Metraipolis › archive › dc_e16a9a4e5ec4

HARPO: Hallucination-Aware Reinforcement Learning for Faithful and Creative Language Generation — arXiv:2610.03063

Source
arXiv
Original
https://arxiv.org/abs/2610.03063
License (as archived)
CC0-1.0
Archived
2026-10-05T22:51:41.691Z
SHA-256 of the archived text
11498405fa023ef8fb23d38ac38466719d103dc181eb9ad8e10ab28167b1b567
Stable URL of this page
https://metraipolis.com/documento/dc_e16a9a4e5ec4

How to cite

HARPO: Hallucination-Aware Reinforcement Learning for Faithful and Creative Language Generation — arXiv:2610.03063 — arXiv, https://arxiv.org/abs/2610.03063 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e16a9a4e5ec4 (Metraipolis archive dc_e16a9a4e5ec4, sha256 11498405fa023ef8fb23d38ac38466719d103dc181eb9ad8e10ab28167b1b567)
BibTeX
@misc{dce16a9a4e5ec4,
  title = {HARPO: Hallucination-Aware Reinforcement Learning for Faithful and Creative Language Generation — arXiv:2610.03063},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.03063},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e16a9a4e5ec4; sha256 11498405fa023ef8fb23d38ac38466719d103dc181eb9ad8e10ab28167b1b567},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.03063. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large Language Models (LLMs) are prone to generating hallucinated content, which compromises their reliability in knowledge-intensive tasks. To address this challenge without sacrificing creativity, we propose HARPO, a reinforcement learning framework designed to jointly optimize faithfulness and creativity. HARPO incorporates a Hallucination-Aware Generative Reward Model (HA-GRM), trained via verifiable feedback, to assess both faithfulness and writing quality. A Selective Activation Mechanism (SAM) activates writing rewards only for outputs judged hallucination-free by HA-GRM, while a data curriculum progressively shifts training from creative writing to hallucination-centric tasks. On RAGTruth, our Qwen3-4B-based HA-GRM achieves a response-level F1 score of 78.08%, compared with 66.37% for the supervised fine-tuning baseline. Experiments on Qwen2.5 and Qwen3 models from 1.7B to 8B parameters show improvements in both faithful generation and writing quality. On Qwen3-4B, HARPO reduces the HA-GRM-judged hallucination rate on MultiHopRAG from 3.29% to 1.02%, while increasing the Arena-Hard-v2.0 creative-writing score from 16.95% to 27.54%.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.03063 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-05T22:46:10.568Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).