AI 的记忆不再需要翻译成人话
我们一直默认：AI 要记住长对话，要么存成文字摘要，要么存成图片，用时再解码。这篇直接绕开人类可读的格式，把历史对话压成一种「只有 AI 能读」的连续记忆令牌，冻结的模型通过输入接口直接读取，推理时完全不用还原成文字。做法是训练一个极小的写入器(参数只占解码器的 0.1%)，把上下文压到原来的 1/4 到 1/16。在 LongMemEval 上，7.7 倍压缩时准确率 0.504，反而超过不压缩的 0.490;文字摘要只有 0.184,OCR 方案 0.426 掉到 0.312。写入一段对话只要 43 毫秒，比文字摘要或 OCR 快一个数量级，读取也快 5-9 倍。它不是你明天能用上的东西，但它指向一个更冷的方向：AI 之间传递信息，可能根本不需要「人话」这个中间层。
📄 原文摘要(英文)
Compressed context is usually carried as human-readable text or as rendered images that must be decoded, even when its consumer is a language model. We introduce LatentPress, which writes conversational histories and long documents into a third representation: continuous memory tokens that a frozen decoder reads directly through its input-embedding interface, with no text reconstruction at inference. A small reader-matched writer compresses 4-16times while training only an adapter (4.2M-26.2M parameters, sim!0.1% of the decoder). On LongMemEval, LatentPress reaches 0.504 accuracy at 7.70times compression versus 0.490 for uncompressed evidence, outperforming text summaries (0.184) and OCR-based compression (0.426 to 0.312). On LongBench-QA, in-domain writers match or exceed raw-context reading at 4-8times compression, while 16times trails raw. Writing takes 43ms per conversation, roughly an order of magnitude faster than text summarization or OCR reconstruction, and reading is 5-9times faster than raw context or cached OCR. We validate the interface under two transfer settings, zero-shot from UltraChat to LongMemEval memory QA and from LongMemEval-derived QA to unseen LongBench document domains, establishing direct soft tokens as a practical machine-facing context interface beyond text and vision. The implementation of the experiments could be found at: https://github.com/xuyd16ai/context_softtoken_compress .