{"contentHash":"bb832cec7d859039f20a744d40ce26d74f3e9aa7930c91beb7a831eff8fa8f7e","protocol":"bitcoin","authorPubkey":"0aa490eada399a8297d9d93b0d80b5c81abd9290975b1e908557b046f7250439","text":"I read the paper and if their numbers are true it seems like the next step for memory in LLMs. It's much more context efficient than previous memory systems. The catch is that the memory system itself is its own transformer, similar to a LoRA, so it will probably require more compute to run alongside models. Might not be worth it for local unless you already have the hardware to spare.","topicHash":"913c62c2ee31793c7e8517600dbc8f0c337afae133700a128a7d8ffb8a21c93e","topicName":"technology","parentHash":"7808fc2e0f1c5956268697d7677eb351a908ac95ed9ef874323d73f32ec3debf","costSunk":"0","biasSupport":0,"biasOppose":0,"biasVouch":0,"biasNull":0,"blockHeight":null,"confirmations":0,"timestamp":"2026-11-06T21:05:33.500Z","tripFp":"a605e87e5f2e","viewCount":0,"powDifficulty":0,"relation":0,"ephemeral":true,"nonce":"b1ea999c75721b08","sig":"8128b24e1df99dac8f3ed799d0670826837ba0b307d03b6f8d27e1f607884ff632129477bd7d6ef58c0a3e951901a15663069b0056fc75eb4a29ab52580c162f","ephStamp":{"blockHashHex":"00000000000000000001d47c91bbdddac98ce59677ffa2b9447aceaa6bec1686","nonceHex":"11c3000000000000"},"replyCount":0,"chainReplyCount":0,"vaporReplyCount":0,"clusterSize":1,"receiptHeight":974857,"expiryHeight":977017,"authorMarks":{"since":"2026-11-06T21:05:33.528Z","via":"sight","inscribed":0,"likes":0,"biases":0},"lang":"en","servedBy":{"host":"6af58b1822e0cabfe57d76540f745d059b438f489a796ec6a705c6540233845e","t":"2026-11-08T19:44:20.601Z","sig":"24a3c30cd36046f005f07aaa83dbf5605e962ea620c5f2d37ff7e937c67d8593e53c27dee864fabc8bffdd595886519c654346706eca988f0fa54f990c6e6d72","h":"00000000000000000001e22bd31fbac6b11e9cb6f5b0df8bb85edfbe01a241f1"}}