diff --git a/paper/results-summary.md b/paper/results-summary.md index 07e9782..bf42be8 100644 --- a/paper/results-summary.md +++ b/paper/results-summary.md @@ -13,8 +13,9 @@ sexual society**: reality-checking (grounding) plus **recombination across many parents** (sexual reproduction, not teacher→pupil copying) plus **quality-diversity** selection. The payoff is not merely arrested collapse but a population whose **offspring exceed their parents** and whose general capability **climbs** while specialties are re-earned — and removing -any one operator breaks it. **122 tests pass**; three of the core predictions are exact closed -forms, so the headline curves sit on analytic targets rather than eyeballing. +any one operator breaks it, and a first **real-LLM prototype** (up to 7B on HPC) confirms the +recombination claims in trained weights. **131 tests pass**; three of the core predictions are exact +closed forms, so the headline curves sit on analytic targets rather than eyeballing. --- @@ -145,7 +146,47 @@ Fisher–Muller effect): a rugged landscape you need diversity to explore basins, sex to recombine them, and grounding to select on reality. -### 5. Positioning — what is prior art, what is ours +### 5. The claims tested in real LLM weights (Layer 2 prototype) + +The analytic operators above make three claims that are testable in *real language-model weights*: +"merge, don't average" (E4), the Fisher–Muller generalist-from-specialists (E8), and directed sex +(E10). We tested them with a small, reproducible pipeline — LoRA **specialists** fine-tuned on three +*disjoint*, procedurally-generated task families (`lists`, `strings`, `arith`) with an **exact-match +verifier** as reality's "no" — recombined by various operators and scored on held-out tasks. Runs are +at two scales: **Qwen2.5-0.5B** locally and **Qwen2.5-7B** on Imperial College's CX3 HPC (one L40S). +This is a prototype (three families, one seed), so read it as **signs, not magnitudes**; the analytic +layer carries the quantities. + +- **Recombining specialists yields a generalist that exceeds every parent (E8).** At 7B, a weight-space + **merge of the three specialists reaches 0.87 overall vs 0.77 for the best single specialist**, and + beats every specialist on every family — the Fisher–Muller signature, in real weights. At 0.5B the + same sign is present but marginal, because a weak base *dilutes* (see below). +- **Union beats averaging — when there is headroom (E4).** Two ways to recombine: **fusion** (average + the LoRA deltas — a "model soup") versus **union** (keep each specialist intact and *route* each + input to the right one — a Mixture-of-Experts). On tasks with headroom, **routing beats fusion** + because averaging dilutes a specialist's contribution: at 0.5B routing 0.74 > soup 0.64; on *hard* + 7B tasks routing 0.50 > soup 0.40, with fusion diluting the fragile string-cipher specialist so badly + (0.67→0.30) that the soup even loses to the best single specialist. This is E8's `max > mean` in real + weights. +- **Directed sex helps — when there is headroom (E10).** Breeding a population of recombinant offspring + (specialists merged at many weightings), scoring each on a held-out split with the verifier, and + keeping the fittest **beats the single a-priori soup** where the soup is suboptimal: 0.69 > 0.64 at + 0.5B, and 0.49 > 0.39 on hard 7B tasks — recovering most of routing's benefit from one deployable + model. This is offspring selection (which biology cannot preview) realised in weight space. +- **The one law that unifies them: headroom, not scale.** Our first 7B runs on *easy* tasks showed the + opposite — fusion beating union, and directed selection tied with the soup. That was a **saturation + artefact**: the easy families saturated 7B at 1.00, leaving no room to lose to dilution, so the naive + soup was already optimal. On *hard* (unsaturated) 7B tasks the original ordering returns. **The + operative variable is headroom**: union-preserving merging and offspring selection pay off whenever + there is room to lose to dilution — a weak base *or* hard tasks — and only the degenerate corner of + easy tasks on a strong base makes naive averaging look sufficient. This both confirms E4/E8/E10 in + real weights and delimits exactly when the refinements matter. + +Honest scope: three lexically-distinct families (so the router is trivially accurate — an ambiguous- +skill benchmark is the interesting next stress test), one seed, small LoRA. The full *grounded* +society on LLMs (collapse under dry self-training, the dynamic society) is the HPC-scale next step. + +### 6. Positioning — what is prior art, what is ours An independent 2026 paper (Riis, *Drift and selection in LLM text ecosystems*) rigorously formalises **collapse = Wright–Fisher drift** (martingale of minority mass, rare-first extinction, de Bruijn @@ -166,8 +207,11 @@ engineered cure and its integration. 1. **Never inherit dry, and ground where it matters** — a little reality (`g*≈5%`) protects most diversity, but it protects only what it overlaps, and it can't hold the deep tail. -2. **Merge, don't average** — union-preserving model-merging realises the multi-teacher benefit; - averaging cancels it. +2. **Merge, don't average — where there is headroom** — union-preserving model-merging (routing / + max-merge) realises the multi-teacher benefit; averaging dilutes it. Confirmed in real LLM weights, + with a sharp caveat: dilution only bites when the task leaves room for it — on easy tasks a strong + base's naive soup already composes to the ceiling, so averaging looks fine. The refinement matters + exactly in proportion to how far the naive soup is from optimal. 3. **Sex, with no parent limit** — recombining decorrelated specialists yields offspring that exceed any parent; more, complementary parents climb higher. 