Metraipolis › archive › dc_481fc46cb7e3
Gacha Decoding: Eliciting Diverse Generations Through Instruction Following — arXiv:2610.01382
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.01382
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T12:04:52.902Z
- SHA-256 of the archived text
2a7bd0241a6ba8c362825c8e9dcc73a6693890812cb5f9b94a885c00f52b4846- Stable URL of this page
https://metraipolis.com/documento/dc_481fc46cb7e3
How to cite
Gacha Decoding: Eliciting Diverse Generations Through Instruction Following — arXiv:2610.01382 — arXiv (en), https://arxiv.org/abs/2610.01382 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_481fc46cb7e3 (Metraipolis archive dc_481fc46cb7e3, sha256 2a7bd0241a6ba8c362825c8e9dcc73a6693890812cb5f9b94a885c00f52b4846)
BibTeX
@misc{dc481fc46cb7e3,
title = {Gacha Decoding: Eliciting Diverse Generations Through Instruction Following — arXiv:2610.01382},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.01382},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_481fc46cb7e3; sha256 2a7bd0241a6ba8c362825c8e9dcc73a6693890812cb5f9b94a885c00f52b4846},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.01382. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We introduce Gacha Decoding, an inference-time method for eliciting diverse language model generations that scales with model capability. Across open-ended domains (in-the-wild chat, creative writing, planning for image generation, and protein design), Gacha Decoding significantly outperforms existing generation diversity approaches at equal quality (up to 2.4x Vendi over the next-best prior approach), reaching the same number of high-quality modes with over an order of magnitude fewer samples (11.0x) and discovering novel modes that no other approach surfaces. Our key insight is to treat diversity as an instruction-following problem: rather than relying on the LM's token entropy, we combine its instruction-following capability with randomness from an external RNG tool to scalably identify and realize distinct modes of the response space. This approach of "planning with dice" enables Gacha to invert the long-observed tension between diversity and model capability. As the underlying LM becomes a better instruction follower, diversity under Gacha Decoding consistently improves--even as its token entropy and diversity under prior approaches decline. Together, our results highlight that instruction following, rather than token entropy alone, can drive generation diversity. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2610.01382 — 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 2026-10-06T10:21:27.944Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).