DebFlow: Automating Agent Creation via Agent Debate

• URL: http://arxiv.org/abs/2503.23781v1
• Date: Mon, 31 Mar 2025 06:56:13 GMT
• Title: DebFlow: Automating Agent Creation via Agent Debate
• Authors: Jinwei Su, Yinghui Xia, Ronghua Shi, Jianhui Wang, Jianuo Huang, Yijin Wang, Tianyu Shi, Yang Jingsong, Lewei He,
• Abstract summary: DebFlow is a framework that employs a debate mechanism to optimize and integrates reflexion to improve.<n>We evaluated our method across six benchmark datasets including HotpotQA, MATH, and ALFWorld.<n>During training, our framework reduces resource consumption by 37% compared to the state-of-the-art baselines.
• Score: 3.7606626616500947
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models (LLMs) have demonstrated strong potential and impressive performance in automating the generation and optimization of workflows. However, existing approaches are marked by limited reasoning capabilities, high computational demands, and significant resource requirements. To address these issues, we propose DebFlow, a framework that employs a debate mechanism to optimize workflows and integrates reflexion to improve based on previous experiences. We evaluated our method across six benchmark datasets, including HotpotQA, MATH, and ALFWorld. Our approach achieved a 3\% average performance improvement over the latest baselines, demonstrating its effectiveness in diverse problem domains. In particular, during training, our framework reduces resource consumption by 37\% compared to the state-of-the-art baselines. Additionally, we performed ablation studies. Removing the Debate component resulted in a 4\% performance drop across two benchmark datasets, significantly greater than the 2\% drop observed when the Reflection component was removed. These findings strongly demonstrate the critical role of Debate in enhancing framework performance, while also highlighting the auxiliary contribution of reflexion to overall optimization.

Related papers

• DARS: Dynamic Action Re-Sampling to Enhance Coding Agent Performance by Adaptive Tree Traversal [55.13854171147104]
  Large Language Models (LLMs) have revolutionized various domains, including natural language processing, data analysis, and software development. We present Dynamic Action Re-Sampling (DARS), a novel inference time compute scaling approach for coding agents. We evaluate our approach on SWE-Bench Lite benchmark, demonstrating that this scaling strategy achieves a pass@k score of 55% with Claude 3.5 Sonnet V2.
  arXiv  Detail & Related papers  (2025-03-18T14:02:59Z)
• HuixiangDou2: A Robustly Optimized GraphRAG Approach [11.91228019623924]
  Graph-based Retrieval-Augmented Generation (GraphRAG) addresses this by structuring it as a graph for dynamic retrieval.<n>We introduce HuixiangDou2, a robustly optimized GraphRAG framework.<n>Specifically, we leverage the effectiveness of dual-level retrieval and optimize its performance in a 32k context.
  arXiv  Detail & Related papers  (2025-03-09T06:20:24Z)
• Towards Robust Universal Information Extraction: Benchmark, Evaluation, and Solution [66.11004226578771]
  Existing robust benchmark datasets have two key limitations.<n>They generate only a limited range of perturbations for a single Information Extraction (IE) task.<n>Considering the powerful generation capabilities of Large Language Models (LLMs), we introduce a new benchmark dataset for Robust UIE, called RUIE-Bench.<n>We show that training with only textbf15% of the data leads to an average textbf7.5% relative performance improvement across three IE tasks.
  arXiv  Detail & Related papers  (2025-03-05T05:39:29Z)
• Dynamic Noise Preference Optimization for LLM Self-Improvement via Synthetic Data [51.62162460809116]
  We introduce Dynamic Noise Preference Optimization (DNPO) to ensure consistent improvements across iterations.<n>In experiments with Zephyr-7B, DNPO consistently outperforms existing methods, showing an average performance boost of 2.6%.<n> DNPO shows a significant improvement in model-generated data quality, with a 29.4% win-loss rate gap compared to the baseline in GPT-4 evaluations.
  arXiv  Detail & Related papers  (2025-02-08T01:20:09Z)
• ScoreFlow: Mastering LLM Agent Workflows via Score-based Preference Optimization [51.280919773837645]
  We develop ScoreFlow, a high-performance framework for agent workflow optimization.<n>ScoreFlow incorporates Score-DPO, a novel variant of the direct preference optimization method that accounts for quantitative feedback.<n>It achieves an 8.2% improvement over existing baselines across question answering, coding, and mathematical reasoning.
  arXiv  Detail & Related papers  (2025-02-06T18:47:49Z)
• Automatic Evaluation for Text-to-image Generation: Task-decomposed Framework, Distilled Training, and Meta-evaluation Benchmark [62.58869921806019]
  We propose a task decomposition evaluation framework based on GPT-4o to automatically construct a new training dataset. We design innovative training strategies to effectively distill GPT-4o's evaluation capabilities into a 7B open-source MLLM, MiniCPM-V-2.6. Experimental results demonstrate that our distilled open-source MLLM significantly outperforms the current state-of-the-art GPT-4o-base baseline.
  arXiv  Detail & Related papers  (2024-11-23T08:06:06Z)
• RA-ISF: Learning to Answer and Understand from Retrieval Augmentation via Iterative Self-Feedback [19.28222902440827]
  Large language models (LLMs) demonstrate exceptional performance in numerous tasks but still heavily rely on knowledge stored in their parameters. Retrieval-augmented generation (RAG) methods address this issue by integrating external knowledge. We propose Retrieval Augmented Iterative Self-Feedback (RA-ISF), a framework that iteratively decomposes tasks and processes them in three submodules to enhance the model's problem-solving capabilities.
  arXiv  Detail & Related papers  (2024-03-11T16:01:05Z)
• Re-Evaluating LiDAR Scene Flow for Autonomous Driving [80.37947791534985]
  Popular benchmarks for self-supervised LiDAR scene flow have unrealistic rates of dynamic motion, unrealistic correspondences, and unrealistic sampling patterns. We evaluate a suite of top methods on a suite of real-world datasets. We show that despite the emphasis placed on learning, most performance gains are caused by pre- and post-processing steps.
  arXiv  Detail & Related papers  (2023-04-04T22:45:50Z)

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.