4. **Match recombination to entanglement** — merge freely for complementary skills; sparingly for @@ -230,6 +274,13 @@ that lets a society climb instead of collapse. consensus (the exact analogue of training on the internet's AI-generated crowd); remove **sex** and it gets stuck; remove **diversity** and it converges too fast to a worse answer. Each failure is distinct; only the full society wins. +8. **It shows up in real language models — with one clean caveat.** We merged LoRA-specialised + Qwen models (0.5B locally, 7B on a university GPU cluster): the recombined model beats every + specialist (Fisher–Muller, for real), and *routing* / *offspring-selection* beat naive averaging — + but only when the tasks are hard enough to leave room. On easy tasks a strong model's plain average + is already at ceiling, so the fancier operators don't help. The lesson is precise: **these + recombination tricks matter exactly in proportion to how far the naive average is from the best you + could do** — which is a genuinely useful thing to know before you spend compute on them. ### Why it is novel and why it matters @@ -240,9 +291,10 @@ that lets a society climb instead of collapse. - **The sexual-reproduction frame is, we believe, genuinely new for AI** — model merging reframed as meiotic recombination, with a rigorous account of when it helps (complementary skills), when it hurts (entangled skills), and how to make it reliably win (directed sex, unbounded parents). -- **It is validated, not vibes.** Three core predictions are exact closed forms; the neural and - image results confirm the *signs* in real trained weights; 122 automated tests pass; the whole - study is laptop-reproducible from a seed. +- **It is validated, not vibes.** Three core predictions are exact closed forms; the neural, image, + and **real-LLM** results confirm the *signs* in real trained weights (up to 7B on HPC); 131 + automated tests pass; the whole study is laptop-reproducible from a seed (the LLM tier statistically + reproducible on one GPU). - **It gives concrete design rules** for anyone building self-improving or model-merging systems: ground where it matters, merge-don't-average, match recombination to skill-entanglement, select offspring, keep diversity, and gate irreversible consolidation on health. diff --git a/paper/the-lamarckian-society-v5.md b/paper/the-lamarckian-society-v5.md index 0d3fd99..2a75277 100644 --- a/paper/the-lamarckian-society-v5.md +++ b/paper/the-lamarckian-society-v5.md @@ -35,9 +35,11 @@ you skipped a definition: it knew when it learns something new. We have tried to keep the big picture legible on every page, and to be candid about what is argument -and what is evidence. The evidence is from **deliberately small models** — mathematics, small neural -networks, image generators, and evolutionary simulations. None of it is a large language model yet. -We will say so repeatedly, because the gap matters. +and what is evidence. The evidence is mostly from **deliberately small models** — mathematics, small +neural networks, image generators, and evolutionary simulations. A first bridge to real language +models exists — a prototype that recombines LoRA-specialised Qwen models up to 7B on a GPU cluster, +which confirms the recombination signs (below) — but the *full grounded society* has not yet been +built on a large language model. We will say so repeatedly, because the gap matters. --- @@ -67,13 +69,16 @@ AI has one advantage biology lacks: its "sex" has **no two-parent limit**, its m for complementarity, and its offspring can be **screened before they are kept**. We call this *directed sex*, and it turns recombination from a gamble into a reliable engine. -We support the argument with a set of **minimal models** (not language models): a mathematically exact -account of collapse and its cure; reproductions of the same effects in small trained neural networks -and in a generator of handwritten digits; and evolutionary simulations in which a population of models -climbs a "fitness landscape" that stands in for reality. In these, every claim above either holds or -fails visibly, and removing any single ingredient breaks the system in a distinct way. The scope is -honest: these are existence proofs and design rules, and the eventual test is to build the same thing -out of real language models. We also note where our diagnosis is no longer novel — the reading of +We support the argument with a set of **minimal models**: a mathematically exact account of collapse +and its cure; reproductions of the same effects in small trained neural networks and in a generator of +handwritten digits; and evolutionary simulations in which a population of models climbs a "fitness +landscape" that stands in for reality. In these, every claim above either holds or fails visibly, and +removing any single ingredient breaks the system in a distinct way. A first **language-model +prototype** then confirms the recombination claims in real weights — merging LoRA-specialised Qwen +models (up to 7B on a GPU cluster) produces a generalist that exceeds every parent, and shows the +sharp condition under which the "merge, don't average" refinement matters (below). The scope is +honest: these are existence proofs and design rules, and the eventual test is to build the *whole +grounded society* out of real language models. We also note where our diagnosis is no longer novel — the reading of collapse as genetic drift has since been derived independently — and locate our contribution in the **cure** rather than the diagnosis. @@ -207,6 +212,19 @@ knowledge; specialisation is re-earned each generation," and it is why the refra teacher→pupil to *sexual reproduction* is not cosmetic: **copying can only recover a ceiling; recombination can exceed it.