Card: what is not in the repository, the run table by provenance cell, names in the before/after table

README.md

@@ -28,7 +28,7 @@ reviews (Qwen3-Embedding-0.6B, 1,024-d) and an MLP maps the pooled result to one
 
 | Listed in | Checkpoints | What they are |
 |---|---|---|
-| `rank_ckpt_plan.json` | 156 | every five-fold run a number of the paper reads, at the epoch the validation rank rule of Section 5 selects |
+| `rank_ckpt_plan.json` | 156 | the five-fold runs the paper's numbers read (all but one control, see below), each at the epoch the validation rank rule of Section 5 selects |
 | `i2bce_sel_pool.json` | 5 | I + SimCLR on the five folds (Figure 8), selected by the same rule |
 | `fixed_split_ckpt_plan.json` | 8 | the fixed-split runs of Table 3 and Section 6.2 that train with 512-sentence anchors: seven runs, one of them with an earlier checkpoint kept beside the selected one |
 | `rank_ckpt_plan.json`, field `init_ckpt` | 5 | the BYOL checkpoints the five two-stage runs started from |
@@ -39,8 +39,10 @@ the file name and the file's SHA-256; `i2bce_sel_pool.json` is the selection as
 script wrote it, with the rank score of every checkpoint of the run. The same four files are in
 `contrast_experiment/w9/eval/` of the code repository, where the evaluation script reads them.
 
-Other fixed-split towers the paper reads (the a+u @ views row of Table 3, the VICReg rows of
-Table A7, the towers of Figure 9) are not in this repository.
+Not in this repository: the other fixed-split towers the paper reads (the a+u @ views row of
+Table 3, the VICReg rows of Table A7, and the towers of Figure 9, Figures A1 and A3 and the
+neighborhood statistics of Appendix L), and the DINO-style control of Section 4.3 (Table A1, one
+fold), whose training code is not part of the code release either.
 
 ## File names
 
@@ -66,13 +68,11 @@ row reads at least one of the row's runs. That file also names the notebook that
 | Runs | What they are | Pack | Folds | Read by |
 |---|---|---|---|---|
 | `ai2ce_icetf_d512` | CE + anchor-in-I | 512 | 5 | Table 2, Section 6.2, Table A14 |
-| `ce_cetf_d512` | CE | 512 | 5 | Table 2, Table A2, Table A3, Table A10, Table A14 |
+| `ce_cetf_d512`, `i2ce_icetf_d512` | CE; I-CE, the recipe | 512 | 5 | Table 2, Table A2, Table A3, Table A4, Table A5, Table A6, Table A10, Table A14, Table A18, Table S1 |
 | `d1r2_i2ce_icetf_d512`, `d1r4_i2ce_icetf_d512`, `d1r5_i2ce_icetf_d512`, `d1r5_i3ce_icetf_d512`, `d1r6_i2ce_icetf_d512` | I-CE, V = 3; I-CE, V = 5; I-CE, V = 6; I-CE, V = 6 with λ = 3; I-CE, V = 7 | 512 | 0 to 2 | Table A5 |
-| `i2ce_icetf_d512` | I-CE, the recipe | 512 | 5 | Table 2, Table A2, Table A3, Table A10, Table A14, Table A18, Table S1 |
-| `i2ce_icetf_wllm_d512` | I-CE, LLM-rewritten documents | 512 | 5 | Table A18 |
+| `i2ce_icetf_wllm_d512`, `nodoc_i2ce_icetf_d512` | I-CE, LLM-rewritten documents; I-CE without the document view | 512 | 5 | Table A18 |
 | `i2cet05_icetf_d512`, `i2cet10_icetf_d512` | I-CE, τ = 0.05; I-CE, τ = 0.10 | 512 | 5 | Table A3 |
 | `i3ce_icetf_d512`, `i4ce_icetf_d512`, `ice_icetf_d512` | I-CE, λ = 3; I-CE, λ = 4; I-CE, λ = 1 | 512 | 0 to 2 | Table A4 |
-| `nodoc_i2ce_icetf_d512` | I-CE without the document view | 512 | 5 | Table A18, Table S1 |
 | `slot16i2cemean_icetf_d512`, `slot1i2cemean_icetf_d512`, `slot2i2cemean_icetf_d512`, `slot8i2cemean_icetf_d512` | I-CE, 16 latent queries; I-CE, 1 latent query; I-CE, 2 latent queries; I-CE, 8 latent queries | 512 | 0 to 2 | Table A6 |
 | `vfai2ce_icetf_d512` | I-CE, views-first anchors | 512 | 5 | Table S1 |
 | `ce_cetf_g1024`, `i2ce_icetf_g1024` | CE; I-CE, the recipe | 1,024 | 5 | Table A2, Table A3, Table A10, Table A14 |
@@ -83,7 +83,8 @@ row reads at least one of the row's runs. That file also names the notebook that
 | `slot8i2cemean_icetf_gpool` | I-CE, 8 latent queries | pool | 5 | Table A6 |
 | `swin168step84loop2i2ce_icetf_gpool`, `swin168step84loop2i2ce_icetf_gpool_bw` | I-CE with the windowed teacher; I-CE, two-stage: BYOL warm start, then the windowed teacher | pool | 5 | Table A12, Table S2, Table S3, Table S4 |
 
