3/26/2026
Research Shows Verbatim Recall of Copyrighted Books in LLMs
Research study demonstrating verbatim recall of copyrighted books in finetuned LLMs across cross-author and within-author scenarios
3/26/2026
PLDR-LLMs Reason At Self-Organized Criticality
Theoretical analysis of LLM reasoning properties at self-organized criticality with connections to phase transitions and scaling functions.
3/26/2026
Environment Maps: Structured Environmental Representations for Long-Horizon Agents
Environment Maps: Persistent agent-agnostic representation for reducing cascading errors in long-horizon LLM-based software automation tasks.
3/26/2026
Evaluating a Multi-Agent Voice-Enabled Smart Speaker for Care Homes: A Safety-Focused Framework
Safety-focused evaluation framework for multi-agent voice-enabled smart speaker in care homes covering resident data access and task scheduling.
3/26/2026
Can LLM Agents Be CFOs? A Benchmark for Resource Allocation in Dynamic Enterprise Environments
EnterpriseArena: Benchmark evaluating LLM agents as CFOs for resource allocation under uncertainty in dynamic business environments.
3/26/2026
GTO Wizard Benchmark
Public API and evaluation framework for benchmarking poker algorithms against GTO Wizard, a superhuman HUNL poker agent.
3/26/2026
Grounding Vision and Language to 3D Masks for Long-Horizon Box Rearrangement
Method for long-horizon 3D box rearrangement using vision-language grounding and 3D masks for multi-step planning from natural language.
3/26/2026
LLMs Do Not Grade Essays Like Humans
Evaluation comparing LLM essay scoring with human grading across GPT and Llama models, finding weak agreement in standard settings.
3/26/2026
Efficient Benchmarking of AI Agents
Study on efficient benchmarking of AI agents showing how task subsets can preserve agent rankings while reducing evaluation costs.
3/26/2026
Learning-guided Prioritized Planning for Lifelong Multi-Agent Path Finding in Warehouse Automation
Learning-guided prioritized planning combining ML and search-based solvers for lifelong multi-agent pathfinding in warehouse automation.
3/26/2026
VehicleMemBench: An Executable Benchmark for Multi-User Long-Term Memory in In-Vehicle Agents
VehicleMemBench: Benchmark for evaluating long-term memory in multi-user in-vehicle agents handling preference conflicts and temporal dynamics.
3/26/2026
SCoOP: Semantic Consistent Opinion Pooling for Uncertainty Quantification in Multiple Vision-Language Model Systems
SCoOP: Training-free uncertainty quantification framework for multi-VLM systems using semantic-consistent opinion pooling.
3/26/2026
The DeepXube Software Package for Solving Pathfinding Problems with Learned Heuristic Functions and Search
DeepXube: Free open-source Python package for pathfinding using learned heuristic functions from deep RL and search algorithms.
3/26/2026
DUPLEX: Agentic Dual-System Planning via LLM-Driven Information Extraction
DUPLEX: Neuro-symbolic agentic architecture combining LLMs with schema-guided information extraction for robust robotic task planning in long-horizon domains.
3/26/2026
AnalogAgent: Self-Improving Analog Circuit Design Automation with LLM Agents
AnalogAgent: LLM-based agentic framework for automated analog circuit design using multi-model loops to preserve domain-specific insights and context.
3/26/2026
From Pixels to Digital Agents: An Empirical Study on the Taxonomy and Technological Trends of Reinforcement Learning Environments
Empirical study analyzing 2000+ RL papers to create quantitative taxonomy of reinforcement learning environments and technological trends.
3/26/2026
Language-Grounded Multi-Agent Planning for Personalized and Fair Participatory Urban Sensing
MAPUS: LLM-based multi-agent framework for personalized and fair participatory urban sensing modeling participants as autonomous agents with preferences.
3/26/2026
ELITE: Experiential Learning and Intent-Aware Transfer for Self-improving Embodied Agents
ELITE framework for self-improving embodied agents using vision-language models with experiential learning and intent-aware transfer to bridge vision-action gap.
