AI
Assessing Quality of Experience in Natural Language Generation of German Text
Researchers have created a new dataset called TextQ-German to evaluate the quality of text generated...
AI
Institutional Newspapers Pipeline: Deriving billions of high quality tokens from historical newspapers
Researchers have developed a system to extract high-quality data from historical newspaper scans. Th...
AI
Structure, Association, and Decision Value: Representation-Based Difficulty Estimation for Adaptive Inference in African-Language NLI
Researchers have investigated whether internal representation statistics can provide useful difficul...
AI
Introducing the Privacy-HSD Trade-off: Hate Speech Detection, but not at the Cost of Privacy
Researchers have identified a trade-off between detecting hate speech and protecting online users' p...
AI
Grading the Graders: Verification Autonomy Levels (L0-L5) for LLM Reasoning
Researchers have proposed a new framework for evaluating the effectiveness of ve...
AI
Institutional Books - Enriched Text: A customizable multilingual open-source pipeline for denoising, deduplicating, and annotating OCR text at scale
Researchers from Harvard University have developed an open-source pipeline calle...
AI
Comment-level Topic Drift Analysis in the Reddit Corpus
Researchers have developed a method to analyze topic drift in large datasets of ...
AI
ChildSafeAds Shared Task 2026: Commercial Content in Child-Facing YouTube Videos
The ChildSafeAds Shared Task 2026 is a dataset of YouTube videos likely to reach...