2 days ago
-
Fact-check corrections: refresh Hugging Face cards and results dataset
3 days ago
-
Training paused: size-1/seed1 evaluated; diagrams and Hugging Face card videos; remaining runs postponed
-
Run size-1/seed1 trained
-
Register run size-1/seed1 (2-single-seed) before training
-
Run specialist-finetune-campaign-05/seed1 evaluated
-
Run specialist-finetune-campaign-05/seed1 trained
-
Register run specialist-finetune-campaign-05/seed1 (2-single-seed) before training
-
Run specialist-finetune-campaign-04/seed1 evaluated
-
Run specialist-finetune-campaign-04/seed1 trained
-
Register run specialist-finetune-campaign-04/seed1 (2-single-seed) before training
-
Run specialist-finetune-campaign-03/seed1 evaluated
-
Run specialist-finetune-campaign-03/seed1 trained
-
Register run specialist-finetune-campaign-03/seed1 (2-single-seed) before training
-
Run specialist-finetune-campaign-02/seed1 evaluated
-
Run specialist-finetune-campaign-02/seed1 trained
-
Register run specialist-finetune-campaign-02/seed1 (2-single-seed) before training
-
Run specialist-finetune-campaign-01/seed1 evaluated
-
Run specialist-finetune-campaign-01/seed1 trained
-
Register run specialist-finetune-campaign-01/seed1 (2-single-seed) before training
-
Run specialist-campaign-05/seed1 evaluated
-
Run specialist-campaign-05/seed1 trained
-
Register run specialist-campaign-05/seed1 (2-single-seed) before training
-
Run specialist-campaign-04/seed1 evaluated
-
Run specialist-campaign-04/seed1 trained
-
Register run specialist-campaign-04/seed1 (2-single-seed) before training
-
Run specialist-campaign-03/seed1 evaluated
-
Run specialist-campaign-03/seed1 trained
-
Register run specialist-campaign-03/seed1 (2-single-seed) before training
-
Run specialist-campaign-02/seed1 evaluated
-
Run specialist-campaign-02/seed1 trained
-
Register run specialist-campaign-02/seed1 (2-single-seed) before training
-
Run specialist-campaign-01/seed1 evaluated
-
Run specialist-campaign-01/seed1 trained
-
Register run specialist-campaign-01/seed1 (2-single-seed) before training
-
Run pqn/seed1 evaluated
-
Run pqn/seed1 trained
-
Register run pqn/seed1 (2-single-seed) before training
-
Run blind/seed1 evaluated
-
Run blind/seed1 trained
-
Register run blind/seed1 (2-single-seed) before training
-
Run generalist/seed1 evaluated
-
Run generalist/seed1 trained
-
Register run generalist/seed1 (2-single-seed) before training
-
Pilot generalist and training-free baselines scored on test, out-of-range and unfiltered test tracks (after the recipe freeze)
-
Re-score both first models and the training-free baselines with order-independent evaluation; regenerate reports, paper and README
-
Run generalist/seed0 evaluated
-
Run generalist/seed0 trained
-
Register run generalist/seed0 (1-first-models) before training
-
Re-score specialist-campaign-01/seed0 with the corrected energy flag; add the certified 50,000-seed training-set summary
-
Run specialist-campaign-01/seed0 evaluated