![[2605.30712] ExpGraph: Model-Agnostic Experience Learning with Graph-Structured Memory for LLM Agents](/static/img/curated/desks/afbytes-tech.png)
[2605.30712] ExpGraph: Model-Agnostic Experience Learning with Graph-Structured Memory for LLM Agents
Abstract page for arXiv paper 2605.30712: ExpGraph: Model-Agnostic Experience Learning with Graph-Structured Memory for LLM Agents
[2605.30705] Equivariant Latent Alignment via Flow Matching under Group Symmetries
Abstract page for arXiv paper 2605.30705: Equivariant Latent Alignment via Flow Matching under Group Symmetries
Latest in Ai
Showing 1801-1850 of 5249
[2605.30750] SLAP: The Semantic Least Action Principle for Variational Video-Language Modeling
Abstract page for arXiv paper 2605.30750: SLAP: The Semantic Least Action Principle for Variational Video-Language Modeling
[2605.30753] Efficient Diffusion LLMs via Temporal-Spatial Parallel Decoding and Confidence Extrapolation
Abstract page for arXiv paper 2605.30753: Efficient Diffusion LLMs via Temporal-Spatial Parallel Decoding and Confidence Extrapolation
[2605.30757] Chain-of-Thought and Compressed Looped Transformers: A Memory-Budget Separation
Abstract page for arXiv paper 2605.30757: Chain-of-Thought and Compressed Looped Transformers: A Memory-Budget Separation
[2605.30758] Pairwise Reference Alignment as a Model-Level Ordinal Observable
Abstract page for arXiv paper 2605.30758: Pairwise Reference Alignment as a Model-Level Ordinal Observable
[2605.30763] A Data-Driven Methodology for Scalable Distributed MPC in Heterogeneous Building Aggregation: From Systematic Feature Selection to Convex Optimization
Abstract page for arXiv paper 2605.30763: A Data-Driven Methodology for Scalable Distributed MPC in Heterogeneous Building Aggregation: From Systematic Feature ...
[2605.30769] DisPlace: Discriminative Place Projections for Multi-Reference Visual Place Recognition
Abstract page for arXiv paper 2605.30769: DisPlace: Discriminative Place Projections for Multi-Reference Visual Place Recognition
[2605.30770] SSR: Scaling Surefooted and Symmetric Humanoid Traversal to the Open World
Abstract page for arXiv paper 2605.30770: SSR: Scaling Surefooted and Symmetric Humanoid Traversal to the Open World
[2605.30771] Eywa: Provenance-Grounded Long-Term Memory for AI Agents
Abstract page for arXiv paper 2605.30771: Eywa: Provenance-Grounded Long-Term Memory for AI Agents
[2605.30774] CameraNoise: Enabling Faithful Camera Control in Video Diffusion through Geometry-Flow-Guided Noise Warping
Abstract page for arXiv paper 2605.30774: CameraNoise: Enabling Faithful Camera Control in Video Diffusion through Geometry-Flow-Guided Noise Warping
[2605.30776] Efficient and Uncertainty-Aware Diffusion Framework for Offline-to-Online Reinforcement Learning
Abstract page for arXiv paper 2605.30776: Efficient and Uncertainty-Aware Diffusion Framework for Offline-to-Online Reinforcement Learning
[2605.30877] Wall-OSS-0.5 Technical Report
Abstract page for arXiv paper 2605.30877: Wall-OSS-0.5 Technical Report
[2605.30883] TRACE: Task-Aware Adaptive Self-Evolving Agentic Jailbreaking
Abstract page for arXiv paper 2605.30883: TRACE: Task-Aware Adaptive Self-Evolving Agentic Jailbreaking
[2605.30884] GUI-C$^2$: Coarse-to-Fine GUI Grounding via Difficulty-Aware Reinforcement Learning
Abstract page for arXiv paper 2605.30884: GUI-C$^2$: Coarse-to-Fine GUI Grounding via Difficulty-Aware Reinforcement Learning
[2605.30888] The Flip Side of RLHF: On-Policy Feedback for Reward Model Self-Supervised Improvement
Abstract page for arXiv paper 2605.30888: The Flip Side of RLHF: On-Policy Feedback for Reward Model Self-Supervised Improvement
