Linguistic Loopholes in LLM Unlearning: From a 174-Language Benchmark to Coverage-Aware Unlearning
Paper recorded by Signals 4 on 2026-09-30 in cs.AI. Abstract reproduced from arXiv; link to the original below.
Published 2026-09-30 on arXiv · recorded by Signals 4 on 2026-10-01
Category: cs.AI · 人工智能 · first seen 2026-10-01
Abstract
Unlearning a fact in one language does not guarantee its removal in others as changing the query or even the requested answer language can reopen seemingly forgotten knowledge -- a cross-lingual loophole. The most straightforward solution to this challenge -- unlearning in all languages -- is neither scalable nor desirable as it amplifies damage to unrelated model capabilities. We introduce the ta
Read on arXiv →
Cite this page: Linguistic Loopholes in LLM Unlearning: From a 174-Language Benchmark to Coverage-Aware Unlearning: the #11 most recent of 480 cs.AI papers we have recorded (as of 2026-09-30). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/linguistic-loopholes-in-llm-unlearning-from-a-174-language-benchmark-to-coverage.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable:
papers.json
Get 4 AI signals a day by email — free.
Get 4 AI signals a day by email — free