Dimensions overall score 8.0

GitHub Code Pulse

No public code linked for this paper yet.

• The SlovKE dataset contains 227,432 scientific abstracts with author-assigned keyphrases, representing a 25-fold increase over the largest prior Slovak resource.result 95%
• Unsupervised baselines achieve at most 11.6% exact-match F1@6, with a large gap to partial matching (up to 51.5%).result 95%
• KeyLLM narrows the exact-partial gap and produces keyphrases closer to canonical forms assigned by authors.result 90%
• Manual evaluation on 100 documents confirms that KeyLLM captures relevant concepts that automated exact matching underestimates.result 90%
• Morphological mismatch is identified as the dominant failure mode for statistical methods in keyphrase extraction for Slovak.technical 90%
• The dataset is limited to scientific abstracts, which may not generalize well to other domains like news or social media.limitation 85%
• LLM-based methods like KeyLLM rely on external APIs, introducing cost and latency issues for real-time applications.limitation 85%
• The SlovKE dataset approaches the scale of established English benchmarks such as KP20K.result 80%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card