Update paper title to current Sci Rep revision; remove internal version numbers

Title now matches manuscript: "Discovery and promotion of unknown sounds into operational detection targets for underwater passive acoustic monitoring under false alarm constraints". Revision label simplified to "revision in review" since this is the first author-side revision at Scientific Reports.

README.md

@@ -25,21 +25,22 @@ via Domain-Adaptive Pretraining (DAPT) on a 5,673-h global ocean
 soundscape corpus (World-DAPT).
 
 This model serves as the "ears" for underwater soundscapes described in our
-paper: **"A stethoscope for the ocean: Open-world discovery in underwater
-soundscapes"** (*Scientific Reports*, **revision 2.2** in review).
+paper: **"Discovery and promotion of unknown sounds into operational
+detection targets for underwater passive acoustic monitoring under false
+alarm constraints"** (*Scientific Reports*, **revision** in review).
 
-> **About revision 2.2** (May 2026). The original December 2026 release used
+> **About this revision** (May 2026). The original December 2026 release used
 > SimCLR/InfoNCE-based DAPT under AMP fp16, which suffered a numerical
 > instability that prevented BEATs encoder weight updates (the
 > `beats_dapt_topup_encoder.pt` weights were therefore byte-identical to
-> Microsoft's BEATs AS-2M PRETRAIN). Revision 2.2 corrects this by re-running
-> DAPT with **Masked Audio Modeling (MAM)** and a **k-means k=1024 tokeniser**
-> under bfloat16 precision on a larger 5,673-h World-DAPT corpus. The
-> superseded buggy weights have been **removed** from this repository (see
-> *Reproducibility of the original buggy state* below for how to recreate
-> them if needed).
+> Microsoft's BEATs AS-2M PRETRAIN). The current revision corrects this by
+> re-running DAPT with **Masked Audio Modeling (MAM)** and a **k-means
+> k=1024 tokeniser** under bfloat16 precision on a larger 5,673-h
+> World-DAPT corpus. The superseded buggy weights have been **removed**
+> from this repository (see *Reproducibility of the original buggy state*
+> below for how to recreate them if needed).
 
-## Model Details (revision 2.2 canonical)
+## Model Details (current canonical revision)
 
 - **Architecture:** BEATs (Audio Transformer; Microsoft)
 - **Self-supervised pretraining:** Masked Audio Modeling (MAM) with k=1024
@@ -52,7 +53,7 @@ soundscapes"** (*Scientific Reports*, **revision 2.2** in review).
 - **Input:** 16 kHz mono waveform
 - **Backbone init:** BEATs AS-2M (iter3+)
 
-## Available files (revision 2.2 canonical)
+## Available files (current canonical revision)
 
 | File | SHA-256 | Size |
 |---|---|---|
@@ -66,7 +67,7 @@ above encoder. Single-seed Event F1 = 0.483; n=10 mean ± std = 0.475 ± 0.017
 ## Reproducibility of the original (buggy) state
 
 The original December 2026 release contained two files that have been
-removed in revision 2.2:
+removed in the current revision:
 
 | Removed file | Replacement / how to recreate |
 |---|---|
@@ -83,12 +84,11 @@ prevented any weight updates.
 These weights are designed to be used with the official code repository:
 
 **GitHub Repository:** [alohajazz/openworld-soundscape-cced2-dgpu](https://github.com/alohajazz/openworld-soundscape-cced2-dgpu)
-(see branch `revision-2.2-restructure` until merged into `main`)
 
 ```python
 from huggingface_hub import hf_hub_download
 
-# Download canonical revision-2.2 weights
+# Download canonical revision weights
 encoder_path = hf_hub_download(
     repo_id="BiologgingSolutions/OceanBEATs",
     filename="beats_dapt_mam_step120000.pt",
@@ -129,11 +129,11 @@ the released weights does not grant any rights under those patents.
 If you use this model in your research, please cite our paper:
 
 ```bibtex
-@article{noda2026stethoscope,
-  title={A stethoscope for the ocean: Open-world discovery in underwater soundscapes},
-  author={Noda, Takuji and Koizumi, Takuya and others},
+@article{noda2026discovery,
+  title={Discovery and promotion of unknown sounds into operational detection targets for underwater passive acoustic monitoring under false alarm constraints},
+  author={Noda, Takuji and Koizumi, Takuya},
   journal={Scientific Reports},
-  note={Revision 2.2, in review},
+  note={Revision, in review},
   year={2026}
 }
 ```
@@ -147,8 +147,8 @@ ignored `center_sec`, returning per-file constant embeddings) was discovered
 and fixed on 2026-05-08. The fix affects only the **extraction code** in the
 GitHub repository — **encoder weights in this repository are byte-identical
 before and after the fix** (the bug occurred downstream of the encoder
-forward pass). All revision-2.2 result tables (Tables 2/3/4 and Fig 3) were
-re-computed with the corrected window-aware extractor; updated paper
+forward pass). All current-revision result tables (Tables 2/3/4 and Fig 3)
+were re-computed with the corrected window-aware extractor; updated paper
 artifacts are tracked under
 [`paper_artifacts/winaware_2026-05-09/`](https://github.com/alohajazz/openworld-soundscape-cced2-dgpu/tree/main/paper_artifacts/winaware_2026-05-09)
 and
@@ -158,9 +158,9 @@ strict 0–8 kHz in-band consistency (Nyquist of the 16-kHz BEATs input);
 species whose dominant call energy lies above 8 kHz are listed in the
 GitHub `REVISION2.md`. SHA-256 fingerprints of
 `beats_dapt_mam_step120000.pt` and `sed_head_56_fulldata_ep8.pt` are
-unchanged from the revision-2.2 release listed in the table above.
+unchanged from the current revision listed in the table above.
 
-### Revision 2.2 (May 2026)
+### Current revision (May 2026)
 - DAPT method changed from SimCLR/InfoNCE to Masked Audio Modeling (MAM)
   with k-means k=1024 tokeniser; precision changed from AMP fp16 to bfloat16
   (corrects the original numerical instability that prevented weight updates)