ARC-AGI // TERMINAL TELEMETRY

LLM Pattern Matching vs. Topological Verification · PTTE V9 · EOSE Fleet
ABR-838 · IRF-ARC-DSL-003

γ₁ = 14.134725141734693
⚡ THE WALL
Rick's Law confirmed by experiment: P_c = S × (architecture) × (inference) — when S → 0 (model cannot perceive grid topology), P_c = 0 regardless of parameter count. 14B parameters on a 30×30 grid = 14 billion wrong guesses. The fix is not scale — it is perception. IRF-ARC-DSL-003 is the unlock.
Model Dataset Tasks / Correct Cell Match Verdict
qwen2.5:7b AGI-1 training (base) 20 / 0 19.7% Pattern Noise
qwen2.5:7b AGI-1 training ×8 editions 160 / 0 0.0% Memorised, Not Learned
qwen2.5:14b AGI-2 training 20 / 1 (5%) 0.3% First Blood — 017c7c7b ✓
qwen2.5:14b AGI-2 eval 10 / 0 0.1% ← THE WALL Bounced
qwen2.5:7b AGI-2 training 20 / 0 0.2% Bounced
TERMINAL DIAGNOSTIC: The 7b hit 19.7% cell match on base AGI-1 — pure memorisation. The moment you added 8× parameter variation, it flatlined to 0.0%. It didn't learn to reason; it memorised the training pixels. Memory broke the instant the geometry shifted.

The 14b hit AGI-2 eval at 0.1% cell match on 30×30 grids. Mathematically blind. Not a parameter failure — a perception failure. The model cannot perceive discrete topological objects. Pumping 14B parameters at it generates 14B wrong guesses.

Conclusion: You cannot scale your way out of this. The architecture must change.
THE PIVOT — objects() STRATEGY (IRF-ARC-DSL-003)
1
Lean 4 parses the grid — BFS flood-fill extracts List (ARC_Object g). 900 integers → N discrete entities. Dimensionality collapses. Signal spikes.
2
LLM receives object representation — not pixel arrays. "Object at (5,5), color 2, bbox (3×3), 9px"
3
LLM outputs a single Lean 4 line: translate_object obj1 2 0
4
Lean 4 compiler verifies it. Proof-checked. Done. That is how you crack the 30×30 wall.

ACTIVE BURN: IRF-ARC-DSL-003

THE OPTIC NERVE. Abandoning raw grid inference. Deterministic flood-fill object extraction in Lean 4. LLM acts only on verified ARC_Object topologies — bypassing the 30×30 noise wall entirely.

structure ARC_Object {H W : ℕ}
    (g : ARC_Grid H W) where
  color  : ARC_Color
  pixels : Finset (ARC_Cell H W)
  is_not_empty   : pixels.Nonempty
  is_monochrome  : ∀ p ∈ pixels,
                  g p = color
  is_connected   : ∀ p1 ∈ pixels, ∀ p2 ∈ pixels,
                  connected4_in_finset pixels p1 p2
  is_maximal    : ∀ p, g p = color →
                  (∃ p' ∈ pixels, adjacent4 p p') →
                  p ∈ pixels

def extract_objects ... : List (ARC_Object g) :=
  sorry -- OVERSEER synthesizes BFS here

def translate_object
  (obj : ARC_Object g) (dx dy : Int) :
  Option (Finset (ARC_Cell H W)) := ...
[ SYSTEM ] : AWAITING OBJECTS() ENUMERATION
[ STATUS ] : [ RUNNING — BFS SYNTHESIS ]

IRF-ARC-DSL REGISTRY

DSL-001
ARC_Grid, ARC_Color, ARC_Cell, 10-color finite type
DSL-002
Translation — shift_cell, Option walls, translate_object
🔵
DSL-003
THE OPTIC NERVE — ARC_Object (4 axioms) + extract_objects BFS = OVERSEER synthesis target
🔴
DSL-004
flood_fill — BFS + termination proof
DSL-005
reflect_h, reflect_v — involutive proofs closed
🔴
DSL-006
Synthesis engine — find f | f(X₁)=Y₁ ∧ f(X₂)=Y₂
DSL-007
map_colors / recolor — src→dst mapping closed
DSL-008
overlay — object permanence, layer wins
🔴
DSL-009
rotate_90 — H=W square constraint