{"slug": "core-weakly-supervised-coarse-to-fine-risk-evidence-learning-in-driving-videos", "title": "CoRE: Weakly Supervised Coarse-to-Fine Risk Evidence Learning in Driving Videos", "summary": "Researchers introduced CoRE, a weakly supervised coarse-to-fine framework that learns fine-grained prediction support from coarse video supervision, achieving strong temporal localization on the DoTA benchmark and competitive performance on UCF-Crime without requiring fine-grained annotations. The framework trains a video-level predictor, then uses structured interventions to generate prediction-effect targets that are distilled into a student model for direct inference.", "body_md": "arXiv:2608.25344v1 Announce Type: new\nAbstract: Perceived risk in driving evolves over time and may be supported by specific scene entities, yet supervision is typically limited to coarse video-level judgments. Learning \\emph{when} supporting evidence emerges and \\emph{which entities} support a risk predictor would ordinarily require costly temporal- and entity-level annotations. We introduce \\textbf{CoRE}, a weakly supervised coarse-to-fine framework that learns fine-grained prediction support from coarse video supervision. CoRE first trains a video-level predictor and then freezes it. Structured interventions over candidate temporal regions or entity tracks measure how each candidate changes the coarse prediction, producing graded prediction-effect targets. These targets are distilled into a student that directly predicts temporal and entity support from the original video, without requiring interventions at inference. We evaluate this learning principle across three complementary settings: RISEE tests perceived-risk support from subjective clip-level judgments without temporal or entity-level risk annotations; DoTA provides independent temporal event annotations for evaluating weakly supervised traffic-anomaly localization; and UCF-Crime tests whether the same coarse-to-fine mechanism extends to a standard non-driving anomaly-detection benchmark. Across these settings, CoRE learns informative fine-grained support from coarse supervision, with strong temporal localization on DoTA and competitive performance on UCF-Crime. These results show that coarse video predictions can provide useful supervision for recovering the fine-grained evidence supporting them, without requiring corresponding fine-grained labels.", "url": "https://wpnews.pro/news/core-weakly-supervised-coarse-to-fine-risk-evidence-learning-in-driving-videos", "canonical_source": "https://arxiv.org/abs/2608.25344", "published_at": "2026-08-27 04:00:00+00:00", "updated_at": "2026-08-27 04:21:58.750794+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "computer-vision"], "entities": ["CoRE", "RISEE", "DoTA", "UCF-Crime"], "alternates": {"html": "https://wpnews.pro/news/core-weakly-supervised-coarse-to-fine-risk-evidence-learning-in-driving-videos", "markdown": "https://wpnews.pro/news/core-weakly-supervised-coarse-to-fine-risk-evidence-learning-in-driving-videos.md", "text": "https://wpnews.pro/news/core-weakly-supervised-coarse-to-fine-risk-evidence-learning-in-driving-videos.txt", "jsonld": "https://wpnews.pro/news/core-weakly-supervised-coarse-to-fine-risk-evidence-learning-in-driving-videos.jsonld"}}