-I + SimCLR (`i2bce_icetf`, files without a pack or a suffix, five folds) backs Figure 8.
+I + SimCLR (`i2bce_icetf`, files without a pack or a suffix, five folds) backs Figure 8. The
+fold-0 run of `i2ce_icetf_g2048` also gives the game vectors of Figure 10.
 
 ## The fixed-split runs
 
@@ -136,7 +137,9 @@ python contrast_experiment/w9/eval/k10_eval.py table  scores.json
 
 `cells` takes the 161 five-fold checkpoints of `rank_ckpt_plan.json` and `i2bce_sel_pool.json`;
 `--only <regex>` keeps a subset, for example `--only "_d512$"` for the 512-sentence runs. The
-script does not score the fixed-split towers.
+script does not score the fixed-split towers. In its table the five families of the main
+comparison go by the paper's names (`I-CE`, `CE`, `SimCLR`, `VICReg`, `BYOL`) and every other
+family by its plan key without the fold.
 
 ## The 512-sentence packs and the `_d512` runs
 
@@ -156,13 +159,17 @@ Stripped hit@1 on the ten 4,096-sentence draws, five-fold mean ± std, before an
 
 | Tower | on the packs as first built | on the corrected packs (`_d512`) |
 |---|---|---|
-| I-CE | 0.669 ± 0.031 | 0.715 ± 0.010 |
-| CE | 0.638 ± 0.016 | 0.676 ± 0.013 |
-| CE + anchor-in-I | 0.663 ± 0.014 | 0.725 ± 0.015 |
-| I-CE, no document view | 0.650 ± 0.022 | 0.683 ± 0.025 |
-| I-CE, LLM-rewritten documents | 0.680 ± 0.034 | 0.711 ± 0.009 |
-| τ = 0.05 | 0.601 ± 0.029 | 0.642 ± 0.026 |
-| τ = 0.10 | 0.510 ± 0.022 (four folds) | 0.520 ± 0.015 |
+| I-CE (`i2ce_icetf`) | 0.669 ± 0.031 | 0.715 ± 0.010 |
+| CE (`ce_cetf`) | 0.638 ± 0.016 | 0.676 ± 0.013 |
+| CE + anchor-in-I (`ai2ce_icetf`) | 0.663 ± 0.014 | 0.725 ± 0.015 |
+| I-CE without the document view (`nodoc_i2ce_icetf`) | 0.650 ± 0.022 | 0.683 ± 0.025 |
+| I-CE, LLM-rewritten documents (`i2ce_icetf`, `_wllm`) | 0.680 ± 0.034 | 0.711 ± 0.009 |
+| I-CE, τ = 0.05 (`i2cet05_icetf`) | 0.601 ± 0.029 | 0.642 ± 0.026 |
+| I-CE, τ = 0.10 (`i2cet10_icetf`) | 0.510 ± 0.022 (four folds) | 0.520 ± 0.015 |
+
+The checkpoints behind the left column are the superseded files of the next section, except
+those of CE + anchor-in-I, whose first runs were not uploaded. The other `_d512` families (the
+sweeps on folds 0 to 2 and the views-first anchors) have no earlier counterpart here.
 
 ## Superseded checkpoints