Phase-Scheduled Multi-Agent Systems for Token-Efficient Coordination

• URL: http://arxiv.org/abs/2604.17400v1
• Date: Sun, 19 Apr 2026 12:03:54 GMT
• Title: Phase-Scheduled Multi-Agent Systems for Token-Efficient Coordination
• Abstract summary: Phase-Scheduled Multi-Agent Systems (PSMAS) is a framework that reconceptualizes agent activation as continuous control over a shared attention space modeled on a circular manifold.<n>PSMAS achieves a mean token reduction of 27.3 percent (range 21.4-34.8 percent) while maintaining task performance within 2.1 percentage points of a fully activated baseline.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Multi-agent systems (MAS) powered by large language models suffer from severe token inefficiency arising from two compounding sources: (i) unstructured parallel execution, where all agents activate simultaneously irrespective of input readiness; and (ii) unrestricted context sharing, where every agent receives the full accumulated context regardless of relevance. Existing mitigation strategies - static pruning, hierarchical decomposition, and learned routing - treat coordination as a structural allocation problem and fundamentally ignore its temporal dimension. We propose Phase-Scheduled Multi-Agent Systems (PSMAS), a framework that reconceptualizes agent activation as continuous control over a shared attention space modeled on a circular manifold. Each agent i is assigned a fixed angular phase theta_i in the range [0, 2*pi], derived from the task dependency topology; a global sweep signal phi(t) rotates at velocity omega, activating only agents within an angular window epsilon. Idle agents receive compressed context summaries, reducing per-step token consumption. We implement PSMAS on LangGraph, evaluate on four structured benchmarks (HotPotQA-MAS, HumanEval-MAS, ALFWorld-Multi, WebArena-Coord) and two unstructured conversational settings, and prove stability, convergence, and optimality results for the sweep dynamics. PSMAS achieves a mean token reduction of 27.3 percent (range 21.4-34.8 percent) while maintaining task performance within 2.1 percentage points of a fully activated baseline (p < 0.01, n = 500 per configuration), and outperforms the strongest learned routing baseline by 5.6 percentage points in token reduction with 2.0 percentage points less performance drop. Crucially, we show that scheduling and compression are independent sources of gain: scheduling alone accounts for 18-20 percentage points of reduction, robust to compression degradation up to alpha = 0.40.

