Issues

• SkyServe: Serving AI Models across Regions and Clouds with Spot Instances

  pentium3/sys_reading#463 · Filter by author pentium3 opened on Sep 28, 2026

• MAPS: Memory-Aware Predictive Scheduling Framework for Large Language Model Serving

  pentium3/sys_reading#462 · Filter by author pentium3 opened on Sep 21, 2026

• HybridFlow: Resource-Adaptive Subtask Routing for Efficient Edge-Cloud LLM Inference

  pentium3/sys_reading#461 · Filter by author pentium3 opened on Sep 21, 2026

• Agent JIT Compilation for Latency-Optimizing Computer-Use Agent Planning and Scheduling

  pentium3/sys_reading#460 · Filter by author pentium3 opened on Sep 21, 2026

• Concur: Proactive Agent-Level Admission Control for Efficient Agentic Batch Inference

  pentium3/sys_reading#459 · Filter by author pentium3 opened on Sep 21, 2026

• EvoRoute: Experience-Driven Self-Routing LLM Agent Systems

  pentium3/sys_reading#458 · Filter by author pentium3 opened on Sep 21, 2026

• Timely Machine: Awareness of Time Makes Test-Time Scaling Agentic

  pentium3/sys_reading#457 · Filter by author pentium3 opened on Sep 21, 2026

• GlimpRouter: Efficient Collaborative Inference by Glimpsing One Token of Thoughts

  pentium3/sys_reading#456 · Filter by author pentium3 opened on Sep 21, 2026

• Beyond Accuracy: Unveiling Inefficiency Patterns in Tool-Integrated Reasoning

  pentium3/sys_reading#455 · Filter by author pentium3 opened on Sep 21, 2026

• LLM-as-Scheduler (LAS): Agentic Workflow Dynamic Scheduling

  pentium3/sys_reading#454 · Filter by author pentium3 opened on Sep 21, 2026

• SpecAgent: A Speculative Retrieval and Forecasting Agent for Code Completion

  pentium3/sys_reading#453 · Filter by author pentium3 opened on Sep 21, 2026

• Efficient Reasoning for LLMs through Speculative Chain-of-Thought

  pentium3/sys_reading#452 · Filter by author pentium3 opened on Sep 21, 2026

• SpecBranch: Speculative Decoding via Hybrid Drafting and Rollback-Aware Branch Parallelism

  pentium3/sys_reading#451 · Filter by author pentium3 opened on Sep 21, 2026

• Cascadia: An Efficient Cascade Serving System for Large Language Models

  pentium3/sys_reading#450 · Filter by author pentium3 opened on Sep 21, 2026

• Demystifying and Enhancing the Efficiency of LLM Based Search Agents

  pentium3/sys_reading#449 · Filter by author pentium3 opened on Sep 21, 2026

• SpecExit: Accelerating Large Reasoning Model via Speculative Exit

  pentium3/sys_reading#448 · Filter by author pentium3 opened on Sep 21, 2026

• ECHO: Elastic Speculative Decoding with Sparse Gating for High-Concurrency Scenarios

  pentium3/sys_reading#447 · Filter by author pentium3 opened on Sep 21, 2026

• PipeSD: Cloud-Edge Collaborative Pipeline Inference Framework with Speculative Decoding

  pentium3/sys_reading#446 · Filter by author pentium3 opened on Sep 21, 2026

• Accelerating RL Post-Training Rollouts via System-Integrated Speculative Decoding

  pentium3/sys_reading#445 · Filter by author pentium3 opened on Aug 25, 2026

• Inference in the Shadows: Taming Memory Bandwidth Contention in Mobile LLM Inference with Sereno

  pentium3/sys_reading#444 · Filter by author pentium3 opened on Jul 19, 2026
  1

• JITServe: SLO-aware LLM Serving with Imprecise Request Information

  pentium3/sys_reading#443 · Filter by author pentium3 opened on Jul 19, 2026
  1

• Beyond Prediction: Tail-Aware Scheduling for LLM Inference

  pentium3/sys_reading#442 · Filter by author pentium3 opened on Jul 11, 2026
  1

• HybridFlow: Resource-Adaptive Subtask Routing for Efficient Edge-Cloud LLM Inference

  pentium3/sys_reading#441 · Filter by author pentium3 opened on Jul 11, 2026

• Chimera: Latency- and Performance-Aware Multi-agent Serving for Heterogeneous LLMs

  pentium3/sys_reading#440 · Filter by author pentium3 opened on Jul 2, 2026
  1

• Pythia: Exploiting Workflow Predictability for Efficient Agent-Native LLM Serving

  pentium3/sys_reading#439 · Filter by author pentium3 opened on Jul 2, 2026
  1