![[2606.01483] MURMUR: An Efficient Inference System for Long-Form ASR](/static/img/curated/desks/afbytes-tech.w400.webp)
[2606.01483] MURMUR: An Efficient Inference System for Long-Form ASR
Abstract page for arXiv paper 2606.01483: MURMUR: An Efficient Inference System for Long-Form ASR
![[2606.01482] Beyond Topical Similarity: Contrastive Evidence Retrieval with Interpretable Attention Alignment in RAG](/static/img/curated/desks/afbytes-tech.w1200.webp)
[2606.01482] Beyond Topical Similarity: Contrastive Evidence Retrieval with Interpretable Attention Alignment in RAG
Abstract page for arXiv paper 2606.01482: Beyond Topical Similarity: Contrastive Evidence Retrieval with Interpretable Attention Alignment in RAG
[2606.01485] Perception First: A Frontier Native-Video Model with Self-Consistency for Implicit Video Question Answering
Abstract page for arXiv paper 2606.01485: Perception First: A Frontier Native-Video Model with Self-Consistency for Implicit Video Question Answering
[2606.01488] A Reproducible UAV-Assisted VANET Dataset Generator for Fragmentation Risk Analysis in Intelligent Transportation Systems
Abstract page for arXiv paper 2606.01488: A Reproducible UAV-Assisted VANET Dataset Generator for Fragmentation Risk Analysis in Intelligent Transportation Syst...
[2606.01490] LLM Consortium for Software Design Refinement: A Controlled Experiment on Multi-Agent Collaboration Topologies
Abstract page for arXiv paper 2606.01490: LLM Consortium for Software Design Refinement: A Controlled Experiment on Multi-Agent Collaboration Topologies
[2606.01493] Splatshot: 3D Face Avatar Generation from a Single Unconstrained Photo
Abstract page for arXiv paper 2606.01493: Splatshot: 3D Face Avatar Generation from a Single Unconstrained Photo
[2606.01495] CART: Context-Anchored Recurrent Transformer -- A Parameter-Efficient Architecture with Learned Stability
Abstract page for arXiv paper 2606.01495: CART: Context-Anchored Recurrent Transformer -- A Parameter-Efficient Architecture with Learned Stability
[2606.01498] TimeSage-MT: A Multi-Turn Benchmark for Evaluating Agentic Time Series Reasoning
Abstract page for arXiv paper 2606.01498: TimeSage-MT: A Multi-Turn Benchmark for Evaluating Agentic Time Series Reasoning
[2606.01502] Move the Query, Not the Cache: Characterizing Cross-Instance Latent Attention Redistribution Across GPU Fabrics
Abstract page for arXiv paper 2606.01502: Move the Query, Not the Cache: Characterizing Cross-Instance Latent Attention Redistribution Across GPU Fabrics
[2606.01503] On the Limits of Token Reduction for Efficient Unified Vision Language Training
Abstract page for arXiv paper 2606.01503: On the Limits of Token Reduction for Efficient Unified Vision Language Training
[2606.01508] Agent Operating Systems (AOS): Integrating Agentic Control Planes into, and Beyond, Traditional Operating Systems
Abstract page for arXiv paper 2606.01508: Agent Operating Systems (AOS): Integrating Agentic Control Planes into, and Beyond, Traditional Operating Systems
[2606.01542] Self-Conditioned Positional HNSW for Overlap-Aware Retrieval in Chunked-Document RAG Systems: Method and Industrial Evidence-Quality Audit
Abstract page for arXiv paper 2606.01542: Self-Conditioned Positional HNSW for Overlap-Aware Retrieval in Chunked-Document RAG Systems: Method and Industrial Ev...
