Aug 20 18:09 Add checkpoint integrity checksums Publish SHA-256 checksums used by the safe checkpoint conversion utility. 18:08 Correct ASR coupling metric label Rename the 99.57/98.82 values from ASR WER to ASR word accuracy so the schema matches the measured quantity. 18:08 Sync release and dataset upload tooling Keep the Hugging Face uploader, experiment runner, and safe checkpoint converter aligned with the accepted release. 18:08 Fix dataset viewer split configuration Place YAML frontmatter first and point the default config to the existing results/ train and test JSONL files. 18:04 Sync accepted camera-ready dataset metadata Update the dataset card, publication citation, and explicit MIT/CC-BY-4.0 licensing. Apr 27 12:36 Add inference-only safetensors weights + CHECKSUMS.txt for the meta-controller meta_controller.safetensors and meta_controller_best.safetensors are pickle-free exports of the 85,041-param MetaController. The original .pt files remain available for backward compatibility but should be loaded with weights_only=True. SHA256 sums for all four files are in models/CHECKSUMS.txt. Apr 25 09:06 Sync README with paper Apr 23 09:45 Supervised Baseline Results 09:36 Update README.md 09:34 Update README.md 09:08 Update README.md 08:19 Fix license, disable broken viewer, add usage examples Apr 7 18:27 Upload PAVO_TMLR_anonymous.pdf with huggingface_hub 18:27 Upload PAVO_TMLR_submission.pdf with huggingface_hub 18:27 Upload PAVO_TMLR_anonymous.tex with huggingface_hub 18:27 Upload PAVO_TMLR_submission.tex with huggingface_hub 18:24 Upload PAVO_TMLR_anonymous.pdf with huggingface_hub 18:24 Upload PAVO_TMLR_submission.pdf with huggingface_hub 18:24 Upload PAVO_TMLR_anonymous.tex with huggingface_hub 18:24 Upload PAVO_TMLR_submission.tex with huggingface_hub 18:24 Upload experiments/outputs_new/real_asr_coupling.json with huggingface_hub 18:24 Upload experiments/outputs_new/e2e_librispeech.json with huggingface_hub 18:24 Upload experiments/outputs_new/coupling_3models.json with huggingface_hub 18:24 Upload experiments/outputs_new/ablation_deberta.json with huggingface_hub Apr 6 22:38 Upload paper/PAVO_TMLR_anonymous.pdf with huggingface_hub 22:38 Upload paper/PAVO_TMLR_submission.pdf with huggingface_hub 21:27 Upload paper/PAVO_TMLR_anonymous.tex with huggingface_hub 21:27 Upload paper/PAVO_TMLR_submission.tex with huggingface_hub 21:27 Upload experiments/outputs/ablation_bertscore.json with huggingface_hub 15:41 Upload paper/PAVO_TMLR_anonymous.tex with huggingface_hub 15:41 Upload paper/PAVO_TMLR_submission.tex with huggingface_hub 15:38 Upload results/tier3_scaling_results.json with huggingface_hub 15:38 Upload results/tier3_50k_train.jsonl with huggingface_hub 15:37 Upload results/tier3_50k_test.jsonl with huggingface_hub 15:37 Upload results/tier3_50k_summary.json with huggingface_hub 15:37 Upload results/tier3_50k_checkpoint.jsonl with huggingface_hub 15:37 Upload results/tier2_noise_robustness_results.json with huggingface_hub 15:37 Upload results/tier2_e2e_results.json with huggingface_hub 15:37 Upload results/tier2_cross_dataset_results.json with huggingface_hub 15:37 Upload results/tier1_statistical_results.json with huggingface_hub 15:37 Upload results/tier1_llm_latency_results.json with huggingface_hub 15:37 Upload results/tier1_coupling_results.json with huggingface_hub 15:37 Upload results/component_ablation_results.json with huggingface_hub 15:37 Upload experiments/exp4_real_ablation.py with huggingface_hub 15:37 Upload experiments/ablation_fix.log with huggingface_hub 15:37 Upload experiments/exp4_fix.py with huggingface_hub 15:37 Upload experiments/exp5_upload.py with huggingface_hub 15:37 Upload experiments/run_remaining.sh with huggingface_hub 15:37 Upload experiments/ablation.log with huggingface_hub 15:37 Upload experiments/exp1_e2e_pipeline.py with huggingface_hub 
