LLMs1 min read
A retrieval-augmented generation framework extracts crash mechanisms from narratives to recommend site-specific traffic safety countermeasures. Evaluated on 312 fatal crashes, it achieved an F1-score of 0.82 with LLMs guided by engineering reasoning.
By OpenSmartRoute editorial · written through the router by writer-small
From arXiv cs.CL - “Crash Narrative-Guided Countermeasure Recommendation Using Large Language Models: A Retrieval-Augmented Generation Framework for Intersection Safety”
On average, the system recommended 3.91 countermeasures per location with 3.14 matches against actual data, closely aligning with the observed average of 3.86. This demonstrates the potential for interpretable and scalable tools in transportation safety analysis.
Source: https://arxiv.org/abs/2609.15997
Keep reading
LLMs1 min read
A new study finds that non-atomic and rigid criteria in clinical benchmarks can artificially inflate or deflate LLM evaluation scores, with rewriting disjunctive bundles shifting results by up to 15.9 percentage points.
LLMs1 min read
A study of ten frontier models using ten extrinsic audit instruments found that while all tools detected bias, their rankings were statistically indistinguishable from chance.
writer-smallollama-qwen3-5-4b/qwen3.5:4beditorial-writerDrafted automatically from the source and reviewed automatically; it may contain errors. Refer to the source for the authoritative text. How routing works
LLMs1 min read
A new study finds few-shot prompting effects vary by task and are driven by demonstration content rather than prompt length. Replacing demonstrations with random text isolates the impact of representation shifts caused by content.