** +This is no longer only a simulation. In a first language-model prototype — LoRA specialists on +disjoint task families, recombined and judged by an exact verifier — a merge of three specialist Qwen +models (7B, on a GPU cluster) **beats every single specialist**, overall and on every family: the +Fisher–Muller effect, in real weights. The same prototype pins down *when* the finer "inherit the +union, don't average" rule actually bites. Keeping each parent whole and **routing** each input to the +right one beats the tail-thinning average — but only when the task is hard enough to leave room to +lose: on easy tasks a strong model's plain average is already at the ceiling, so the crude soup is +fine, whereas on hard tasks the average dilutes a hard-won specialist so badly it falls below even the +best single parent, and routing wins by a wide margin. The rule is therefore precise: **the union +beats the average in exact proportion to how far the average is from the best attainable** — a caveat +that sharpens rather than weakens the claim, and that a practitioner needs before spending compute on +the fancier operator. + Two caveats keep this honest, and both are results, not hand-waving. *Sex can backfire.* When the parents' skills are not cleanly separable but **entangled** — when the @@ -223,8 +241,11 @@ and cannot inspect an offspring before it is born. An AI has none of those limit **generate many candidate offspring and keep only the fittest**, screening them against reality before committing. We call this **directed sex**, and in our simulations it converts the outbreeding-depression catastrophe into a reliable gain: where blind recombination collapses on entangled problems, directed -recombination matches or beats the best parent every time. This is a genuine advantage of engineered -reproduction over the biological kind, and we think it is one of the more useful ideas in the paper. +recombination matches or beats the best parent every time. The language-model prototype shows the same +sign where it can: breeding many recombined Qwen offspring and keeping the one the verifier scores +highest beats the single averaged soup on hard tasks (and, unsurprisingly, does nothing extra on easy +tasks the soup already solves). This is a genuine advantage of engineered reproduction over the +biological kind, and we think it is one of the more useful ideas in the paper. So the picture of §5 is: single-teacher copying is asexual and collapses (Muller's ratchet = model collapse); the cure is to *ground* every birth in reality and to reproduce *sexually*, recombining @@ -443,9 +464,10 @@ refute, or you evolve toward what is merely convincing.** ## 13. What we built, what we found, and what is still open The previous drafts of this paper promised a "companion paper" that *would* make this concrete. That -work now exists — as a set of **minimal, laptop-reproducible models**, not language models — and it is -worth stating plainly what it does and does not show. (A separate results document gives the numbers; -here is the shape.) +work now exists — mostly as a set of **minimal, laptop-reproducible models**, with a first bridge to +**real language models** (a LoRA-merge prototype, up to 7B on a GPU cluster) — and it is worth stating +plainly what it does and does not show. (A separate results document gives the numbers; here is the +shape.) **What we built and found.** @@ -471,6 +493,15 @@ here is the shape.) produces below-parent offspring (outbreeding depression) — and *directed* recombination (choose mates, screen offspring, unbounded parents) reliably fixes it. This is the concrete evidence for the paper's central reframing. +- *The recombination claims, in real language models — with a sharp condition.* Merging LoRA-specialised + Qwen models (up to 7B on a GPU cluster) produces a generalist that beats every specialist parent + (Fisher–Muller, for real); and keeping parents intact and *routing*, or *breeding and screening* + offspring, beats the naive average — but *only when the task leaves headroom*. On easy tasks a strong + model's plain average is already at the ceiling and the refinements add nothing; on hard tasks the + average dilutes a specialist below even the best single parent, and the union-preserving operators win + clearly. The practical rule is exact: these tricks pay off in proportion to how far the naive average + is from the best attainable. This is a prototype (three task families, one seed), so we read it as + signs, not magnitudes; the *whole grounded society* on a language model remains the open step. - *The whole society, and why every part is needed.* In a population evolving on a "reality" landscape, the full system — grounding + sexual recombination + preserved diversity — climbs to the top while keeping its specialists. Remove *grounding* and it collapses into a confident, wrong consensus (a @@ -499,10 +530,11 @@ the **institutions** that let contemporaries correct one another before error is we do not solve; and the **calibration** of everything the results left as knobs — how many parents, how complementary, at what ratio of inherited-to-real data, and how healthy a lineage must be before its knowledge is safe to make irreversibly innate. These are, at least, *measurable* — which is the -difference between an open problem and a hole. And the largest gap of all: every result here is from a -minimal model. The real test is to build the same grounded, sexually-reproducing, diversity-preserving -society out of actual open-weight language models, and see whether the three signs survive contact with -a system too big to write down. Theory and toys here; the real thing, next. +difference between an open problem and a hole. And the largest gap of all: the *recombination* claims +now hold in real language models, but the *society* — the grounded, diversity-preserving, continually +reproducing loop — does not yet. The real test is to build that whole system out of actual open-weight +language models, and see whether all the signs survive contact with a system too big to write down. +The operators, checked; the living society, next. ---