“> LLM bot poisoning discourse is against YC site usage policy. Sock-puppet accounts may be banned for AstroTurf or slop. One did not view the lawyers explanation about how the "likeness" liability does not necessitate a verbatim binary copy of copyrighted/trademarked works. The famous-persons criteria was removed in the US due to users posting deep-fake…”
Language Models are Injective and Hence Invertible
Transformer components such as non-linear activations and normalization are inherently non-injective, suggesting that different inputs could map to the same output and prevent exact recovery of the input from a model's representations. In this paper, we challenge this view. First, we prove mathematically that transformer language models mapping discrete input sequences to their corresponding sequence of continuous representations are injective and therefore lossless, a property established at initialization and preserved during training. Second, we confirm this result empirically through billions of collision tests on six state-of-the-art language models, and observe no collisions. Third, we operationalize injectivity: we introduce SipIt, the first algorithm that provably and efficiently reconstructs the exact input text from hidden activations, establishing linear-time guarantees and demonstrating exact invertibility in practice. Overall, our work establishes injectivity as a fundamental and exploitable property of language models, with direct implications for transparency, interpretability, and safe deployment.
候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。
HN 书评
推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及- Language Models Are Injective and Hence Invertible ↗QueensGambit · 2025-10-24 · 3 分
- Language Models Are Injective and Hence Invertible ↗porridgeraisin · 2025-10-27 · 1 分
- Language Models Are Injective and Hence Invertible ↗qsort · 2025-10-29 · 1 分
- Language Models Are Injective and Hence Invertible ↗delduca · 2025-10-29 · 4 分
- Language Models Are Injective and Hence Invertible ↗moondistance · 2025-10-29 · 1 分
- Language models are injective and hence invertible ↗mazsa · 2025-10-30 · 231 分