AusTSafetyEval (T for Transport) tests whether a large language model (LLM) can reason about evidence in an accident investigation: given the factual part of an Australian Transport Safety Bureau (ATSB) final report — the narrative of the occurrence and the contextual sections, with the investigator's analysis and findings withheld — can it tell whether that material settles a specific evidentiary question, answer it if so, and say precisely what the material does and does not establish if not? Answerability datasets usually mark a question unanswerable when the text lacks the needed evidence, and score the label a model returns. However a complete accident narrative can still leave the…
Organization
AusSafetyEval
AusSafetyEval
Models in Library0
Datasets in Library1
Models on Hugging Face—
Followers1
Compare the published evidence
This profile connects the 1 model and dataset record shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
Published tasks: Question answering.
Reported licenses: cc-by-4.0. Open a record to review its specific license and conditions.
Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.