3/18/2026
Large Language Models for Wireless Communications: From Adaptation to Autonomy
Survey of LLM applications in wireless communications covering adaptation, autonomy, and intelligent system design for complex communication networks.
3/18/2026
From Image Generation to Infrastructure Design: a Multi-agent Pipeline for Street Design Generation
Multi-agent pipeline for street design generation combining image generation and infrastructure design for urban planning visualization.
3/18/2026
Hilbert: Recursively Building Formal Proofs with Informal Reasoning
Hilbert system combines informal LLM reasoning with formal theorem proving in Lean 4 for verifiable mathematical proofs.
3/18/2026
ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory
ReasoningBank memory framework enables LLM agents to learn from interaction history and distill generalizable reasoning strategies for continuous tasks.
3/18/2026
Zephyrus: An Agentic Framework for Weather Science
Zephyrus agentic framework combines weather foundation models with LLM reasoning for interactive scientific workflows in meteorology.
3/18/2026
Impatient Users Confuse AI Agents: High-fidelity Simulations of Human Traits for Testing Agents
Study showing AI agents fail under realistic user behavior variations; proposes high-fidelity human trait simulations for robust agent testing.
3/18/2026
PREFINE: Personalized Story Generation via Simulated User Critics and User-Specific Rubric Generation
PREFINE framework enables personalized story generation using simulated user critics and rubric generation without explicit user feedback.
3/18/2026
Efficient LLM Safety Evaluation through Multi-Agent Debate
Multi-agent debate framework using small language models for cost-efficient LLM safety evaluation, with HAJailBench benchmark for jailbreak testing.
3/18/2026
Alignment-Aware Quantization for LLM Safety
Alignment-Aware Quantization: PTQ method for efficient LLM deployment that preserves behavioral alignment and safety properties, not just minimizing reconstruction error.
3/18/2026
SpatialBench: Benchmarking Multimodal Large Language Models for Spatial Cognition
SpatialBench: benchmark measuring multimodal LLM spatial cognition across hierarchical abilities for real-world physical environment interaction.
3/18/2026
Analyzing Planner Design Trade-offs for MAPF under ADG-based Realistic Execution
Analysis of multi-agent path finding algorithm design trade-offs under realistic robot execution constraints for warehouse and manufacturing applications.
3/18/2026
Stepwise Think-Critique: A Unified Framework for Robust and Interpretable LLM Reasoning
Stepwise Think-Critique: unified framework integrating reasoning and verification in LLMs for robust, interpretable problem-solving with intertwined evaluation.
3/18/2026
Token-Level LLM Collaboration via FusionRoute
FusionRoute: token-level collaboration method enabling multiple specialized LLMs to work together, combining domain expertise efficiency with generalization.
3/18/2026
VisTIRA: Closing the Image-Text Modality Gap in Visual Math Reasoning via Structured Tool Integration
VisTIRA: tool integration approach addressing modality gap where VLMs underperform on visual math problems compared to text-based versions.
3/18/2026
LogicSkills: A Structured Benchmark for Formal Reasoning in Large Language Models
LogicSkills: benchmark isolating three fundamental logical reasoning skills in LLMs: formal symbolization, countermodel construction, and logical inference.
3/18/2026
Dynamics Within Latent Chain-of-Thought: An Empirical Study of Causal Structure
Empirical study of latent chain-of-thought in LLMs using structural causal models to analyze intermediate computation steps beyond correlation-based probes.
3/18/2026
Zero-Shot Time Series Foundation Models for Annual Institutional Forecasting Under Data Sparsity
Benchmark comparing zero-shot Time Series Foundation Models against classical methods for annual institutional demand forecasting under data sparsity.
3/18/2026
MemPO: Self-Memory Policy Optimization for Long-Horizon Agents
MemPO: self-memory policy optimization approach enabling long-horizon agents to proactively manage memory content aligned with task objectives.
3/18/2026
Model Medicine: A Clinical Framework for Understanding, Diagnosing, and Treating AI Models
Model Medicine: framework for understanding, diagnosing, and treating AI model disorders using biological organism principles as analogy for model analysis.
3/18/2026
UIS-Digger: Towards Comprehensive Research Agent Systems for Real-world Unindexed Information Seeking
UIS-Digger: LLM-based research agent system for unindexed information seeking, addressing blind spots where vital information isn't captured by search engines.
3/18/2026
Measuring AI Agents' Progress on Multi-Step Cyber Attack Scenarios
Evaluation of frontier AI models' autonomous cyber-attack capabilities on multi-step scenarios, tracking capability trends across 18 months of model releases.
3/18/2026
Automating Skill Acquisition through Large-Scale Mining of Open-Source Agentic Repositories: A Framework for Multi-Agent Procedural Knowledge Extraction
Framework for automated skill acquisition in modular AI agents through mining open-source repositories to extract procedural knowledge and specialized expertise.
3/18/2026
Relationship-Aware Safety Unlearning for Multimodal LLMs
Framework for unlearning relational safety failures in multimodal LLMs where combinations of benign concepts become unsafe when linked by specific relations.
3/18/2026
Gradient Atoms: Unsupervised Discovery, Attribution and Steering of Model Behaviors via Sparse Decomposition of Training Gradients
Gradient Atoms: unsupervised method for discovering and attributing model behaviors via sparse decomposition of training gradients without requiring predefined queries.
