Do More Agents Help? Controlled and Protocol-Aligned Evaluation of LLM Agent Workflows

• URL: http://arxiv.org/abs/2606.05670v1
• Date: Thu, 04 Jun 2026 03:50:47 GMT
• Title: Do More Agents Help? Controlled and Protocol-Aligned Evaluation of LLM Agent Workflows
• Abstract summary: BenchAgent places single-agent, fixed multi-agent (MAS) and evolving MAS under one normalized execution and logging protocol.<n> BenchAgent evaluates these substrate-internal across ten reasoning, coding, and tool-use benchmarks with GPT-4.1.<n>On the PAE GAIA snapshot, a Claude-Code-style runtime workflow reaches 66.72% overall and 69.23% on Level 3.
• Score: 8.909548943361637
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Does adding more agents help an LLM workflow once compared systems share the same benchmark loader, tool access, answer contract, usage accounting, and trajectory logging? We introduce BenchAgent, an evaluation framework that places single-agent, fixed multi-agent (MAS), and evolving MAS workflows under one normalized execution and logging protocol. BenchAgent evaluates these substrate-internal workflows across ten reasoning, coding, and tool-use benchmarks with GPT-4.1, and separately reports a Protocol-Aligned External (PAE) GAIA study of a runtime-generated workflow. Under SI conditions, at most one of six tested MAS exceeds the matched single-agent anchor on benchmark-balanced average accuracy: EvoAgent lies within the Wilson one-run guidance, while the remaining five trail by 2.56-11.29 points and occupy more expensive accuracy-cost trade-offs. On the PAE GAIA snapshot, a Claude-Code-style runtime workflow reaches 66.72% overall and 69.23% on Level 3, more than 20 points above the strongest non-Claude baseline, Jarvis, a fixed MAS.

Related papers

• ClawArena-Team: Benchmarking Subagent Orchestration and Dynamic Workflows in Language-Model Agents [61.184954205350984]
  ClawArena-Team is a benchmark of 41 multi-turn, multimodal, multi-agent scenarios spanning 258 evaluation rounds and 72 staged updates.<n>All scoring is execution-based with no LLM judge: an overall score -- the Subagent-Management Score (SMS) -- multiplies task correctness by a least-privilege and modality-routing factor.
  arXiv  Detail & Related papers  (2026-06-30T06:06:08Z)
• CFAgentBench: A Reproducible Environment and Benchmark for Autonomous Construction-Finance Agents [0.0]
  CFAgentBench is a benchmark for autonomous construction-finance agents.<n>It contains 1,014 machine-gradeable task specifications across 8 domains and 77 families.<n>The strongest agent reaches pass1 = 0.67 but only pass5 = 0.38.
  arXiv  Detail & Related papers  (2026-06-20T11:34:52Z)
• ExecuGraph: A Multi-Agent, Execution-Grounded Framework for Reliable Backend Code Synthesis with Large Language Models [2.3004655342211078]
  Large Language Models generate plausible backend code, but a single-pass paradigm provides no guarantee of correctness or runtime reliability.<n>We present ExecuGraph, a multi-agent framework that places execution-based validation at the center of backend code.<n>A per-agent ablation, retry-budget sweep, error-class taxonomy, and test-source audit are reported.
  arXiv  Detail & Related papers  (2026-06-18T05:58:17Z)
• EngiAI: A Multi-Agent Framework and Benchmark Suite for LLM-Driven Engineering Design [0.4499833362998488]
  We introduce a benchmark suite with three evaluation dimensions.<n>We present EngiAI, a Multi-Agent System (MAS) implementation built on LangGraph.
  arXiv  Detail & Related papers  (2026-05-19T12:12:09Z)
• MetaCogAgent: A Metacognitive Multi-Agent LLM Framework with Self-Aware Task Delegation [11.486913474115402]
  We propose MetaCogAgent, a multi-agent large language model (LLM) framework where each agent is equipped with a Metacognitive Self-Assessment Unit.<n>We show that MetaCogAgent achieves 82.4% task accuracy while using 5% fewer API calls than AutoGen and 34% fewer than ensemble voting.
  arXiv  Detail & Related papers  (2026-05-17T07:12:04Z)
• Claw-Eval-Live: A Live Agent Benchmark for Evolving Real-World Workflows [67.92316850084575]
  We introduce Claw-Eval-Live, a live benchmark for workflow agents.<n>Each release is constructed from public workflow-demand signals.<n>Claw-Eval-Live records execution traces, audit logs, service state, and post-run workspace artifacts.
  arXiv  Detail & Related papers  (2026-04-30T17:23:19Z)
• Graph-of-Agents: A Graph-based Framework for Multi-Agent LLM Collaboration [57.23877089929136]
  Graph-of-Agents (GoA) is a new graph-based framework for modeling multi-agent LLM communication.<n>GoA achieves superior performance using only 3 selected agents, outperforming recent multi-agent LLM baselines.
  arXiv  Detail & Related papers  (2026-04-18T21:13:03Z)
• RewardHackingAgents: Benchmarking Evaluation Integrity for LLM ML-Engineering Agents [0.9821874476902969]
  LLM agents increasingly perform end-to-end ML engineering tasks where success is judged by a single scalar test metric.<n>This creates a structural vulnerability: an agent can increase the reported score by compromising the evaluation pipeline rather than improving the model.<n>We introduce RewardHackingAgents, a workspace-based benchmark that makes two compromise vectors explicit and measurable.
  arXiv  Detail & Related papers  (2026-03-11T22:06:44Z)
• Beyond Rule-Based Workflows: An Information-Flow-Orchestrated Multi-Agents Paradigm via Agent-to-Agent Communication from CORAL [0.15199492741752027]
  We propose an Information-Flow-Orchestrated Multi-Agent Paradigm via Agent-to-Agent (A2A) Communication.<n>We evaluate our approach on the general-purpose benchmark GAIA, using the representative workflow-based MAS as the baseline.<n>Our method achieves 63.64% accuracy, outperforming OWL's 55.15% by 8.49 percentage points with comparable token consumption.
  arXiv  Detail & Related papers  (2026-01-14T21:35:51Z)
• Self-Abstraction from Grounded Experience for Plan-Guided Policy Refinement [61.35824395228412]
  Large language model (LLM) based agents are increasingly used to tackle software engineering tasks.<n>We propose Self-Abstraction from Grounded Experience (SAGE), a framework that enables agents to learn from their own task executions.
  arXiv  Detail & Related papers  (2025-11-08T08:49:38Z)
• Runaway is Ashamed, But Helpful: On the Early-Exit Behavior of Large Language Model-based Agents in Embodied Environments [54.67512489842682]
  Large language models (LLMs) have demonstrated strong planning and decision-making capabilities in complex embodied environments.<n>We take a first step toward exploring the early-exit behavior for LLM-based agents.
  arXiv  Detail & Related papers  (2025-05-23T08:23:36Z)
• SOPBench: Evaluating Language Agents at Following Standard Operating Procedures and Constraints [59.645885492637845]
  SOPBench is an evaluation pipeline that transforms each service-specific SOP code program into a directed graph of executable functions.<n>Our approach transforms each service-specific SOP code program into a directed graph of executable functions and requires agents to call these functions based on natural language SOP descriptions.<n>We evaluate 18 leading models, and results show the task is challenging even for top-tier models.
  arXiv  Detail & Related papers  (2025-03-11T17:53:02Z)

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.