Back to paper
Full reasoning
Sep 11, 2026
cs.AI
Share this scorecard
Every score comes with the model's reasoning on the paper page.
Suggested post
"K-Bench: A Benchmark for LLM Unlearning in Agentic Deployments" scored 7.5/10 predicted impact on @kurateorg — Rigor 8.5 · Reproducible 8.5 · Novelty 8.0