Papers

New on arXiv

Sun 19 Jul 2026

151 papers

Multi-Agent Scheduling with LLM-Assisted Contract Net Negotiation for Stream Processing in Mobile Edge ComputingLLM agents and tool use → Software engineering coding agents
Poor Man's Agentic Modeling: Simulating Large LLM-Agent Societies on a LaptopLLM agents and tool use → Social simulation and multi-agent debate
A Consensus-Based Framework for Relative Preference Evaluation of Large Language ModelsLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Staged Depth-Pruning Distillation of a Flow-Matching Text-to-Speech Teacher: A Compact Hindi Speech SynthesizerDiffusion models → Audio, speech and music +1
Empirical Grounding Improves the Realism of LLM Agents Simulating Human Behavior During DisruptionsLLM agents and tool use → Social simulation and multi-agent debate
Decoder-Preserving Sparse Autoencoders: Which Readouts Survive Sparse Compression?Mechanistic interpretability → Sparse autoencoder training and evaluation
TimeLens2: Generalist Video Temporal Grounding with Multimodal LLMsMultimodal LLMs / vision-language models → Video understanding with MLLMs
Grounded verification of chemical and materials reasoning: detection is the bottleneckLLM reasoning → Clinical and biomedical applications
Feature-Guided Diffusion for Non-Differentiable Inverse RenderingDiffusion models → 3D, point cloud and scene generation
SkyVLaM: Multimodal Large Language Model for UAV Video Understanding in Remote SensingMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Quantifying Diversity of Thought: A Predictive Law of Weighted LLM Ensemble LiftLLM reasoning → Reasoning benchmarks and evaluation
Self-Modifying Lean Proof Agents with Verifier-Grounded Benchmark CoevolutionLLM reasoning → Mathematical reasoning
DeeperRadar: End-to-End MIMO Radar Design and Multi-Modal Fusion for Autonomous Vehicle PerceptionEnd-to-end autonomous driving → End-to-end system architecture
Agentic ERP: Multi-Agent Large Language Model Architecture for Autonomous Enterprise Resource PlanningLLM agents and tool use → General LLM agent frameworks and benchmarks
Rethinking the Suitability of Reinforcement Learning Algorithms Under Practical Transfer ConstraintsDeep reinforcement learning methods → Safe and constrained RL
Rationalizing Boltzmann Rationality: An Axiomatic Characterization of Entropy-Regularized PoliciesDeep reinforcement learning methods → RL theory and sample complexity
WAR: Workload-Aware Rollouts for Synchronous Agentic Reinforcement LearningLLM agents and tool use → Reinforcement learning for LLM agents +1
DRNOISE: Benchmarking Deep Research Agents in Misleading Evidence EnvironmentsLLM agents and tool use → Web, search, and research agents
SAGA: Synthetic Agentic Graph Architecture for Temporal Benchmark GenerationLLM agents and tool use → Web, search, and research agents
Lossless but Not Free: An Empirical Anatomy of Speculative Decoding on Consumer HardwareEfficient LLM inference and serving → Speculative decoding
AIGB-R1: Self-Evolving Generative Auto-Bidding via Hierarchical Planner-Executor OptimizationSynthetic data and self-improvement for LLMs → Self-evolving LLM agents
Between Safe Boundaries: Exploiting Temporal Consistency for Jailbreaking Text-To-Video Generation ModelsLLM safety and security → Vision-language and text-to-image safety
An Explicit World Model Based on Data-First Ontology: DaoQL Multimodal Storage Validation and Counterfactual Reasoning EvaluationWorld models → World models for language agents +1
Debate-on-Graph: Reliable and Adaptive Reasoning of Large Language Model on Uncertain Knowledge GraphLLM reasoning → Knowledge- and tool-augmented reasoning
Asynchronous Multimodal Diffusion Policy Composition via Latency-Aware Guidance FusionDiffusion models → Diffusion and flow policies for robotics +1
EvolvingWorld: An Open-Schema Framework for Co-Evolving Role-Play Agents and World Model in Interactive Literary WorldWorld models → World models for language agents +1
