From 9e83aa7034bcfe18b2b6e08ec8708de258911d5d Mon Sep 17 00:00:00 2001 From: Giorgio Gilestro Date: Mon, 7 Sep 2026 13:49:51 +0100 Subject: [PATCH] =?UTF-8?q?llm=5Fsociety=20pilot=20v2:=20the=20contrast=20?= =?UTF-8?q?is=20alive=20=E2=80=94=20grounded=20society=20climbs,=20ungroun?= =?UTF-8?q?ded=20decays?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit full: best 0.583 -> 0.617 and holds (a bred offspring beats every founder); no_grounding: monotone decay to 0.533 despite identical elitism (conformist selection keeps the wrong elite), with the confident-but-wrong gap growing to +0.24. One seed - direction only; the 4-arm x 3-seed CX3 campaign is gated on GG's review. Co-Authored-By: Claude Fable 5 Claude-Session: https://claude.ai/code/session_01BkRLcc18rwT2Lysu6PbG7v --- tasks/workorder-llm-society.md | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/tasks/workorder-llm-society.md b/tasks/workorder-llm-society.md index bc62356..f8efc07 100644 --- a/tasks/workorder-llm-society.md +++ b/tasks/workorder-llm-society.md @@ -74,10 +74,13 @@ disagreement), consensus accuracy, conformity−truth gap. E11's three panels re first distillation generation and plateau at the same mutation–selection floor (best ≈0.40, mean ≈0.35, consensus 0.63→0.4 in both). Diagnosis: inheritance too lossy; the ~25%/generation distillation tax swamps the selection differential. v1 archived at `results/llm_society_pilot1`. -- [ ] Pilot v2 (n_inherit=600, ep3, founders 600×3, elitism=1 all arms, n_parents=3, n_conf=90) — - raises inheritance fidelity so selection has something to act on; elites = overlapping - generations, applied identically in every arm (documented honestly; reproduction here is lossy - distillation, unlike E11's faithful genotype copy) +- [x] Pilot v2 (n_inherit=600, ep3, founders 600×3, elitism=1 all arms, n_parents=3, n_conf=90): + **the contrast is alive.** `full` climbs 0.583 → 0.617 best true fitness and holds (a bred + offspring beat every founder at gen 3); `no_grounding` decays monotonically 0.583 → 0.533 + despite identical elitism — conformist selection keeps the wrong elite. Self-consumption + signature present: in `no_grounding` the population agrees with itself more (conformity 0.60 → + 0.73) while the consensus gets less right (0.63 → 0.49); the confident-but-wrong gap reaches + +0.24 vs +0.16 in `full`. One seed, N=6: direction only. `results/llm_society`. - [ ] GG gate: review pilot curves before the campaign - [ ] `hpc/llm_society.pbs` array job (arm × seed) → campaign - [ ] Figure `paper/pnas/make_figs.py` panel(s); fold into main.md (replaces the "open" cell)