uiai

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
uiai [2026/03/17 01:40] – [Definition: Third-party action] pedroortegauiai [2026/03/17 10:30] (current) – [Definition: Third-party action] pedroortega
Line 398: Line 398:
 Generate $(\dot{\gamma}_j,\dot{x}_j)_{j \ge k}$ as follows: Generate $(\dot{\gamma}_j,\dot{x}_j)_{j \ge k}$ as follows:
  
-**Shared prefix:** Set $\dot{\gamma}_{\le k-1} := \gamma_{\le k-1}$, $\dot{x}_{\le k-1} := x_{\le k-1}$.+  * **Shared prefix:** Set $\dot{\gamma}_{\le k-1} := \gamma_{\le k-1}$, $\dot{x}_{\le k-1} := x_{\le k-1}$.
  
-**Force an $\mathcal{A}$-block start:** Set $\dot{\gamma}_k := 1$.+  * **Force an $\mathcal{A}$-block start:** Set $\dot{\gamma}_k := 1$.
  
-**Evolve branch chronologically:** For $j \ge k$, first sample the next substrate symbol by $\dot{x}_j \sim \mu(\cdot \mid \underline{a\hat{o}}_{<t} a_t\,w\,\dot{x}_{k:j-1})$, so $\mu$ emits the content of the forced $\mathcal{A}$-block in the branch, conditioned on the shared past and the already-emitted branch block prefix. Then sample the next gate value by+  * **Evolve branch chronologically:** For $j \ge k$, first sample the next substrate symbol by $\dot{x}_j \sim \mu(\cdot \mid \underline{a\hat{o}}_{<t} a_t\,w\,\dot{x}_{k:j-1})$, so $\mu$ emits the content of the forced $\mathcal{A}$-block in the branch, conditioned on the shared past and the already-emitted branch block prefix. Then sample the next gate value by
 $$ $$
   \dot{\gamma}_{j+1} \sim \Gamma(\cdot \mid \dot{\gamma}_{\le j}, \dot{x}_{\le j}).   \dot{\gamma}_{j+1} \sim \Gamma(\cdot \mid \dot{\gamma}_{\le j}, \dot{x}_{\le j}).
Line 416: Line 416:
  
 Note that $k'$ is determined inside the branch and therefore the length of $\dot{a}_{t+1}$ need not match the length of the on-path $\mathcal{A}$-token written by the agent starting at $k$. Note that $k'$ is determined inside the branch and therefore the length of $\dot{a}_{t+1}$ need not match the length of the on-path $\mathcal{A}$-token written by the agent starting at $k$.
 +
 +{{ ::uiai-cf-action.png?600 |Counterfactual Action}}
  
 **Diagram note.**   **Diagram note.**  
-The intended picture is that after the shared on-path prefix ending at $a_3,o_3$, the on-path transcript has factual action $a_4$, while the counterfactual branch replaces that with the world-generated block $\dot{a}_4$ occupying the same would-be action slot.+The diagram shows that after the shared on-path prefix ending at $a_3,o_3$, the on-path transcript has factual action $a_4$, while the counterfactual action $\dot{a}_4$ is spawn from the same prefix but generated by the world.
  
 ==== Definition: Third-party action ==== ==== Definition: Third-party action ====
Line 448: Line 450:
 o_t = w\,\dot{a}_{t+1}\,v. o_t = w\,\dot{a}_{t+1}\,v.
 $$ $$
 +
 +{{ :uiai-third-party.png?600 |}}
  
 **Diagram note.**   **Diagram note.**  
-The intended picture is that inside a long world-written observation token, there is an embedded block $\dot{a}_4$ that occupies an $\mathcal{A}$-position under the tokenization convention, even though on-path it is still written by the world and therefore appears as evidence.+The diagram illustrates a third party action. Inside a long observation token $o_3$one can identify an embedded block $\dot{a}_4$ that is interpreted as a third-party action. Because it is written by the world, it counts as evidence.
  
 It is not hard to see that, for a given potential index $k$, the counterfactual action $\dot{a}_{t+1}$ and the third-party action $\dot{a}_{t+1}$ are the same random block: the difference is only whether the gate sampled $\gamma_k = 1$ (counterfactual, not observed) or $\gamma_k = 0$ (third-party, observed) at position $k$. The precise distinction between the different types of $\mathcal{A}$-tokens is important; we will also refer to them as //factual// (first-person, agent-generated), //counterfactual//, and //third-party// $\mathcal{A}$-tokens. It is not hard to see that, for a given potential index $k$, the counterfactual action $\dot{a}_{t+1}$ and the third-party action $\dot{a}_{t+1}$ are the same random block: the difference is only whether the gate sampled $\gamma_k = 1$ (counterfactual, not observed) or $\gamma_k = 0$ (third-party, observed) at position $k$. The precise distinction between the different types of $\mathcal{A}$-tokens is important; we will also refer to them as //factual// (first-person, agent-generated), //counterfactual//, and //third-party// $\mathcal{A}$-tokens.
  • uiai.1773711617.txt.gz
  • Last modified: 2026/03/17 01:40
  • by pedroortega