3/26/2026
Enhanced Mycelium of Thought (EMoT): A Bio-Inspired Hierarchical Reasoning Architecture with Strategic Dormancy and Mnemonic Encoding
Enhanced Mycelium of Thought (EMoT): bio-inspired hierarchical reasoning architecture for LLMs with four-level hierarchy, strategic dormancy, and mnemonic encoding.
3/26/2026
AI-Supervisor: Autonomous AI Research Supervision via a Persistent Research World Model
AutoProf: multi-agent orchestration framework for autonomous AI research with persistent world model, gap analysis, and inter-agent verification mechanisms.
3/26/2026
Multi-Agent Reasoning with Consistency Verification Improves Uncertainty Calibration in Medical MCQA
Multi-agent framework with specialist agents for medical multiple-choice question answering, improving calibration and confidence scoring through verification.
3/26/2026
The Stochastic Gap: A Markovian Framework for Pre-Deployment Reliability and Oversight-Cost Auditing in Agentic Artificial Intelligence
Markovian framework for auditing reliability and oversight costs in agentic AI systems operating as stochastic policies with sequential decisions and tool calls.
3/26/2026
Mitigating Many-Shot Jailbreaking
Analysis of many-shot jailbreaking technique exploiting long context windows; probes effectiveness and develops mitigation strategies for LLM safety.
3/26/2026
Evidence for Limited Metacognition in LLMs
Novel methodology quantitatively evaluating metacognitive abilities in LLMs, testing self-awareness without relying on model self-reports.
3/26/2026
Leveraging Computerized Adaptive Testing for Cost-effective Evaluation of Large Language Models in Medical Benchmarking
Computerized Adaptive Testing framework grounded in Item Response Theory for cost-effective and scalable evaluation of LLMs in medical benchmarking.
3/26/2026
Beyond Masks: Efficient, Flexible Diffusion Language Models via Deletion-Insertion Processes
Deletion-Insertion Diffusion language models replacing masking paradigm with discrete diffusion processes for improved computational efficiency and generation flexibility.
3/26/2026
Internal Safety Collapse in Frontier Large Language Models
Internal Safety Collapse (ISC) failure mode identified in frontier LLMs where models generate harmful content under certain task conditions; TVD framework presented to trigger and study ISC.
3/26/2026
Visuospatial Perspective Taking in Multimodal Language Models
Evaluation of visuospatial perspective-taking abilities in multimodal language models using adapted tasks from human studies (Director Task, Rotating F task).
3/26/2026
DISCO: Document Intelligence Suite for COmparative Evaluation
DISCO benchmark suite for evaluating OCR pipelines and vision-language models on document parsing and QA across diverse document types including handwritten and multilingual text.
3/26/2026
S-Path-RAG: Semantic-Aware Shortest-Path Retrieval Augmented Generation for Multi-Hop Knowledge Graph Question Answering
S-Path-RAG framework for multi-hop question answering over knowledge graphs using semantic-aware shortest-path retrieval with differentiable path scoring.
3/26/2026
Berta: an open-source, modular tool for AI-enabled clinical documentation
Berta: open-source modular platform for AI-enabled clinical documentation with institutional data governance and workflow integration, deployed at Alberta Health Services.
3/26/2026
DepthCharge: A Domain-Agnostic Framework for Measuring Depth-Dependent Knowledge in Large Language Models
DepthCharge framework for measuring how deeply LLMs sustain accurate responses in domain-specific topics through adaptive probing across arbitrary domains.
3/26/2026
Training a Large Language Model for Medical Coding Using Privacy-Preserving Synthetic Clinical Data
Privacy-preserving synthetic clinical data trains LLM for medical coding automation, improving ICD-10-CM and CPT code assignment from clinical documentation.
3/26/2026
MSA: Memory Sparse Attention for Efficient End-to-End Memory Model Scaling to 100M Tokens
Memory Sparse Attention enables end-to-end LLM scaling to 100M tokens for long-term memory tasks, extending effective context beyond 1M token limits.
