alejandrohdez commited on
Commit
622e1d7
·
verified ·
1 Parent(s): cafbdf4

Simplify fixed proposals and remove cached combination scores

Browse files
Files changed (2) hide show
  1. README.md +4 -2
  2. manifest.json +2 -2
README.md CHANGED
@@ -10,7 +10,7 @@ tags:
10
 
11
  These are the two core inference models for *The Alien Space of Science: Sampling Coherent but Cognitively Unavailable Research Directions*. Run them with the [public code](https://github.com/alejandrohdez00/alien-space-of-science).
12
 
13
- Training data, paper artifacts, and optional cached scores are in [alien-data](https://huggingface.co/datasets/alejandrohdez/alien-data).
14
 
15
  ## Contents
16
 
@@ -32,8 +32,10 @@ CPU and CUDA are supported. Sampling does not use an LLM API. Reconstruction is
32
 
33
  Coherence is negative mean atom-token NLL over all set permutations, excluding EOS. Availability is the median of the top ten author cosine similarities. Fusion is `(1-beta)*z(coherence) + beta*z(-availability)` over the candidate population. Each checkpoint has its own token mapping; preserve the files together.
34
 
35
- ## Training
36
 
37
  Coherence uses 136,827 training and 15,203 validation records from 15,203 papers. Availability uses 43,077,647 usable training and 4,786,404 validation subset/complement records; the training bank contains 39,791 authors from a corpus with 46,329 author atom profiles. Availability validation is by example, not by held-out author.
38
 
 
 
39
  Source: the authors' paper corpus and model training runs. File sizes and checksums are in `manifest.json`. The model weights and repository code are released under the MIT license; see `LICENSE`.
 
10
 
11
  These are the two core inference models for *The Alien Space of Science: Sampling Coherent but Cognitively Unavailable Research Directions*. Run them with the [public code](https://github.com/alejandrohdez00/alien-space-of-science).
12
 
13
+ Training data, paper artifacts, and fixed proposals are in [alien-data](https://huggingface.co/datasets/alejandrohdez/alien-data).
14
 
15
  ## Contents
16
 
 
32
 
33
  Coherence is negative mean atom-token NLL over all set permutations, excluding EOS. Availability is the median of the top ten author cosine similarities. Fusion is `(1-beta)*z(coherence) + beta*z(-availability)` over the candidate population. Each checkpoint has its own token mapping; preserve the files together.
34
 
35
+ ## Training and limitations
36
 
37
  Coherence uses 136,827 training and 15,203 validation records from 15,203 papers. Availability uses 43,077,647 usable training and 4,786,404 validation subset/complement records; the training bank contains 39,791 authors from a corpus with 46,329 author atom profiles. Availability validation is by example, not by held-out author.
38
 
39
+ The models recombine the fixed 273-atom LLM-research catalogue. Availability measures corpus-relative proximity to author repertoires; it is not proof of originality, feasibility, or future scientific value. Decoder outputs are proposals to assess. CPU and GPU calculations can differ numerically.
40
+
41
  Source: the authors' paper corpus and model training runs. File sizes and checksums are in `manifest.json`. The model weights and repository code are released under the MIT license; see `LICENSE`.
manifest.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "files": {
3
  "README.md": {
4
- "bytes": 3038,
5
- "sha256": "fcf2fb73c648b3be26304911a39b26553d83a0c6a1a2c1b72752778ed12dc8f7"
6
  },
7
  "availability_model/COMPLETE": {
8
  "bytes": 44,
 
1
  {
2
  "files": {
3
  "README.md": {
4
+ "bytes": 2881,
5
+ "sha256": "91854db2858ce45d0866bd6648e2bd17200e851c2fb7d79d55c24b0ea99505fe"
6
  },
7
  "availability_model/COMPLETE": {
8
  "bytes": 44,