[2606.01543] PathAR: Structure-First Autoregressive Synthesis of Multimodal Pathology Images
Abstract page for arXiv paper 2606.01543: PathAR: Structure-First Autoregressive Synthesis of Multimodal Pathology Images
[2606.01544] CRePE: Convolution-aware Relative Importance in Post-training Pruning with Efficient Search
Abstract page for arXiv paper 2606.01544: CRePE: Convolution-aware Relative Importance in Post-training Pruning with Efficient Search
[2606.01545] Hierarchical Object Representation for Spatial Robot Perception: Points, Meshes, and Superquadrics
Abstract page for arXiv paper 2606.01545: Hierarchical Object Representation for Spatial Robot Perception: Points, Meshes, and Superquadrics
[2606.01546] Flexible Online Representation Learning Based on Similarity Matching
Abstract page for arXiv paper 2606.01546: Flexible Online Representation Learning Based on Similarity Matching
[2606.01549] ForestMamba: Sparse Mamba with Geometry-guided Queries for 3D Forest Point Cloud Segmentation
Abstract page for arXiv paper 2606.01549: ForestMamba: Sparse Mamba with Geometry-guided Queries for 3D Forest Point Cloud Segmentation
[2606.01552] RoleCDE:Benchmarking and Mitigating Role-Alignment Trade-offs in Role-Playing Agents
Abstract page for arXiv paper 2606.01552: RoleCDE:Benchmarking and Mitigating Role-Alignment Trade-offs in Role-Playing Agents
[2606.01556] TwinQuant: Learnable Subspace Decomposition for 4-Bit LLM Quantization
Abstract page for arXiv paper 2606.01556: TwinQuant: Learnable Subspace Decomposition for 4-Bit LLM Quantization
[2606.01557] Everywhere Learning: Artificial Intelligence with Pointwise Constraints
Abstract page for arXiv paper 2606.01557: Everywhere Learning: Artificial Intelligence with Pointwise Constraints
[2606.01558] Attention-guided Fine-tuning of Multimodal Large Language Models Improves Chain-of-Thought Reasoning
Abstract page for arXiv paper 2606.01558: Attention-guided Fine-tuning of Multimodal Large Language Models Improves Chain-of-Thought Reasoning
[2606.01581] Agent System Operations: Categorization, Challenges, and Future Directions
Abstract page for arXiv paper 2606.01581: Agent System Operations: Categorization, Challenges, and Future Directions
[2606.01584] Identifying High-Confidence Social Biases in LLMs for Trustworthy Conversational Tutoring Agents
Abstract page for arXiv paper 2606.01584: Identifying High-Confidence Social Biases in LLMs for Trustworthy Conversational Tutoring Agents
[2606.01590] Effective Multi-sensor Conditioning for Street-view Novel-view Synthesis
Abstract page for arXiv paper 2606.01590: Effective Multi-sensor Conditioning for Street-view Novel-view Synthesis
[2606.01591] TLG: Temporal-Logic Grounding for Video Question Answering via Source-Annotation Reconstruction and Category-Targeted Reasoning
Abstract page for arXiv paper 2606.01591: TLG: Temporal-Logic Grounding for Video Question Answering via Source-Annotation Reconstruction and Category-Targeted ...
[2606.01592] Question Type, Cognitive Load, and CEFR Alignment: Evaluating LLM-Generated EFL Grammar Drill Exercises
Abstract page for arXiv paper 2606.01592: Question Type, Cognitive Load, and CEFR Alignment: Evaluating LLM-Generated EFL Grammar Drill Exercises
[2606.01594] Distributed Algorithm for Robust Wardrop Equilibrium in Uncertain Aggregative Congestion Games
Abstract page for arXiv paper 2606.01594: Distributed Algorithm for Robust Wardrop Equilibrium in Uncertain Aggregative Congestion Games
[2606.01595] Uncertainty-Calibrated Diffusion for Reliable 3D Molecular Graph Generation
Abstract page for arXiv paper 2606.01595: Uncertainty-Calibrated Diffusion for Reliable 3D Molecular Graph Generation
[2606.01596] Learning Chaotic Dynamics through Second-Order Geometric Supervision
Abstract page for arXiv paper 2606.01596: Learning Chaotic Dynamics through Second-Order Geometric Supervision
[2606.01597] Physics-Informed Modeling and Control of Emergent Behaviors in Robot Swarms
Abstract page for arXiv paper 2606.01597: Physics-Informed Modeling and Control of Emergent Behaviors in Robot Swarms
[2606.01599] TRON: Targeted Rule-Verifiable Online Environments for Visual Reasoning RL
Abstract page for arXiv paper 2606.01599: TRON: Targeted Rule-Verifiable Online Environments for Visual Reasoning RL
[2606.01670] Time-Aware Diffusion based on Preference Disentanglement for Generative Recommendation
Abstract page for arXiv paper 2606.01670: Time-Aware Diffusion based on Preference Disentanglement for Generative Recommendation
[2606.01671] When Meaning Travels: A Granular Lens on Hybrid-MoE's Role in Idiomatic Understanding for Language Models
Abstract page for arXiv paper 2606.01671: When Meaning Travels: A Granular Lens on Hybrid-MoE's Role in Idiomatic Understanding for Language Models
[2606.01672] RDA: Reward Design Agent for Reinforcement Learning
Abstract page for arXiv paper 2606.01672: RDA: Reward Design Agent for Reinforcement Learning
[2606.01676] Overcoming Challenges in Agile and DevOps Integration: A Qualitative Study
Abstract page for arXiv paper 2606.01676: Overcoming Challenges in Agile and DevOps Integration: A Qualitative Study
[2606.01677] UniVocal: Unified Speech-Singing Code-Switching Synthesis
Abstract page for arXiv paper 2606.01677: UniVocal: Unified Speech-Singing Code-Switching Synthesis
[2606.01678] Why Do Self-Harm Prediction Models Struggle to Generalise? Lexical and Semantic Variations in Emergency Department Triage Notes
Abstract page for arXiv paper 2606.01678: Why Do Self-Harm Prediction Models Struggle to Generalise? Lexical and Semantic Variations in Emergency Department Tri...
