent Gating for Functional Plasticity
DynamicGate MLP conditional computation framework using learned structural dropout and input-dependent gating for efficiency.
3/18/2026
FederatedFactory: Generative One-Shot Learning for Extremely Non-IID Distributed Scenarios
FederatedFactory zero-dependency framework for federated learning in non-IID scenarios using generative one-shot learning.
3/18/2026
Robust Physics-Guided Diffusion for Full-Waveform Inversion
Physics-guided diffusion framework for full-waveform inversion combining score-based generative models with wave-equation simulations.
3/18/2026
Fanar 2.0: Arabic Generative AI Stack
Fanar 2.0 Arabic generative AI platform built on 256 H100 GPUs at QCRI with sovereign infrastructure and data pipelines.
3/18/2026
Who Benchmarks the Benchmarks? A Case Study of LLM Evaluation in Icelandic
Study identifying flaws in LLM benchmarks for Icelandic, highlighting issues with synthetic and machine-translated evaluation data.
3/18/2026
PlotTwist: A Creative Plot Generation Framework with Small Language Models
PlotTwist creative plot generation framework using small language models with specialized training for narrative coherence.
3/18/2026
Trained Persistent Memory for Frozen Encoder--Decoder LLMs: Six Architectural Methods
Method for adding persistent memory to frozen encoder-decoder LLMs via trainable adapters in continuous latent space.