Explainable Innovation Engine: Dual-Tree Agent-RAG with Methods-as-Nodes and Verifiable Write-Back

Renwei Meng

cs.AIMar 10, 2026

Explainable Innovation Engine: Dual-Tree Agent-RAG with Methods-as-Nodes and Verifiable Write-Back

Authors: Renwei Meng

Organizations: Anhui University, Hefei, China

Abstract

Retrieval-augmented generation (RAG) improves factual grounding, yet most systems rely on flat chunk retrieval and provide limited control over multi-step synthesis. We propose an Explainable Innovation Engine that upgrades the knowledge unit from text chunks to methods-as-nodes. The engine maintains a weighted method provenance tree for traceable derivations and a hierarchical clustering abstraction tree for efficient top-down navigation. At inference time, a strategy agent selects explicit synthesis operators (e.g., induction, deduction, analogy), composes new method nodes, and records an auditable trajectory. A verifier-scorer layer then prunes low-quality candidates and writes validated nodes back to support continual growth. Expert evaluation across six domains and multiple backbones shows consistent gains over a vanilla baseline, with the largest improvements on derivation-heavy settings, and ablations confirm the complementary roles of provenance backtracking and pruning. These results suggest a practical path toward controllable, explainable, and verifiable innovation in agentic RAG systems. Code is available at the project GitHub repository https://github.com/xiaolu-666113/Dual-Tree-Agent-RAG.

Figures & tables

Figure 1: The complete algorithm process

Figure 2: A schematic diagram of Clustering Abstraction Tree and Method Provenance Tree

Domain	GPT-5.2	Gemini 3.0	Llama4 70B	DeepSeek
Sociology	4.52 $\pm$ 0.35	4.41 $\pm$ 0.38	4.28 $\pm$ 0.41	4.12 $\pm$ 0.45
Computer Science	4.35 $\pm$ 0.37	4.23 $\pm$ 0.39	4.10 $\pm$ 0.42	3.96 $\pm$ 0.44
Biology	4.08 $\pm$ 0.40	3.97 $\pm$ 0.42	3.86 $\pm$ 0.45	3.73 $\pm$ 0.47
Chemistry	4.06 $\pm$ 0.41	3.95 $\pm$ 0.43	3.85 $\pm$ 0.46	3.71 $\pm$ 0.48
Physics	4.04 $\pm$ 0.42	3.93 $\pm$ 0.44	3.83 $\pm$ 0.46	3.70 $\pm$ 0.49
Mathematics	3.60 $\pm$ 0.48	3.45 $\pm$ 0.50	3.28 $\pm$ 0.52	3.10 $\pm$ 0.55

Table 1: Agent-RAG expert scores (

\mu\pm\sigma

Domain	GPT-5.2	Gemini 3.0	Llama4 70B	DeepSeek
Sociology	4.30 $\pm$ 0.36	4.20 $\pm$ 0.38	4.05 $\pm$ 0.41	3.92 $\pm$ 0.44
Computer Science	4.05 $\pm$ 0.38	3.95 $\pm$ 0.40	3.80 $\pm$ 0.43	3.68 $\pm$ 0.45
Biology	3.66 $\pm$ 0.41	3.57 $\pm$ 0.42	3.45 $\pm$ 0.45	3.34 $\pm$ 0.47
Chemistry	3.64 $\pm$ 0.41	3.55 $\pm$ 0.43	3.44 $\pm$ 0.46	3.32 $\pm$ 0.48
Physics	3.62 $\pm$ 0.42	3.53 $\pm$ 0.44	3.42 $\pm$ 0.46	3.30 $\pm$ 0.49
Mathematics	2.78 $\pm$ 0.52	2.62 $\pm$ 0.54	2.45 $\pm$ 0.56	2.28 $\pm$ 0.58

Table 2: Baseline (plain chat) expert scores (

\mu\pm\sigma

Domain	Avg. $\Delta$
Mathematics	+0.83
Biology	+0.41
Chemistry	+0.42
Physics	+0.41
Computer Science	+0.29
Sociology	+0.21

Table 3: Average improvement

\Delta

(Agent-RAG minus Baseline), averaged across backbones.

Backbone	Agent-RAG	Baseline	Avg. $\Delta$
GPT-5.2	4.11	3.68	+0.43
Gemini 3.0	3.99	3.57	+0.42
Llama4 70B	3.87	3.44	+0.43
DeepSeek	3.72	3.31	+0.41

Table 4: Average scores across domains.