[2606.01679] Encoded but Not Routed: Explaining the Table-Chart Gap in Scientific Claim Verification
Abstract page for arXiv paper 2606.01679: Encoded but Not Routed: Explaining the Table-Chart Gap in Scientific Claim Verification
![[2606.01680] Don't Let a Few Network Failures Slow the Entire AllReduce](/static/img/curated/desks/afbytes-datacenters.w400.webp)
[2606.01680] Don't Let a Few Network Failures Slow the Entire AllReduce
Abstract page for arXiv paper 2606.01680: Don't Let a Few Network Failures Slow the Entire AllReduce
[2606.01682] Off-the-Shelf LLMs as Process Scorers: Training-Free Alternative to PRMs for Mathematical Reasoning
Abstract page for arXiv paper 2606.01682: Off-the-Shelf LLMs as Process Scorers: Training-Free Alternative to PRMs for Mathematical Reasoning
[2606.01686] HAIM: Human-AI Music Datasets for AI Music Production Tracking Benchmark
Abstract page for arXiv paper 2606.01686: HAIM: Human-AI Music Datasets for AI Music Production Tracking Benchmark
[2606.01725] Characterization of Multi-Model Agentic AI Systems on General Tasks via Trace-Driven Simulation
Abstract page for arXiv paper 2606.01725: Characterization of Multi-Model Agentic AI Systems on General Tasks via Trace-Driven Simulation
[2606.01730] Evidence-Gated LLM Priors for Multi-Objective Bayesian Optimization
Abstract page for arXiv paper 2606.01730: Evidence-Gated LLM Priors for Multi-Objective Bayesian Optimization
[2606.01734] FlatVPR: Plug-and-play Geo-linear Residual Adapter for Geometric Rectification of Foundation Model Feature Manifolds
Abstract page for arXiv paper 2606.01734: FlatVPR: Plug-and-play Geo-linear Residual Adapter for Geometric Rectification of Foundation Model Feature Manifolds
[2606.01736] Argument Collapse: LLMs Flatten Long-Form Public Debate
Abstract page for arXiv paper 2606.01736: Argument Collapse: LLMs Flatten Long-Form Public Debate
[2606.01737] TrafficRAG: A Multimodal RAG Framework for Traffic Accident Liability Determination
Abstract page for arXiv paper 2606.01737: TrafficRAG: A Multimodal RAG Framework for Traffic Accident Liability Determination
[2606.01738] THRD: A Training-Free Multi-Turn Defense Framework for Jailbreak Attacks on Large Language Models
Abstract page for arXiv paper 2606.01738: THRD: A Training-Free Multi-Turn Defense Framework for Jailbreak Attacks on Large Language Models
[2606.01741] SECUREVENT: Hybrid AI/ML Security Monitoring for Distributed Event-Based Systems
Abstract page for arXiv paper 2606.01741: SECUREVENT: Hybrid AI/ML Security Monitoring for Distributed Event-Based Systems
[2606.01723] Shortcut to Nowhere: Demystifying Deep Spurious Regression
Abstract page for arXiv paper 2606.01723: Shortcut to Nowhere: Demystifying Deep Spurious Regression
Anthropic IPO filing ratifies Wall Street's AI obsession
Tech stocks continue to lead the Nasdaq Composite and S&P 500 in a record-breaking market rally, while Elon Musk's SpaceX is set to price its IPO this month and...