CoWork-X: Experience-Optimized Co-Evolution for Multi-Agent Collaboration System

• URL: http://arxiv.org/abs/2602.05004v1
• Date: Wed, 04 Feb 2026 19:42:14 GMT
• Title: CoWork-X: Experience-Optimized Co-Evolution for Multi-Agent Collaboration System
• Abstract summary: CoWork-X casts peer collaboration as a closed-loop optimization problem across episodes.<n>CoWork-X achieves stable, cumulative performance gains while steadily reducing online latency and token usage.
• Score: 43.33554364574611
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models are enabling language-conditioned agents in interactive environments, but highly cooperative tasks often impose two simultaneous constraints: sub-second real-time coordination and sustained multi-episode adaptation under a strict online token budget. Existing approaches either rely on frequent in-episode reasoning that induces latency and timing jitter, or deliver post-episode improvements through unstructured text that is difficult to compile into reliable low-cost execution. We propose CoWork-X, an active co-evolution framework that casts peer collaboration as a closed-loop optimization problem across episodes, inspired by fast--slow memory separation. CoWork-X instantiates a Skill-Agent that executes via HTN (hierarchical task network)-based skill retrieval from a structured, interpretable, and compositional skill library, and a post-episode Co-Optimizer that performs patch-style skill consolidation with explicit budget constraints and drift regularization. Experiments in challenging Overcooked-AI-like realtime collaboration benchmarks demonstrate that CoWork-X achieves stable, cumulative performance gains while steadily reducing online latency and token usage.