Figure 3: Ablation results: score drop when removing each module. Larger values indicate higher sensitivity.

Figure 4: Cost–quality trade-off: expert score vs. token cost, colored by latency. The curve indicates the Pareto frontier.

Appendix figures & tables1 asset

Supplementary material from the paper’s appendix.

Appendix

Domain	Backbone	n	$\boldsymbol{\bar{\Delta}}$	$\boldsymbol{t}$	$\boldsymbol{d_{z}}$	$\boldsymbol{p_{\mathrm{Holm}}}$ (t / Wilcoxon)
Mathematics	GPT-5.2	100	+0.82	12.30	1.23	$2.0\times 10^{-21}$ / $1.1\times 10^{-19}$
Physics	GPT-5.2	100	+0.42	9.10	0.91	$3.8\times 10^{-15}$ / $7.2\times 10^{-14}$
Chemistry	GPT-5.2	100	+0.42	9.25	0.93	$2.6\times 10^{-15}$ / $6.0\times 10^{-14}$
Biology	GPT-5.2	100	+0.42	9.05	0.90	$4.6\times 10^{-15}$ / $8.1\times 10^{-14}$
Computer Science	GPT-5.2	100	+0.30	7.10	0.71	$2.1\times 10^{-10}$ / $9.0\times 10^{-10}$
Sociology	GPT-5.2	100	+0.22	5.60	0.56	$6.7\times 10^{-7}$ / $2.4\times 10^{-6}$

Appendix

Table 5: Paired significance tests for Agent-RAG vs. Baseline at the question level. We report Holm–Bonferroni corrected

p

-values (

p_{\mathrm{Holm}}

) for the paired

t

-test, the Wilcoxon signed-rank test as a robustness check, and the paired effect size

d_{z}

Agentic RAG

Agentic retrieval-augmented generation lets language models plan searches, use tools, and refine evidence before producing grounded answers.

See all 246 papers

MASS-RAG: Multi-Agent Synthesis Retrieval-Augmented Generation
Apr 20, 2026Xingchen Xiao, Heyan Huang, Runheng Liu +1Retrieval-Augmented Generation Multi-Agent LLM Systems
School of Computer Science and Technology, Beijing Institute of Technology · Department of Mathematical Sciences, Tsinghua University
NeSy-RAG: Neuro-Symbolic RAG for Explainable Question Answering
Aug 6, 2026Jonas Gann, Michael GertzRetrieval-Augmented Generation Conversational QA
Data Science Group, Heidelberg University, Germany
Quantifying Retriever-Generator Alignment in RAG with Local Explanations
Jan 29, 2026Korbinian Randl, Guido Rocchietti, Aron Henriksson +3Gradient-Based Attribution Retrieval-Augmented Generation
Department of Computer and Systems Sciences, Stockholm University, Sweden · BIFOLD, Technische Universität Berlin, Germany · Department of Informatics, Athens University of Economics and Business, Greece +1

Apr 20, 2026J/K move · Enter open · S save

MASS-RAG: Multi-Agent Synthesis Retrieval-Augmented Generation

Xingchen Xiao, Heyan Huang, Runheng Liu, Jincheng Xie

School of Computer Science and Technology, Beijing Institute of Technology · Department of Mathematical Sciences, Tsinghua University

Retrieval-Augmented Generation Multi-Agent LLM Systems Multi-Agent Reasoning

Open paper AskPDF arXiv

Large language models (LLMs) are widely used in retrieval-augmented generation (RAG) to incorporate external knowledge at inference time. However, when retrieved contexts are noisy, incomplete, or heterogeneous, a single generation process often struggles to reconcile evidence effectively. We propose MASS-RAG, a multi-agent synthesis approach to retrieval-augmented generation that structures evidence processing into multiple role-specialized agents. MASS-RAG applies distinct agents for evidence summarization, evidence extraction, and reasoning over retrieved documents, and combines their outputs through a dedicated synthesis stage to produce the final answer. This design exposes multiple intermediate evidence views, allowing the model to compare and integrate complementary information before answer generation. Experiments on four benchmarks show that MASS-RAG consistently improves performance over strong RAG baselines, particularly in settings where relevant evidence is distributed across retrieved contexts.

Abstract

Agentic RAG

Explore similar work

MASS-RAG: Multi-Agent Synthesis Retrieval-Augmented Generation

NeSy-RAG: Neuro-Symbolic RAG for Explainable Question Answering

Quantifying Retriever-Generator Alignment in RAG with Local Explanations