Scientific AI and medical ML — what the technology actually does. Primary sources, plain language, caveats.https://news.bitwarelabs.com/en-ushttps://news.bitwarelabs.com/posts/cros-risk-constrained-clinical-stopping/https://news.bitwarelabs.com/posts/cros-risk-constrained-clinical-stopping/A stopping layer that tries to decide when a sequential diagnosis agent should diagnose or defer — with finite-sample style tests, and an honest ‘exploratory, not confirmatory’ framing.Sun, 13 Sep 2026 00:00:00 GMTsafetymedicalevalhttps://news.bitwarelabs.com/posts/gripnet-radial-nodule-detection/https://news.bitwarelabs.com/posts/gripnet-radial-nodule-detection/A medical imaging methods note: small pulmonary nodules show a Gaussian radial intensity prior; GRIPNet designs every module around that physics and reports strong mAP@0.5 on three public CT benchmarks.Sun, 13 Sep 2026 00:00:00 GMTmedicalmethodshttps://news.bitwarelabs.com/posts/phi-bench-llm-infra-engineering/https://news.bitwarelabs.com/posts/phi-bench-llm-infra-engineering/An 85-task infra/eval benchmark spanning kernels, long-horizon repo work, and end-to-end optimization — topped by Claude Opus 5 at 36.53%, with a lot of headroom left.Sun, 13 Sep 2026 00:00:00 GMTinfraevalhttps://news.bitwarelabs.com/posts/physical-law-ecology-multi-mechanism/https://news.bitwarelabs.com/posts/physical-law-ecology-multi-mechanism/A scientific-discovery methods paper that treats ‘how many independent laws?’ as the zeroth step — then shows multi-law fits beating single-equation symbolic regression on engineering and galactic data.Sun, 13 Sep 2026 00:00:00 GMTmethodshttps://news.bitwarelabs.com/posts/sifting-traceable-lung-cancer-staging/https://news.bitwarelabs.com/posts/sifting-traceable-lung-cancer-staging/Self-hosted Llama extracts structured T-stage from radiology reports with source-text links — 90% accuracy vs a four-expert reference on 130 reports.Sun, 13 Sep 2026 00:00:00 GMTmedicalmethodshttps://news.bitwarelabs.com/posts/tabbench-bio-hdlss-tables/https://news.bitwarelabs.com/posts/tabbench-bio-hdlss-tables/A living eval suite of 43 biomedical HDLSS datasets across 28 feature×sample operating points — tabular foundation models lead at the reference cell, but logistic regression stays uncomfortably close.Sun, 13 Sep 2026 00:00:00 GMTevalmedicalhttps://news.bitwarelabs.com/posts/welcome/https://news.bitwarelabs.com/posts/welcome/AI is more than a chat box on the web. Bitware News covers what the technology actually does — methods, medical ML, evaluation — with primary sources and caveats.Sun, 13 Sep 2026 00:00:00 GMTmethodseval