Benchmark RadarAI BENCHMARK PROFILE
ForgetBench
Proposes a benchmark for evaluating forgetting dynamics in language models under continual knowledge editing, with concept-based and scenario-based QA, but no public artifacts are available.
- Released
- 2026-07-29
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Addresses knowledge retention over time, an important aspect for model updates.
Motivation
Large language models (LLMs) have demonstrated strong capabilities in knowledge acquisition and reasoning, yet their ability to retain previously acquired knowledge under repeated updates remains insufficiently understood.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.