Response ratio predicts instruction-tuning utility better than common proxies
Under a fixed finetuning budget, answer-dense examples beat common quality proxies on perplexity and LLM-as-judge scores.
Stories per day · 30 days
6 stories, newest first · one card per story
Stories tagged #data-quality
CISA Charts New "Quality Era" for Global CVE Program
CISA outlined a quality framework for the CVE program as 2026 disclosures approach a projected 96,000.
Data Overtakes Skills as Top Threat Hunting Challenge, SANS Study Finds
SANS 2026 Threat Hunting Survey: data quality/quantity (50%) overtook skills (45%) as hunters' top barrier, while ransomware remains the most-found threat.
1
CISA outlines improvement plan for CVE program
CISA published a white paper to improve CVE quality as disclosure volume surges.
A hollowed out data layer is making CISOs fly blind into AI attacks
Opinion piece argues two years of SIEM ingest cost-cutting hollowed out data foundations, leaving SOC visibility blind spots as AI-driven attacks accelerate.
1
18 ways to check whether data can be trusted for AI
ETSI published TR 104 180 defining 18 data quality metrics, including fairness and privacy, to assess whether datasets are fit for AI.
2