@inproceedings{osuji-etal-2026-llm,
    title = "{LLM} Multi-Agent Systems for Long Triple Set Data-to-Text Generation",
    author = "Osuji, Chinonso Cynthia  and
      Mille, Simon  and
      Andrade, Mark  and
      Adkins, Jane  and
      O'Connell, Ornait  and
      Dhonnchadha, Elaine U{\{}{\textbackslash}'{\textbackslash}i{\}}  and
      Heffernan, Bl{\textbackslash}'aith{\{}{\textbackslash}'{\textbackslash}i{\}}n  and
      tSaoir, F{\{}{\textbackslash}'{\textbackslash}i{\}}rinne Nic an  and
      Belz, Anya  and
      Ferreira, Thiago Castro  and
      Davis, Brian",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.1712/",
    pages = "34261--34275",
    ISBN = "979-8-89176-395-1",
    abstract = "Generating coherent, semantically accurate text from large structured inputs remains a persistent challenge in data-to-text generation, as single-step LLM mappings from data-to-text limit control over discourse structuring and amplify hallucinations and omissions as input size grows. We introduce a new dataset of extended DBpedia triple sets (up to 199 triples per input), and a modular multi-agent framework: specialised LLM agents handle content ordering, text structuring, and surface realisation under the supervision of an orchestrator and guardrail control loop. The system generates multi-paragraph outputs in English and Irish (low-resource). We compare a three-worker multi-agent configuration against a single-worker multi-task variant and a strong end-to-end baseline. Quality is assessed via human evaluation and LLM-as-a-judge (with truncation-based sanity checks). Results show slightly superior coherence for the multi-agent approach in both languages, with statistically significant inter-rater correlation over all criteria for English and no statistically significant correlation for Irish. Human-LLM alignment is very weak overall, thus exposing key limits in scalable NLG evaluation."
}