3/18/2026
OpenHospital: A Thing-in-itself Arena for Evolving and Benchmarking LLM-based Collective Intelligence
OpenHospital: interactive benchmark arena for evolving and evaluating LLM-based collective intelligence systems using physician and patient agents.
3/18/2026
Why the Valuable Capabilities of LLMs Are Precisely the Unexplainable Ones
Theoretical analysis arguing that LLM's most valuable capabilities are the unexplainable components that cannot be captured by discrete rule systems.
3/18/2026
SAGE: Multi-Agent Self-Evolution for LLM Reasoning
SAGE framework: multi-agent reinforcement learning system for improving LLM reasoning without large human-labeled datasets, using self-play and closed-loop feedback.
3/18/2026
LLAMAFUZZ: Large Language Model Enhanced Greybox Fuzzing
LLAMAFUZZ uses LLMs to enhance greybox fuzzing for structured data, improving mutation strategies beyond random approaches.
3/18/2026
Domain-Oriented Time Series Inference Agents for Reasoning and Automated Analysis
TS-Reasoner agent integrates LLM reasoning with domain-specific code for multi-step time series inference and automated analysis tasks.
3/18/2026
From Vulnerabilities to Remediation: A Systematic Literature Review of LLMs in Code Security
Systematic literature review of LLM security benefits and drawbacks in code generation, vulnerability detection, and remediation tasks.
3/18/2026
Generative AI Training and Copyright Law
Interdisciplinary study on copyright law implications of training generative AI via web scraping, covering fair use and TDM exceptions.
3/18/2026
MASS: MoErging through Adaptive Subspace Selection
MASS method merges multiple fine-tuned models via adaptive subspace selection, improving accuracy over existing merging approaches without retraining.
3/18/2026
Diverse AI Personas Can Mitigate the Homogenization Effect in Human-AI Collaborative Ideation
Research shows diverse AI personas in generative AI reduce homogenization in collaborative creative outputs compared to single-persona systems.
3/18/2026
Boosting Text-to-Chart Retrieval through Training with Synthesized Semantic Insights
CRBench: Real-world benchmark for text-to-chart retrieval using synthesized semantic insights.
3/18/2026
FALCON: False-Negative Aware Learning of Contrastive Negatives in Vision-Language Alignment
FALCON: Method addressing false negatives in vision-language pretraining through contrastive learning.
3/18/2026
Explanation User Interfaces: A Systematic Literature Review
Systematic literature review of explanation user interfaces for interpretable AI systems.
3/18/2026
BiomedSQL: Text-to-SQL for Scientific Reasoning on Biomedical Knowledge Bases
BiomedSQL: Text-to-SQL benchmark for scientific reasoning over biomedical databases requiring domain knowledge.
3/18/2026
VERINA: Benchmarking Verifiable Code Generation
VERINA benchmark for evaluating LLM code generation with joint code, specification, and proof generation.
3/18/2026
Coded Robust Aggregation for Distributed Learning under Byzantine Attacks
Robust aggregation method for distributed learning systems defending against Byzantine attacks.
3/18/2026
Learning to Diagnose Privately: DP-Powered LLMs for Radiology Report Classification
Differential privacy techniques for LLMs applied to radiology report classification tasks.
3/18/2026
Are LLMs Good Text Diacritizers? An Arabic and Yoruba Case Study
Benchmark evaluation of LLM effectiveness for text diacritization in Arabic and Yoruba with MultiDiac dataset.
3/18/2026
Improved Iterative Refinement for Chart-to-Code Generation via Structured Instruction
Structured instruction approach to improve chart-to-code generation in multimodal LLMs with iterative refinement.
3/18/2026
VideoITG: Multimodal Video Understanding with Instructed Temporal Grounding
VideoITG: Frame selection method for efficient video understanding in video-LLMs using instructed temporal grounding.
3/18/2026
Can large language models assist choice modelling? Insights into prompting strategies and current models capabilities
Systematic study of LLM capabilities for discrete choice modeling with analysis of prompting strategies.
3/18/2026
Can LLMs Detect Their Confabulations? Estimating Reliability in Uncertainty-Aware Language Models
Method for detecting LLM confabulations using token-level uncertainty estimation for reliability in agentic applications.
3/18/2026
On-Policy RL Meets Off-Policy Experts: Harmonizing Supervised Fine-Tuning and Reinforcement Learning via Dynamic Weighting
Dynamic weighting approach integrating supervised fine-tuning and reinforcement learning for LLM post-training alignment.
3/18/2026
No More Blind Spots: Learning Vision-Based Omnidirectional Bipedal Locomotion for Challenging Terrain
Vision-based learning framework for omnidirectional bipedal locomotion using depth images on challenging terrain.
3/18/2026
Generalizable End-to-End Tool-Use RL with Synthetic CodeGym
CodeGym: Reinforcement learning framework for training LLM agents to use tools generalizing across new tasks and workflows.
3/18/2026
LANCE: Low Rank Activation Compression for Efficient On-Device Continual Learning
LANCE: Low-rank compression technique for reducing activation memory in on-device continual learning.
3/18/2026
ERGO: Efficient High-Resolution Visual Understanding for Vision-Language Models
ERGO: Two-stage coarse-to-fine pipeline for efficient high-resolution image processing in vision-language models.