Findings, methods and limits explained with the evidence in view.
Edition for 2026-09-16 (America/Denver). Every item is dated from its source, and anything that has aged out of the news window stays in the archive instead.
Lead finding
A new arXiv preprint proposes a 'converge-then-diversify' approach for multi-objective Bayesian optimisation, but the available evidence is only the paper's abstract.
Lead finding
A preprint describes GAVEL, an LLM judge protocol that compares two clinical timelines with the case report and returns a discrepancy type, verdict, and report passage for each…
Lead finding
A new arXiv preprint describes TimeThink, a synthetic-data framework for training timeseries language models to reason about patterns like trend and seasonality. The abstract claims it outperforms baselines,…
Lead finding
A newly posted abstract describes ZGCM-1, a 7B open model for math and agentic search. The claims come from the authors' abstract, not from independent replication.
Lead finding
A preprint on arXiv reports that a multi-agent framework using large language models can manage long-term physical tasks in a zero-shot manner. The abstract claims the agents matched…
All editions · Archive · RSS