Related papers

• Accelerating Unified Multimodal Models with Core-Expansion Routing and Unified Computation Scheduling [61.28599393707943]
  Unified multimodal models jointly support understanding and generation, but incur substantial redundant computation across tokens, layers, and generation timesteps.<n>We propose a task-shared core scorer and progress-conditioned generation expansions, optimized through generation decomposition and cross-task core alignment.<n> Experiments on two representative UMM architectures demonstrate consistent quality--efficiency improvements across both tasks, retaining 98.03% of dense understanding performance with a 1.93times$ end-to-end inference speedup.
  arXiv  Detail & Related papers  (2026-08-29T14:27:55Z)
• PLAN: Parallel Liquid-Inspired Approximation Network for Efficient Representation Learning in Flexible Job Shop Scheduling [11.920717162099818]
  PLAN is a lightweight representation learning framework that reformulates continuous liquid-state dynamics into a discretized and parallelizable formulation.<n>PLAN reduces the average makespan by 1.2%, 1.4%, and 2.3%, respectively, compared with the corresponding state-of-the-art baselines.<n>PLAN also reduces average inference latency by 13.2%, 31.7%, and 26.9%, respectively, with a maximum reduction of 69.2% on the largest instances.
  arXiv  Detail & Related papers  (2026-08-04T02:43:27Z)
• Multi-Timescale Latent-Action DRL for Joint Optimization in Edge-Cloud Networks [48.421359451414155]
  Load imbalance across edge cloud and layers degrades latency performance in.<n> hierarchical edge-cloud computing systems under dynamic task arrivals and heterogeneous resources.<n>We propose a two-time deep learning study with an action space (2MDRL-LA) to jointly optimize service placement.<n>It achieves a 20.8% reduction in average e2e latency and a 13% improvement in resource utilization over the scheme.
  arXiv  Detail & Related papers  (2026-06-30T11:56:15Z)
• Hierarchical Reinforcement Learning for Neural Network Compression (HiReLC): Pruning and Quantization [1.7616042687330635]
  HiReLC is a hierarchical ensemble-reinforcement learning framework for automated joint quantization and structured pruning of deep neural networks.<n>Low-level agents (LLAs) operate independently per block, selecting per- Kernel configurations over a multi-discrete action space.<n>High-level agents (HLAs) coordinate global budget allocation via ensemble voting guided by Fisher Information-based sensitivity estimates.
  arXiv  Detail & Related papers  (2026-06-24T16:19:29Z)
• A Delta-Aware Orchestration Framework for Scalable Multi-Agent Edge Computing [0.0]
  The Synergistic Collapse occurs when scaling beyond 100 agents causes superlinear performance degradation that individual optimizations cannot prevent.<n>Prior work has addressed each contributing factor in isolation: exponential action-space growth, computational redundancy among spatially adjacent agents, and task-agnostic hardware scheduling.<n>We present a framework that addresses all three simultaneously through: (1) Differential Neural Caching, which stores intermediate layer activations and computes only the input deltas; (2) Criticality-Based Action Space Pruning, which organizes agents into priority tiers and reduces coordination complexity from O(n2) to O(n log n) with less than 6% optimal
  arXiv  Detail & Related papers  (2026-04-22T02:54:35Z)
• Dynamic Attentional Context Scoping: Agent-Triggered Focus Sessions for Isolated Per-Agent Steering in Multi-Agent LLM Orchestration [0.0]
  Multi-agent LLM orchestration systems suffer from context pollution when N concurrent agents compete for the orchestrator's context window.<n>We introduce Dynamic Attentional Context Scoping (DACS), a mechanism in which the orchestrator operates in two asymmetric modes.<n>DACS achieves 90.0--98.4% steering accuracy versus 21.0--60.0% for a flat-context baseline.
  arXiv  Detail & Related papers  (2026-04-09T07:26:42Z)
• Prune Redundancy, Preserve Essence: Vision Token Compression in VLMs via Synergistic Importance-Diversity [54.95089105944234]
  Vision-language models (VLMs) face significant computational inefficiencies caused by excessive generation of visual tokens.<n>We propose PruneSID, a training-free Synergistic Importance-Diversity approach featuring a two-stage pipeline.<n>We demonstrate state-of-the-art performance, achieving 96.3% accuracy on LLaVA-1.5 with only 11.1% token retention, and 92.8% accuracy at extreme compression rates (5.6%) on LLaVA-NeXT.
  arXiv  Detail & Related papers  (2026-03-10T10:31:58Z)
• HCFT: Hierarchical Convolutional Fusion Transformer for EEG Decoding [9.572621097681646]
  We propose a lightweight decoding framework named Hierarchical Conencephaloal Fusion Transformer (HCFT)<n>HCFT combines dual-branchal encoders and hierarchical Transformer blocks for multi-scale representation.<n>Results show that HCFT achieves 80.83% average accuracy and a Cohen's kappa of 0.6165 on BCI IV-2b, as well as 99.10% sensitivity, 0.0236 false positives per hour, and 98.82% specificity on CHB-MIT.
  arXiv  Detail & Related papers  (2026-01-18T06:36:30Z)
• Towards Efficient Agents: A Co-Design of Inference Architecture and System [66.59916327634639]
  This paper presents AgentInfer, a unified framework for end-to-end agent acceleration.<n>We decompose the problem into four synergistic components: AgentCollab, AgentSched, AgentSAM, and AgentCompress.<n>Experiments on the BrowseComp-zh and DeepDiver benchmarks demonstrate that through the synergistic collaboration of these methods, AgentInfer reduces ineffective token consumption by over 50%.
  arXiv  Detail & Related papers  (2025-12-20T12:06:13Z)
• Towards a Science of Scaling Agent Systems [79.64446272302287]
  We formalize a definition for agent evaluation and characterize scaling laws as the interplay between agent quantity, coordination structure, modelic, and task properties.<n>We derive a predictive model using coordination metrics, that cross-validated R2=0, enabling prediction on unseen task domains.<n>We identify three effects: (1) a tool-coordination trade-off: under fixed computational budgets, tool-heavy tasks suffer disproportionately from multi-agent overhead, and (2) a capability saturation: coordination yields diminishing or negative returns once single-agent baselines exceed 45%.
  arXiv  Detail & Related papers  (2025-12-09T06:52:21Z)
• Scalable spectral representations for multi-agent reinforcement learning in network MDPs [13.782868855372774]
  A popular model for multi-agent control, Network Markov Decision Processes (MDPs) pose a significant challenge to efficient learning. We first derive scalable spectral local representations for network MDPs, which induces a network linear subspace for the local $Q$-function of each agent. We design a scalable algorithmic framework for continuous state-action network MDPs, and provide end-to-end guarantees for the convergence of our algorithm.
  arXiv  Detail & Related papers  (2024-10-22T17:45:45Z)
• Toward multi-target self-organizing pursuit in a partially observable Markov game [34.22625222101752]
  This work proposes a framework for decentralized multi-agent systems to improve the implicit coordination capabilities in search and pursuit. We model a self-organizing system as a partially observable Markov game (POMG) featured by large-scale, decentralization, partial observation, and noncommunication. The proposed distributed algorithm: fuzzy self-organizing cooperative coevolution (FSC2) is then leveraged to resolve the three challenges in multi-target SOP.
  arXiv  Detail & Related papers  (2022-06-24T14:59:56Z)
• Collaborative Intelligent Reflecting Surface Networks with Multi-Agent Reinforcement Learning [63.83425382922157]
  Intelligent reflecting surface (IRS) is envisioned to be widely applied in future wireless networks. In this paper, we investigate a multi-user communication system assisted by cooperative IRS devices with the capability of energy harvesting.
  arXiv  Detail & Related papers  (2022-03-26T20:37:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.