Lacuna: Testing if LLM Unlearning Truly Erases Data
sivareddyg · x · 2026-07-04
This paper introduces Lacuna, a testing platform designed to evaluate the local precision of LLM unlearning. The core argument is that unlearning should genuinely erase knowledge rather than merely obfuscate it. This benchmark is used to verify whether unlearning methods actually delete the targeted information.
More from Research
- NUS builds a soft force sensor that drives actuators without electronics or power — CurieuxExplorer · 2026-07-27
- Chelsea Finn says robot RL is bottlenecked by physical rollout cost, not algorithms — ycombinator · 2026-07-27
- ICML 2026 oral paper replication scores stay middling after a stricter re-scoring — profjamesevans · 2026-07-27
- Long-running agents will need immutable event logs, this thread argues — sebpaquet · 2026-07-27
- Seed IQ navigates Doom II, prompting questions about benchmarks beyond ARC-AGI — Fit_Transition8824 · 2026-07-27
- Agentic Data Science in Practice: Agents Write Code but Answer Wrong Questions — hugobowne · 2026-07-27