K-Bench: LLM Unlearning in Agentic Deployments - a UTSCybeR Collection

UTSCybeR 's Collections

Diversity Combining for Multi-Path LLM Reasoning

K-Bench: LLM Unlearning in Agentic Deployments

Jev: Semantic Decision Engines for Network Control

K-Bench: LLM Unlearning in Agentic Deployments

updated Oct 7

Benchmark, reference assets and public leaderboard for LLM unlearning when the model runs inside an agent.

K-Bench: A Benchmark for LLM Unlearning in Agentic Deployments

Paper • 2609.12808 • Published Sep 11
UTSCybeR/kbench-assets

Updated Sep 14 • 216
Running

K-Bench Leaderboard

🔑

Explore LLM unlearning benchmark results across memory substrates