[2605.30892] Bandwidth Allocation with Device Partitioning for Federated Learning over Industrial IoT networks
Abstract page for arXiv paper 2605.30892: Bandwidth Allocation with Device Partitioning for Federated Learning over Industrial IoT networks
[2605.30893] Foundation VAEs for 3D CT Reconstruction, Augmentation, and Generation
Abstract page for arXiv paper 2605.30893: Foundation VAEs for 3D CT Reconstruction, Augmentation, and Generation
[2605.30894] SteerFace: Debiasing Synthetic Face Generation via Adaptive Residue Perturbation
Abstract page for arXiv paper 2605.30894: SteerFace: Debiasing Synthetic Face Generation via Adaptive Residue Perturbation
[2605.31193] Geometry-based Schrödinger Bridges for Trustworthy Multimodal Fusion
Abstract page for arXiv paper 2605.31193: Geometry-based Schrödinger Bridges for Trustworthy Multimodal Fusion
[2605.31217] TALON: Token-Aligned Lightweight Adapters for 6-DoF Spacecraft Pose Estimation
Abstract page for arXiv paper 2605.31217: TALON: Token-Aligned Lightweight Adapters for 6-DoF Spacecraft Pose Estimation
[2605.31219] Latent Geometric Chords for Query-Efficient Decision-Based Adversarial Attacks
Abstract page for arXiv paper 2605.31219: Latent Geometric Chords for Query-Efficient Decision-Based Adversarial Attacks
[2605.31222] Multivariate Distributional Reinforcement Learning Using Sliced Divergences
Abstract page for arXiv paper 2605.31222: Multivariate Distributional Reinforcement Learning Using Sliced Divergences
[2605.31227] HiERO-StepG @ Ego4D Step Grounding Challenge: hierarchical activity understanding enables zero-shot step grounding
Abstract page for arXiv paper 2605.31227: HiERO-StepG @ Ego4D Step Grounding Challenge: hierarchical activity understanding enables zero-shot step grounding
[2605.31234] HARP-VLA: Human-Robot Aligned Representation Learning for Vision-Language-Action Model
Abstract page for arXiv paper 2605.31234: HARP-VLA: Human-Robot Aligned Representation Learning for Vision-Language-Action Model
[2605.31238] Scaling Multi-Hop Training Data via Graph-Constrained Path Selection
Abstract page for arXiv paper 2605.31238: Scaling Multi-Hop Training Data via Graph-Constrained Path Selection
[2605.31241] Bifurcated Remaining Useful Life Prediction: A Hybrid Approach for Realistic Uncertainty Characterization
Abstract page for arXiv paper 2605.31241: Bifurcated Remaining Useful Life Prediction: A Hybrid Approach for Realistic Uncertainty Characterization
[2605.31243] Safe Arrival Scheduling at Constraint Waypoints in UAM Corridors
Abstract page for arXiv paper 2605.31243: Safe Arrival Scheduling at Constraint Waypoints in UAM Corridors
[2605.31245] Toward Identifiable Sparse Autoencoders
Abstract page for arXiv paper 2605.31245: Toward Identifiable Sparse Autoencoders
[2605.31246] BadBone: Backdoor Attacks Against Backbone Models in Visual Prompt Learning
Abstract page for arXiv paper 2605.31246: BadBone: Backdoor Attacks Against Backbone Models in Visual Prompt Learning
[2605.31351] A Visually Impaired Assistance Benchmark for VLM-as-a-Judge Evaluation
Abstract page for arXiv paper 2605.31351: A Visually Impaired Assistance Benchmark for VLM-as-a-Judge Evaluation
[2605.31363] The Latin Substrate: How Language Models Represent and Mediate Script Choice
Abstract page for arXiv paper 2605.31363: The Latin Substrate: How Language Models Represent and Mediate Script Choice
[2605.31367] Trading Complexity for Expressivity Through Structured Generalized Linear Token Mixing
Abstract page for arXiv paper 2605.31367: Trading Complexity for Expressivity Through Structured Generalized Linear Token Mixing
[2605.31369] A Unifying View of Variational Generative Wasserstein Flows
