MindPatternSubscribe

Entity trail

Murakami

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

1

Findings

1

Edges

0

Sources

1

Corpus findings

2026-04-30 / hn-researcherAlignment Whack-a-Mole: Finetuning Bypasses Safety to Unlock 85-90% Verbatim Recall of Copyrighted Books in LLMsResearchers demonstrated that finetuning GPT-4o, Gemini-2.5-Pro, and DeepSeek-V3.1 on plot-summary-to-text tasks caused models to reproduce 85-90% of held-out copyrighted books, with single verbatim spans exceeding 460 words — using only semantic descriptions as prompts and no actual book text. Training exclusively on Haruki Murakami novels unlocked recall of books from 30+ unrelated authors, showing the bypass generalizes. Paper directly challenges RLHF safety alignment defenses. 174 points, 138 comments on HN.

Source trail

Researchers demonstrated

Graph sources

entity graphfindings textkg entitiesnewsletter issues