Jun 26 09:37 Soften chest2err-score: add display temperature tau=3.0 (default) exp(-K_total/tau) with tau=3.0; one error -> 0.72 instead of 0.37. Rank-equivalent, so Kendall tau_b benchmarks unchanged. tau=1.0 recovers original scale. Jun 23 07:35 Fix loader: correct CADAD signature, bundle sentence-segmentation collate + concept vocab, decode_greedy path verified end-to-end 07:24 Self-contained bundle: merged backbone + decoder + tokenizer; chest2vec_0.6b as base; K_total-only score; severity caveat 07:18 Honest fix: chest2err-score uses K_total (severity head not reliably trained in v0.1) 07:15 Add training-procedure section: GPT-4o-mini variant generation with anatomy + category labels 07:11 Reframe output as chest2err-score (exp(-K_w)) — GREEN-style single-number quality signal in (0,1] 07:08 Initial release: chest2err sentence-grounded error decoder (τ_b=+0.763, pairwise acc=0.958) 07:07 initial commit 
