ATProto Browser

ATProto Browser

Experimental browser for the Atmosphere

Post

Sakana introduces new AI architecture, ‘Continuous Thought Machines’ to make models reason with less guidance — like human brains https://venturebeat.com/ai/sakana-introduces-new-ai-architecture-continuous-thought-machines-to-make-models-reason-with-less-guidance-like-human-brains/ #AI #reasoning

May 13, 2025, 2:13 AM

Record data

{
  "uri": "at://did:plc:rhkcyc46ubi523e47bhnkpbb/app.bsky.feed.post/3lozfwssalv2c",
  "cid": "bafyreiejqm7sczgtbu2umyx47ixe4cqdvp3mh22c2swkt5gmglbm7kstmi",
  "value": {
    "text": "Sakana introduces new AI architecture, ‘Continuous Thought Machines’ to make models reason with less guidance — like human brains https://venturebeat.com/ai/sakana-introduces-new-ai-architecture-continuous-thought-machines-to-make-models-reason-with-less-guidance-like-human-brains/ #AI #reasoning",
    "$type": "app.bsky.feed.post",
    "embed": {
      "$type": "app.bsky.embed.images",
      "images": [
        {
          "alt": "Text Shot: Rather than relying on fixed, parallel layers that process inputs all at once — as Transformer models do —CTMs unfold computation over steps within each input/output unit, known as an artificial “neuron.”\n\nEach neuron in the model retains a short history of its previous activity and uses that memory to decide when to activate again.\n\nThis added internal state allows CTMs to adjust the depth and duration of their reasoning dynamically, depending on the complexity of the task. As such, each neuron is far more informationally dense and complex than in a typical Transformer model.",
          "image": {
            "$type": "blob",
            "ref": {
              "$link": "bafkreib736phvegwal2ksju5s3cmpnbenvjfzpebablnop3dbezwlsmvsa"
            },
            "mimeType": "image/png",
            "size": 215407
          }
        }
      ]
    },
    "facets": [
      {
        "index": {
          "byteEnd": 288,
          "byteStart": 136
        },
        "features": [
          {
            "uri": "https://venturebeat.com/ai/sakana-introduces-new-ai-architecture-continuous-thought-machines-to-make-models-reason-with-less-guidance-like-human-brains/",
            "$type": "app.bsky.richtext.facet#link"
          }
        ]
      },
      {
        "index": {
          "byteEnd": 292,
          "byteStart": 289
        },
        "features": [
          {
            "tag": "AI",
            "$type": "app.bsky.richtext.facet#tag"
          }
        ]
      },
      {
        "index": {
          "byteEnd": 303,
          "byteStart": 293
        },
        "features": [
          {
            "tag": "reasoning",
            "$type": "app.bsky.richtext.facet#tag"
          }
        ]
      }
    ],
    "createdAt": "2025-05-13T02:13:13.387Z"
  }
}