# ReqGenX: An Empirical Study of Atomic Decomposition, Artifact Regeneration, and Reconstruction for Legacy SRS Documents

arXiv: https://arxiv.org/abs/2607.16564 · HTML: https://mlsanity.com/paper/2607.16564/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 41 references that are in the map.
- First posted on arXiv 18 Jul 2026.

## Builds on (most cited first, 21 of 21)

- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models](https://mlsanity.com/paper/2405.01535/) (2024)
- [Large Language Models for Software Engineering: A Systematic Literature Review](https://mlsanity.com/paper/2308.10620/) (2023)
- [AlignScore: Evaluating Factual Consistency with a Unified Alignment Function](https://mlsanity.com/paper/2305.16739/) (2023)
- [Large Language Models for Software Engineering: Survey and Open Problems](https://mlsanity.com/paper/2310.03533/) (2023)
- [Dense X Retrieval: What Retrieval Granularity Should We Use?](https://mlsanity.com/paper/2312.06648/) (2023)
- [Quantifying Contamination in Evaluating Code Generation Capabilities of Language Models](https://mlsanity.com/paper/2403.04811/) (2024)
- [A Survey on Large Language Models for Software Engineering](https://mlsanity.com/paper/2312.15223/) (2023)
- [Conformity in Large Language Models](https://mlsanity.com/paper/2410.12428/) (2024)
- [Investigating ChatGPT's Potential to Assist in Requirements Elicitation Processes](https://mlsanity.com/paper/2307.07381/) (2023)
- [Using LLMs in Software Requirements Specifications: An Empirical Evaluation](https://mlsanity.com/paper/2404.17842/) (2024)
- [Assessing and Advancing Benchmarks for Evaluating Large Language Models in Software Engineering Tasks](https://mlsanity.com/paper/2505.08903/) (2025)
- [How Effective are Large Language Models in Generating Software Specifications?](https://mlsanity.com/paper/2306.03324/) (2023)
- [Lessons from the Use of Natural Language Inference (NLI) in Requirements Engineering Tasks](https://mlsanity.com/paper/2405.05135/) (2024)
- [Scalable and Domain-General Abstractive Proposition Segmentation](https://mlsanity.com/paper/2406.19803/) (2024)
- [Replication in Requirements Engineering: the NLP for RE Case](https://mlsanity.com/paper/2304.10265/) (2023)
- [TRIAD: Automated Traceability Recovery based on Biterm-enhanced Deduction of Transitive Links among Artifacts](https://mlsanity.com/paper/2312.16854/) (2023)
- [Can LLMs Generate User Stories and Assess Their Quality?](https://mlsanity.com/paper/2507.15157/) (2025)
- [Exploring the Use of LLMs for Requirements Specification in an IT Consulting Company](https://mlsanity.com/paper/2507.19113/) (2025)
- [ReqInOne: A Large Language Model-Based Agent for Software Requirements Specification Generation](https://mlsanity.com/paper/2508.09648/) (2025)

## Abstract (arXiv)

Background: Evaluating automated Software Requirements Specification (SRS) generation is challenging because few datasets provide fine-grained traceability between source requirements, intermediate elicitation artifacts, and generated specifications. Aims: We aim to study whether legacy SRS documents can be transformed into traceable synthetic pre-SRS artifacts that support fine-grained evaluation of LLM-based SRS generation. Method: We conduct an empirical study using ReqGenX, a controlled pipeline that decomposes SRS sections into source-grounded atomic statements, routes atoms to standards-inspired artifact types through multi-LLM plurality voting, and generates artifacts using constrained prompts with iterative judge-guided refinement. We evaluate ReqGenX on seven PURE SRS documents using grounding, quality, information retention, and downstream reconstruction analyses. Results: ReqGenX produces faithful and usable atoms, with median AlignScore values typically between 0.96 and 0.99 and Prometheus scores ranging from 4.34 to 4.85. Generated artifacts remain strongly grounded in their source atoms, with AlignScore values typically between 0.80--0.94 and judge pass rates near 100%; stricter Prometheus evaluation yields pass rates from 54.8% to 97.1%. In a downstream SRS reconstruction case study, artifact-backed atoms remain recoverable from generated SRSs, with SBERT means between 0.69 and 0.75 and AlignScore medians between 0.76 and 0.84. Conclusions: Traceable synthetic pre-SRS artifacts can support more fine-grained evaluation of LLM-based SRS generation, while exposing tradeoffs among faithfulness, information retention, and artifact completeness.