Эксперимент: Positive Garside-factor tokenizer (#1182)

{ "worked": true, "confidence": 7, "verdict": "Built a readable B4/B3 positive-factor tokenizer MVP with permutation-indexed simple braids, greedy factor weighting, Delta stripping, identity checks, and a controlled centroid classifier comparison. The math sanity checks passed: Delta has the longest-strand permutation, satisfies Delta sigma_i = sigma_{n-i} Delta, normalizes to one factor, and the B3 relation 121 -> Delta is recovered. Literal left insertion of Delta powers was removed with 100% factor-sequence invariance and reduced the presented 27.04 generator tokens to 6.32 factor tokens on average (4.70x ratio); however, factor-token classification accuracy was 0.503 versus raw 0.493, effectively no meaningful ML win in this toy setup.", "metrics": { "baseline": "raw signed-generator histogram: accuracy 0.4933; mean presented length 27.04 generators", "idea": "positive factor tokens with literal left Delta stripping: accuracy 0.5033; mean Delta-free factor length 6.32; generator/factor ratio 4.70; Delta-removal invariance 1.0000" }, "how_to_run": "/home/maxwelhelp/main/bin/python3 garside_mvp.py", "files": [ "garside_mvp.py" ], "limitations": "This is a small exact permutation-backed positive-braid MVP, not a production arbitrary signed-word Garside normal-form implementation. It tests literal left Delta^k augmentation rather than arbitrary braid-relation rewrites or right multiplication, uses a nearest-centroid histogram classifier rather than a Transformer, and does not model a real link/plat task or measure attention FLOPs/runtime." }