Related papers

• EASy: Towards Efficient LLM-Based Agentic System [54.10819421625103]
  We propose EASy, a trainable agentic framework that jointly optimize task performance and computational efficiency through reinforcement learning.<n> EASy equips an LLM-based orchestrator with explicit knowledge of the capability and cost profiles of heterogeneous executors, enabling context-sensitive coordination beyond performance-only routing.<n>It further introduces a milestone-plan-act workflow that decomposes complex tasks into manageable milestones, constructs dependency-aware execution graphs, assigns suitable executors, and parallelizes independent steps while adapting subsequent decisions to intermediate outcomes.
  arXiv  Detail & Related papers  (2026-08-05T08:50:36Z)
• Dual-Process Atomic Skill Learning: Decoupling Semantic Reasoning and Real-Time Control [21.97625864571452]
  Language-conditioned Imitation Learning (IL) is essential for enabling robots to perform complex tasks following natural language instructions.<n>We propose Dual-Process Atomic Skill Learning (DASL), a novel hierarchical imitation learning framework that decouples slow semantic reasoning from real-time motion control.<n>DASL significantly outperforms state-of-the-art baselines, excelling in skill acquisition and compositional generalization to unseen instructions.
  arXiv  Detail & Related papers  (2026-07-12T07:38:53Z)
• When Parallelism Pays Off: Cohesion-Aware Task Partitioning for Multi-Agent Coding [19.03671411536556]
  Multi-agent Large Language Model (LLM) systems offer a way to decompose complex tasks, such as coding, through parallelization and context isolation.<n>We formalize multi-agent orchestration as a graph partitioning problem that captures the communication-to-computation trade-off.<n>We present Cohesion-aware Coder (Co-Coder), which builds dependency graphs from static analysis, isolates structural hub files, partitions the graph via community detection, and executes the partition with a dependency-aware scheduler.
  arXiv  Detail & Related papers  (2026-05-31T02:10:12Z)
• Uni-Synergy: Bridging Understanding and Generation for Personalized Reasoning via Co-operative Reinforcement Learning [56.99010101756807]
  We propose Sync-R1, an end-to-end reinforcement learning framework that jointly optimize personalized understanding and generation.<n> Sync-R1 enables personalized comprehension to guide content creation, while the resulting generation quality reciprocally refines understanding.<n>We also introduce Dynamic Group Scaling (DGS), which adaptively filters low-potential trajectories to reduce gradient variance and accelerate convergence.<n> Experimental results demonstrate that Sync-R1 achieves state-of-the-art performance, showcasing superior cross-task reasoning and robust personalization.
  arXiv  Detail & Related papers  (2026-05-11T12:18:26Z)
• Improving the Efficiency of Language Agent Teams with Adaptive Task Graphs [52.26652574704317]
  Large language models (LLMs) are increasingly deployed in teams, yet existing coordination approaches often occupy two extremes.<n>We introduce Language Agent Teams for Task Evolution (LATTE), a framework for coordinating LLM teams inspired by distributed systems.
  arXiv  Detail & Related papers  (2026-05-07T14:19:17Z)
• Effective Strategies for Asynchronous Software Engineering Agents [52.76455094324273]
  We introduce a structured multi-agent coordination paradigm grounded in three core SWE primitives: centralized task delegation, asynchronous execution, and isolated workspaces.<n>We find that CAID improves accuracy over single-agent baselines by 26.7% absolute on paper reproduction tasks (PaperBench) and 14.3% on Python library development tasks (Commit0).
  arXiv  Detail & Related papers  (2026-03-23T02:26:35Z)
• EmCoop: A Framework and Benchmark for Embodied Cooperation Among LLM Agents [18.802912315746564]
  EmCoop is a benchmark framework for studying cooperation in embodied multi-agent systems.<n>Our framework separates a high-level cognitive layer from a low-level embodied interaction layer.<n>We propose generalizable, process-level metrics that diagnose collaboration quality and failure modes.
  arXiv  Detail & Related papers  (2026-02-27T22:28:33Z)
• CoDA: A Context-Decoupled Hierarchical Agent with Reinforcement Learning [12.710191300398924]
  We introduce CoDA, a reinforcement learning framework that decouples high-level planning from low-level execution.<n>CoDA achieves significant performance improvements over state-of-the-art baselines on complex multi-hop question-answering benchmarks.
  arXiv  Detail & Related papers  (2025-12-14T14:41:29Z)
• The Era of Agentic Organization: Learning to Organize with Language Models [107.41382234213893]
  We introduce asynchronous thinking (AsyncThink) as a new paradigm of reasoning with large language models.<n> Experiments demonstrate that AsyncThink achieves 28% lower inference latency compared to parallel thinking.<n>AsyncThink generalizes its learned asynchronous thinking capabilities, effectively tackling unseen tasks without additional training.
  arXiv  Detail & Related papers  (2025-10-30T16:25:10Z)
• ParaCook: On Time-Efficient Planning for Multi-Agent Systems [62.471032881396496]
  Large Language Models (LLMs) exhibit strong reasoning abilities for planning long-horizon, real-world tasks.<n>We present ParaCook, a benchmark for time-efficient collaborative planning.
  arXiv  Detail & Related papers  (2025-10-13T16:47:07Z)
• HAWK: A Hierarchical Workflow Framework for Multi-Agent Collaboration [3.2588674134593942]
  Multi-agent systems face persistent challenges in cross-platform interoperability, dynamic task scheduling, and efficient resource sharing.<n>We propose Hierarchical Agent (Hawk), a modular framework comprising five layers-User, Operator, Agent, Resource-and supported by sixteen standardized interfaces.<n>Hawk delivers an end-to-end pipeline covering task parsing, workflow orchestration, intelligent scheduling, resource invocation, and data synchronization.
  arXiv  Detail & Related papers  (2025-07-05T15:03:53Z)
• Multi-Agent Collaboration via Evolving Orchestration [55.574417128944226]
  Large language models (LLMs) have achieved remarkable results across diverse downstream tasks, but their monolithic nature restricts scalability and efficiency in complex problem-solving.<n>We propose a puppeteer-style paradigm for LLM-based multi-agent collaboration, where a centralized orchestrator ("puppeteer") dynamically directs agents ("puppets") in response to evolving task states.<n> Experiments on closed- and open-domain scenarios show that this method achieves superior performance with reduced computational costs.
  arXiv  Detail & Related papers  (2025-05-26T07:02:17Z)
• Collaborative Speculative Inference for Efficient LLM Inference Serving [22.3288901160418]
  CoSine is a novel speculative inference system that decouples sequential speculative decoding from parallel verification.<n>With equivalent resource costs, CoSine achieves up to a 23.2% decrease in latency and a 32.5% increase in throughput compared to baseline methods.
  arXiv  Detail & Related papers  (2025-03-13T13:03:38Z)
• Haste Makes Waste: Evaluating Planning Abilities of LLMs for Efficient and Feasible Multitasking with Time Constraints Between Actions [56.88110850242265]
  We present Recipe2Plan, a novel benchmark framework based on real-world cooking scenarios.<n>Unlike conventional benchmarks, Recipe2Plan challenges agents to optimize cooking time through parallel task execution.
  arXiv  Detail & Related papers  (2025-03-04T03:27:02Z)
• KV Cache Compression, But What Must We Give in Return? A Comprehensive Benchmark of Long Context Capable Approaches [52.02764371205856]
  Long context capability is a crucial competency for large language models (LLMs) This work provides a taxonomy of current methods and evaluating 10+ state-of-the-art approaches across seven categories of long context tasks.
  arXiv  Detail & Related papers  (2024-07-01T17:59:47Z)

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.