Distilled Reinforcement Learning for LLM Post-trainingKnowledge distillation → On-policy and self-distillation for LLMs
LenGuard-GPC: Length Guarding with Guided-Prompt Consistency for Spatial Reasoning Reinforce LearningLLM reasoning → Spatial and embodied reasoning +1
Constrained Path Reasoning: Measuring When Committed Stages Earn Their CostLLM reasoning → Test-time compute and efficient thinking
Specifying the Delegated-Autonomy Boundary: Requirements Engineering for Agentic AILLM agents and tool use → Agent threats, guardrails and governance
A Systematic Evaluation of Trajectory Data Curation for LoRA Fine-Tuning of Code AgentsLLM agents and tool use → Software engineering coding agents
Non-Asymptotic Best Policy Identification Guarantees in Online Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity
UPAIR: Diagnosing Reasoning States via Uncertainty-Progress Alignment for Selective InterventionLLM reasoning → Test-time compute and efficient thinking
OrderMoE: An expert similarity driven distributed edge MoE inferenceMixture-of-experts → Efficient MoE training and inference systems
How Jailbreak Attacks Inform Safety Alignment: A Defender-Centric, Shapley-Based Evaluation of Jailbreak ContributionsLLM safety and security → Jailbreaking
SlotGuard: Stop Oversharing Private Local Context in LLM Agent TranscriLLM agents and tool use → Agent threats, guardrails and governance
SynH-Rank: Quality-Aware Code Search via Diverse Data Synthesis and Hierarchical Ranking TrainingSynthetic data and self-improvement for LLMs → Code and instruction data synthesis
Teach it to stop, not just to clickLLM agents and tool use → Software engineering coding agents
Persistent Sparse Autoencoders: Learning Feature-Specific Timescales in Language Model RepresentationsMechanistic interpretability → SAE features and their applications
Evidence Interfaces Shape How Retrieval-Augmented Readers Use SupportRetrieval-augmented generation → RAG robustness and adaptive retrieval
Auto Research for Materials: Auditable AI-Scientist Workflows with Held-Out TransferLLM agents and tool use → Web, search, and research agents
OTAP: Structure-Aware Optimal Transport for Evaluating Planning and Execution in Agent TrajectoriesLLM agents and tool use → General LLM agent frameworks and benchmarks
ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of ExpertsMixture-of-experts → Efficient MoE training and inference systems
Searching for Task-Specific Vision Paths: Evolutionary Block Pruning Across Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
EvoGUI: An Evolution-Aware Benchmark for GUI State-Transition UnderstandingLLM agents and tool use → GUI and mobile agents
Solver-Hard Is Not Model-Hard: A Hardness-Controlled Diagnostic for LLM Constraint ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Learning from Synthetic Data without Model Collapse in Iterative Instruction TuningSynthetic data and self-improvement for LLMs → Model collapse from synthetic data
Reward-Driven LLM Agent Workflows: Synthesizing POMDP Routing and Self-Correction for Autonomous Decision-MakingLLM agents and tool use → Reinforcement learning for LLM agents +1
Feasibility-Aware Security-Constrained Unit Commitment via Hybrid Soft Actor-Critic with Quantum-Sampled FeaturesDeep reinforcement learning methods → Quantum reinforcement learning
Regularize or Localize: When Training-Time KV-Cache Geometry Pays Under QuantizationEfficient LLM inference and serving → KV cache quantization
Schema-Bound LLM Control of Scientific Instrumentation through Model Context Protocol SkillsLLM agents and tool use → Tool learning and function calling
English Word Sense Disambiguation in 2026: When the Labels Become the Bottleneck
EditCLEVR: A Paired-Scene Intervention Benchmark for Compositional Faithfulness of Object-Centric Representations
Visible-Light Imaging Diagnosis of Neutral Particle Emission Tomography in the Tokamak Divertor: An Efficient Transformer-based Surrogate Model
