giorgio/MachineSex
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Discussion: creative diversity — stylistic collapse and the evolutionary remedy

Browse source 
New Discussion block before the Outlook: homogenisation of writing style is
the same tail-drift phenomenon at a different locus (a voice = a distribution
over rare stylistic variants), so the paper's remedies — diverse grounding,
decorrelated lineages, union-preserving recombination, difference-rewarding
selection — transfer, explicitly flagged as untested here. Adds three verified
references (Guo et al. 2024; Padmakumar & He 2024; Doshi & Hauser 2024,
Sci. Adv.); first-appearance order re-verified 1..69.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BkRLcc18rwT2Lysu6PbG7v
This commit is contained in:
Giorgio Gilestro 2026-09-07 10:46:58 +01:00
parent 6325286a0e
commit cc27522270
3 changed files with 21 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

16
paper/pnas/main.md
View file

@ -409,6 +409,19 @@ system: populations of learners where every genotype, environment, and mating de
and manipulable — where the evolution of sex can be studied with interventions (unbounded parents,
offspring preview, directed mating) that no living system permits.
**Creative diversity.** Collapse is not confined to facts and skills. Homogenisation of *style* is
already measurable: models trained on model output lose lexical and syntactic diversity across
generations (67), writing produced with model assistance is individually better but collectively less
diverse than writing produced without it (68, 69), and the house styles of the large assistants are
recognisable enough that their tics serve as signatures. In this framework these are the same
phenomenon at a different locus. A voice is a distribution over rare stylistic variants, exactly the
tail that drift erases first and that blending inheritance averages into a common register. The
remedies transfer unchanged, though they are untested here: grounding on stylistically diverse human
sources, decorrelated lineages maintained as distinct voices rather than merged into one,
union-preserving recombination over blending, and selection that rewards being different as well as
being good. Whether these preserve measured stylistic diversity at scale is an open experiment that
the framework specifies.
**Outlook: the evolution of language models.** The Introduction's premise, that the model ecosystem is
an evolving population, is also a forecast about where these results matter next. Language-model
development is consolidating around exactly the operators studied here: synthetic-data flywheels
@ -526,3 +539,6 @@ publication; every figure in this paper regenerates from committed artifacts wit
64. M. Wortsman, et al., Model soups: Averaging weights of multiple fine-tuned models improves accuracy without increasing inference time. *Proc. Int. Conf. Mach. Learn.* (2022). https://doi.org/10.48550/arXiv.2203.05482.
65. Y. Cao, et al., An empirical study and theoretical explanation on task-level model-merging collapse. arXiv [Preprint] (2026). https://doi.org/10.48550/arXiv.2603.09463.
66. A. Livnat, C. Papadimitriou, Sex as an algorithm: The theory of evolution under the lens of computation. *Commun. ACM* **59**, 84–93 (2016).
67. Y. Guo, G. Shang, M. Vazirgiannis, C. Clavel, The curious decline of linguistic diversity: Training language models on synthetic text. *Findings Assoc. Comput. Linguist.: NAACL* (2024). https://doi.org/10.48550/arXiv.2311.09807.
68. V. Padmakumar, H. He, Does writing with language models reduce content diversity? *Int. Conf. Learn. Represent.* (2024). https://doi.org/10.48550/arXiv.2309.05196.
69. A. R. Doshi, O. P. Hauser, Generative AI enhances individual creativity but reduces the collective diversity of novel content. *Sci. Adv.* **10**, eadn5290 (2024).