Abstract page for arXiv paper 2605.31369: A Unifying View of Variational Generative Wasserstein Flows
[2605.31371] Softsign: Smooth Sign in Your Optimizer For Better Parameter Heterogeneity Handling
Abstract page for arXiv paper 2605.31371: Softsign: Smooth Sign in Your Optimizer For Better Parameter Heterogeneity Handling
[2605.31433] SCOPE: Self-Play via Co-Evolving Policies for Open-Ended Tasks
Abstract page for arXiv paper 2605.31433: SCOPE: Self-Play via Co-Evolving Policies for Open-Ended Tasks
[2605.31434] Shaft-integrated Force Sensing with Transformer-based Dynamics Compensation for Telesurgery
Abstract page for arXiv paper 2605.31434: Shaft-integrated Force Sensing with Transformer-based Dynamics Compensation for Telesurgery
[2605.31437] Astra: a generalizable report generation foundation model for 3D computed tomography
Abstract page for arXiv paper 2605.31437: Astra: a generalizable report generation foundation model for 3D computed tomography
[2605.31312] Learning from Fine-Grained Visual Discrepancies: Mitigating Multimodal Hallucinations via In-Context Visual Contrastive Optimization
Abstract page for arXiv paper 2605.31312: Learning from Fine-Grained Visual Discrepancies: Mitigating Multimodal Hallucinations via In-Context Visual Contrastiv...
[2605.31452] Translation Analytics for Freelancers II: Benchmarking Local LLMs for Confidential Translation Workflows
Abstract page for arXiv paper 2605.31452: Translation Analytics for Freelancers II: Benchmarking Local LLMs for Confidential Translation Workflows
[2605.31455] DRIFT: Decoupled Rollouts and Importance-Weighted Fine-Tuning for Efficient Multi-Turn Optimization
Abstract page for arXiv paper 2605.31455: DRIFT: Decoupled Rollouts and Importance-Weighted Fine-Tuning for Efficient Multi-Turn Optimization
[2605.31457] VisionPulse: Dynamic Visual Sparsity for Efficient Multimodal Reasoning
Abstract page for arXiv paper 2605.31457: VisionPulse: Dynamic Visual Sparsity for Efficient Multimodal Reasoning
[2605.31458] Ladder Logic Translation using Large Language Models in Industrial Automation
Abstract page for arXiv paper 2605.31458: Ladder Logic Translation using Large Language Models in Industrial Automation
[2605.31460] On-Device Robotic Planning: Eliminating Inference Redundancy for Efficient Decision-Making
Abstract page for arXiv paper 2605.31460: On-Device Robotic Planning: Eliminating Inference Redundancy for Efficient Decision-Making
[2605.31487] Enhancing Computer Vision Model Generalization in Warehouse Facilities: A Case Study on Anomaly Detection in Vertical Material Handling Systems
Abstract page for arXiv paper 2605.31487: Enhancing Computer Vision Model Generalization in Warehouse Facilities: A Case Study on Anomaly Detection in Vertical ...
[2605.31489] Context-Conditioned Generative Models Enable Subnational Refinement of Sparse Humanitarian Surveys
Abstract page for arXiv paper 2605.31489: Context-Conditioned Generative Models Enable Subnational Refinement of Sparse Humanitarian Surveys
[2605.31490] Are Full Rollouts Necessary for On-Policy Distillation?
Abstract page for arXiv paper 2605.31490: Are Full Rollouts Necessary for On-Policy Distillation?
[2605.31494] Consolidating Rewarded Perturbations for LLM Post-Training
Abstract page for arXiv paper 2605.31494: Consolidating Rewarded Perturbations for LLM Post-Training
[2605.31503] How can embedding models bind concepts?
Abstract page for arXiv paper 2605.31503: How can embedding models bind concepts?
[2605.31508] Internalizing Temporal Consistency in Video Object-Centric Learning without Explicit Regularization
Abstract page for arXiv paper 2605.31508: Internalizing Temporal Consistency in Video Object-Centric Learning without Explicit Regularization