Histopathological Spectrum-Guided Prostate Stratification via Segmentation-Assisted Diagnostic Transformer
On the Depth Scalability of Logic Gate Networks
EmoEUS: Uncertainty Supervision for Multimodal Emotion Recognition in Conversation
ChemHyperMag: Physics-informed magnetic hypergraph learning improves molecular ADMET prediction
Physics-Guided Masked Multi-Task Network for Edge-Friendly Battery Health Diagnostics from Sto-chastically Fragmented Charging Profiles
Dynamic Loss Balancing for Joint SOH and RUL Prediction of Lithium-Ion Batteries via a Rotary SOH-Injected Prior Battery Transformer
An Explainable FFT-Based Spatial-Frequency Fusion Framework for Deepfake Detection
Intermittent Control Is Not Diluted Control: A Switching Effect in Artificial Agency
Abliteration Is Not a Scalpel: Off-Target Effects of Refusal Removal on Decision Disposition Across Model Families
The Librarian Who Refused to Code: Model-Dependent Identity Enactment in LLM Code Generation
Kernelized Linear Attention: Breaking the Capacity Wall with Symmetric Cones
CORAL: Learning Amyloid Fibril Ligand Docking with Cooperative Binding Rewards
Efficient Sequential Evaluation of Large Language Models
Does Super-Resolution Preserve Defect Evidence? A Low-False-Call Benchmark for Semiconductor Inspection
The PanAf-SBR Dataset: Social Behaviour Recognition for Wild Great Apes
CoEvoP&R: Co-Evolving Placement Objectives with Routing Feedback via Large Language Models
Automating Visual Recognition of Leprosy in Wild Chimpanzees
A Phased Development Framework Enabling Islanded Operation of Sustainable AI Data Centers With Onsite Grid-Following and Grid-Forming Energy Architectures
Kernel Regression with Tensor Trains and Hadamard Overparameterization
Mathematical Discovery in the Wild: AI-Guided Proofs in Banach Space Theory
Team DACTYL at PAN 2026: Bayesian Data Mixing and Empirical X-risk Minimization for AI-text Detection
Econometrics with Pre-Trained Embeddings for Unstructured Data
Chebyshev Manifold Adaptation
Taurus: Accelerating Out-of-Core Graph Neural Network Inference on Billion-Scale Graphs
Stringological sequence prediction II: Right-to-left automaticity and related complexity measures
EII-SCL: Harnessing Emotional Inertia for Multimodal Emotion Recognition in Conversation
A multiverse-consensus pipeline for reproducible feature selection in untargeted LC-MS metabolomics
STAR: Skeletal Token Alignment and Rearrangement for Interaction Recognition
Understanding From Human Perspective: A Multi-agent System for Interactive Egocentric Medical Image Segmentation
Orthogonal Knowledge Refreshing for Domain-Incremental Object Detection
When Drift Detectors cry Wolf: False Alarm Rates in continuous ML Monitoring
MIS-HCC: Hierarchical Channel Clustering for Efficient Medical Image Segmentation
Interpreting Quantum Learning Models via Stochastic Processes
From Perception to Assistance: Open-Vocabulary Shared Autonomy for Robotic Manipulation
TAPAS: Throughput-adaptive Perception for Autonomous Systems
The Optimization Trilemma: Efficiency, Comfort and Fairness in Decentralized Multi-agent Coordination
Learning-Driven Adaptive Audit Scheduling: A Sequential Decision Approach to Off-Chain Data Integrity
Lookahead Branching for Neural Network Verification
An Iterative Geometric Approach to Optimizing Separating Hyperplanes
The Resolution of Causal Heterogeneity
Node4All: Learning Node Representation Beyond Datasets
Safety That Does Not Transfer: Cross-Lingual Clinical Correctness Drift in Deployable Medical Language Models
PACE: Polar Axis-Conditioned Estimation for PairUAV Relative Localization
Coordinated Disentanglement with Iterative Mode Discovery Under Hidden Correlations
Should Missing Modalities Always Be Necessary to Repair for Multi-modal Sentiment Analysis?
VecFontLLM: Anchor-Guided Direct Synthesis of Chinese Vector Fonts

All 151