Two Calls Beat Five Agents: Evaluating Multi-Agent Pipelines Against Self-Refinement for Local Language Models

• URL: http://arxiv.org/abs/2607.26922v1
• Date: Wed, 29 Jul 2026 13:53:02 GMT
• Title: Two Calls Beat Five Agents: Evaluating Multi-Agent Pipelines Against Self-Refinement for Local Language Models
• Abstract summary: We investigate Parishad, a structured multi-agent system involving five roles, by deploying it on Qwen2.5-7B-Instruct.<n>Our results demonstrate that communication format and implementation details determine outcomes more than architectural complexity.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Multi-agent LLM pipeline systems break down the task among multiple roles for better reasoning, but are benchmarked mainly with large-scale commercial models. In this study, we investigate Parishad, a structured multi-agent system involving five roles, by deploying it on Qwen2.5-7B-Instruct, a local model, on two datasets: GSM8K (500 questions) and HumanEval (164 questions), compared with prompting directly and two-call self-refinement. The multi-agent system drops GSM8K accuracy from 75.0\% to 45.0\% with JSON data format due to the error accumulation problem. With plaintext format, the accuracy is restored to 82.0\%. A two-call self-refinement strategy (V1) can achieve 86.2\% accuracy on GSM8K, with 7.4$\times$ lower token usage. However, the same V1 implementation on HumanEval---where direct accuracy is already 96.3\%---actively destroys performance (66.5\%). A task-aware gated redesign (V2) applied to HumanEval preserves accuracy at 95.1\%. Our results demonstrate that communication format and implementation details determine outcomes more than architectural complexity, and that simpler approaches match or outperform multi-agent pipelines for local 7B model deployment. All code and data are released.

Related papers

• ReDraft, Don't Just Distill: Reference-Driven Revision for Continual VLLM Post-Training [66.72708893922605]
  Post-training of large multimodal models should add new capabilities while preserving those from pre-training.<n>SFT gives explicit target supervision that learns a task from near-zero accuracy, but its off-policy targets move the model far enough to cause forgetting.<n>We introduce ReDraft, which obtains both from the model's own failures.
  arXiv  Detail & Related papers  (2026-09-15T04:59:03Z)
• AgentWeave: Routing Before Reasoning for Efficient Function Calling in Tool-Rich Language Models [0.0]
  We introduce AgentWeave, a deterministic pre-inference routing layer that constructs a bounded model-visible action space.<n>Relative to all-tools exposure, AgentWeave presents 70.18% fewer tools, uses 61.70% fewer input tokens, and exhibits 50.95% lower mean local-model latency.
  arXiv  Detail & Related papers  (2026-08-24T10:38:08Z)
• Output Format x Model Identity: Interaction Effects in Single-Round Coding Agent Performance [4.342406076796542]
  We conduct a controlled experiment with 3 models (DeepSeek V4, Doubao 2.0, Qwen 3.7 Max)<n>We identify a distinct failure mechanism -- format misuse where agents execute it with excessive scope.<n>We propose a model-specific strategy, a tooldesign principle that constrains format semantics to the agent's own step.
  arXiv  Detail & Related papers  (2026-07-23T10:18:42Z)
• 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)
• Efficient Visual Pointing for Embodied AI:Agent-Driven Data Synthesis, Cross-Block Attention, and Iterative Correction [55.11480729304395]
  We describe our PointArena 2026 solution, which achieves 77.2% overall accuracy and ranks second on the benchmark.<n>The ap proach targets three failure modes. First, agent-driven syn thesis builds large semantic and anchor-relative candidate pools.<n>Second, a determinis tic steerable-data pipeline creates a verified 10,000-sample main set, plus reserve samples, using masks, templates, and path verification.
  arXiv  Detail & Related papers  (2026-06-29T06:39:03Z)
• When More Documents Hurt RAG: Mitigating Vector Search Dilution with Domain-Scoped, Model-Agnostic Retrieval [2.8753422388037486]
  Retrieval-augmented generation degrades when scaled to large document collections.<n>Top-k retrieval increasingly returns semantically similar but contextually incorrect chunks.<n>We propose MASDR-RAG ( Multi-Agent Scoped Domain Retrieval for RAG) to address this dilution.
  arXiv  Detail & Related papers  (2026-06-09T18:26:24Z)
• SkillDAG: Self-Evolving Typed Skill Graphs for LLM Skill Selection at Scale [54.70985426016736]
  We present SkillDAG, which models inter-skill relationships as a typed directed graph.<n>Each search returns vector matches, typed-edge neighbors, and conflict signals, and a propose-then-commit protocol lets the agent register execution-backed edges.<n>On ALFWorld and SkillsBench with MiniMax-M2.7, SkillDAG reaches 67.1% success and 27.3% reward.
  arXiv  Detail & Related papers  (2026-06-02T02:45:21Z)
• When Correct Isn't Usable: Improving Structured Output Reliability in Small Language Models [2.064923532131528]
  Deployed language models must produce outputs that are both correct and format-compliant.<n>We study this structured-output reliability gap using two mathematical benchmarks -- GSM8K and MATH.<n>We develop AloLab, an iterative system-prompt requiring only black-box API access to the target model.
  arXiv  Detail & Related papers  (2026-05-04T09:07:44Z)
• Benchmarking System Dynamics AI Assistants: Cloud Versus Local LLMs on CLD Extraction and Discussion [0.0]
  We present a systematic evaluation of large language model families -- spanning both proprietary cloud APIs and locally-hosted open-context models.<n>On CLD extraction, cloud models achieve 77--89% overall pass rates; the best local model reaches 77% (KimiK2.5GGUFQ3, zero-shot engine mid-tier cloud)<n>On Discussion Discussion, the best local models achieve 50--100% model building steps; only 0-50% error fixing prompts expose memory limits in local deployments.
  arXiv  Detail & Related papers  (2026-04-20T17:53:29Z)
• RSAgent: Learning to Reason and Act for Text-Guided Segmentation via Multi-Turn Tool Invocations [52.752467948588816]
  We propose RSAgent, an agentic Multimodal Large Language Model (MLLM) which interleaves reasoning and action for segmentation via multi-turn tool invocations.<n> RSAgent queries a segmentation toolbox, observes visual feedback, and revises its spatial hypothesis using historical observations to re-localize targets and iteratively refine masks.<n>Extensive experiments show that RSAgent achieves a zero-shot performance of 66.5% gIoU on ReasonSeg test, improving over Seg-Zero-7B by 9%, and reaches 81.5% cIoU on RefCOCOg, demonstrating state-of-the-art performance
  arXiv  Detail & Related papers  (2025-12-30T06:50:11Z)
• DoVer: Intervention-Driven Auto Debugging for LLM Multi-Agent Systems [48.971606069204825]
  DoVer is an intervention-driven debug framework for large language model (LLM)-based multi-agent systems.<n>It augments hypothesis generation with active verification through targeted interventions.<n>DoVer flips 18-28% of failed trials into successes, achieves up to 16% milestone progress, and validates or refutes 30-60% of failure hypotheses.
  arXiv  Detail & Related papers  (2025-12-07T09:23:48Z)
• ToolACE: Winning the Points of LLM Function Calling [139.07157814653638]
  ToolACE is an automatic agentic pipeline designed to generate accurate, complex, and diverse tool-learning data.<n>We demonstrate that models trained on our synthesized data, even with only 8B parameters, achieve state-of-the-art performance on the Berkeley Function-Calling Leaderboard.
  arXiv  Detail & Related papers  (2024-09-02T03:19:56Z)

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.