3/26/2026
Beyond Accuracy: Introducing a Symbolic-Mechanistic Approach to Interpretable Evaluation
Position paper proposes mechanism-aware evaluation combining symbolic rules and mechanistic interpretability to distinguish genuine generalization from shortcuts.
3/26/2026
Cluster-R1: Large Reasoning Models Are Instruction-following Clustering Agents
Cluster-R1 reframes instruction-following clustering as generative task, enabling reasoning models to autonomously infer corpus structure while respecting user instructions.
3/26/2026
MedMT-Bench: Can LLMs Memorize and Understand Long Multi-Turn Conversations in Medical Scenarios?
MedMT-Bench stress-tests LLMs on long-context memory, interference robustness, and safety in multi-turn medical conversations with realistic clinical scenarios.
3/26/2026
From Physician Expertise to Clinical Agents: Preserving, Standardizing, and Scaling Physicians' Medical Expertise with Lightweight LLM
Lightweight LLM framework captures and scales physician expertise for clinical decision-making agents using individualized diagnostic methodologies.
3/26/2026
Chitrakshara: A Large Multilingual Multimodal Dataset for Indian languages
Chitrakshara multimodal dataset provides multi-image and Indian language coverage for training Vision-Language Models beyond English-centric datasets.
3/26/2026
Qworld: Question-Specific Evaluation Criteria for LLMs
Qworld framework generates question-specific evaluation criteria for LLMs on open-ended tasks, capturing context-dependent response quality requirements.
3/26/2026
Navigating the Concept Space of Language Models
ConceptMap tool enables scalable exploratory discovery of human-interpretable concepts in sparse autoencoders trained on LLM activations.
3/26/2026
Konkani LLM: Multi-Script Instruction Tuning and Evaluation for a Low-Resource Indian Language
Konkani-Instruct-100k synthetic dataset and benchmarks address LLM performance gaps for low-resource Indian language across multiple scripts via instruction tuning.
3/26/2026
Did You Forget What I Asked? Prospective Memory Failures in Large Language Models
Cognitive psychology-inspired study reveals LLMs drop formatting instruction compliance by 2-21% under concurrent task load, identifying prospective memory vulnerabilities.
3/26/2026
Generating Hierarchical JSON Representations of Scientific Sentences Using LLMs
Fine-tuned lightweight LLM generates hierarchical JSON representations of scientific sentences preserving semantic meaning for structured knowledge extraction.
3/26/2026
MDKeyChunker: Single-Call LLM Enrichment with Rolling Keys and Key-Based Restructuring for High-Accuracy RAG
MDKeyChunker pipeline enables structure-aware chunking of Markdown documents and single-call LLM enrichment with metadata extraction for improved RAG accuracy.
3/26/2026
Mixture of Demonstrations for Textual Graph Understanding and Question Answering
Mixture of Demonstrations approach improves GraphRAG performance for domain-specific QA by selecting high-quality demonstrations to reduce irrelevant retrieved information.
3/26/2026
CAPTCHA Solving for Native GUI Agents: Automated Reasoning-Action Data Generation and Self-Corrective Training
Native GUI agent framework ReCAP adds CAPTCHA-solving capability to vision-language models using self-corrective training and automated reasoning-action data generation.
3/26/2026
Synthetic Mixed Training: Scaling Parametric Knowledge Acquisition Beyond RAG
Synthetic Mixed Training combines synthetic QAs and documents to improve LLM knowledge acquisition beyond RAG performance in data-constrained domains.
3/26/2026
Safe Reinforcement Learning with Preference-based Constraint Inference
Safe reinforcement learning approach using preference-based constraint inference for learning complex, subjective safety constraints with minimal expert demonstrations.
3/26/2026
AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization
AI agent optimizes operator performance on Huawei Ascend NPUs by addressing knowledge bottleneck through episodic learning for tiling and kernel programs.