Update results: 40/400 (10.0%) solved — +9 new tasks, 0 regressions, greedy stacker working

arc_results/RESULTS.md

@@ -1,17 +1,33 @@
 # PEMF Solver — ARC-AGI Training Set Evaluation
 
-## Results
+## Results (v2 — object layer + greedy stacker)
 
-| Metric | Value |
-|---|---|
-| **Total tasks** | 400 |
-| **Solved** (σ=0 all train pairs) | **31 (7.8%)** |
-| **≥1 pair solved** | 59 (14.8%) |
-| **Total pairs** | 1,302 |
-| **Pairs solved** | 146 (11.2%) |
-| **Total time** | 17.1s (0.04s/task) |
+| Metric | v1 | **v2** | Change |
+|---|---|---|---|
+| **Tasks solved** | 31 (7.8%) | **40 (10.0%)** | **+9 (+29%)** |
+| ≥1 pair solved | 59 (14.8%) | — | — |
+| Pairs solved | 146 (11.2%) | — | — |
+| Transforms | 19 | **33** | +14 |
+| Total time | 17.1s | **50.8s** | +greedy stacker |
+| Regressions | — | **0** | — |
 
-## Solved Tasks (31)
+## 9 Newly Solved Tasks
+
+| Task ID | Transform | Category |
+|---|---|---|
+| 1f85a75f | ExtractLargestObject | **Object extraction** |
+| 22168020 | ConnectSameColorH | **Connect** |
+| 22eb0ac0 | ConnectSameColorH | **Connect** |
+| 23b5c85d | ExtractSmallestObject | **Object extraction** |
+| 4347f46a | DrawBorder | **Border** |
+| 746b3537 | tile_to_target | Tiling (new pair coverage) |
+| be94b721 | CropToContent | Crop (new pair coverage) |
+| ded97339 | overlay(ConnectSameColorV, ConnectSameColorH) | **Greedy stacker** |
+| eb5a1d5d | CompressGrid | **Compress** |
+
+**6 of the 9 new solves use new transforms.** 1 uses the greedy stacker overlay composition.
+
+## All 40 Solved Tasks
 
 | Task ID | Transform | Family |
 |---|---|---|
@@ -19,13 +35,18 @@
 | 1190e5a7 | KroneckerSelfSimilarInv | Self-similar |
 | 1cf80156 | CropToContent | Crop |
 | 1e0a9b12 | GravityDown | Gravity |
+| 1f85a75f | ExtractLargestObject | Object |
 | 2013d3e2 | CropToContent | Crop |
+| 22168020 | ConnectSameColorH | Connect |
+| 22eb0ac0 | ConnectSameColorH | Connect |
 | 239be575 | tile_to_target | Tiling |
+| 23b5c85d | ExtractSmallestObject | Object |
 | 28bf18c6 | CropToContent | Crop |
 | 2dee498d | tile_to_target | Tiling |
 | 3906de3d | GravityUp | Gravity |
 | 3af2c5a8 | MirrorTileH | Mirror |
 | 3c9b0459 | Rotate_180 | Rotation |
+| 4347f46a | DrawBorder | Border |
 | 6150a2bd | Rotate_180 | Rotation |
 | 62c24649 | MirrorTile4Way | Mirror |
 | 67a3c6ac | Reflect_v | Reflection |
@@ -33,62 +54,46 @@
 | 68b16354 | Reflect_h | Reflection |
 | 6d0aefbc | MirrorTileH | Mirror |
 | 6fa7a44f | MirrorTileV | Mirror |
+| 746b3537 | tile_to_target | Tiling |
 | 74dd1130 | Transpose | Transpose |
 | 7b7f7511 | tile_to_target | Tiling |
 | 8be77c9e | MirrorTileV | Mirror |
 | 9172f3a0 | Upscale_3x | Upscale |
 | 9dfd6313 | Transpose | Transpose |
 | a416b8f3 | tile_to_target | Tiling |
+| be94b721 | CropToContent | Crop |
 | c59eb873 | Upscale_2x | Upscale |
 | c9e6f938 | MirrorTileH | Mirror |
 | d10ecb37 | tile_to_target | Tiling |
-| d631b094 | GravityUp | Gravity |
+| d631b094 | ExtractLargestObject | Object |
 | d9fac9be | tile_to_target | Tiling |
-| de1cd16c | Rotate_270 | Rotation |
+| de1cd16c | KeepSmallestObject | Object |
+| ded97339 | overlay(ConnectSameColorV, ConnectSameColorH) | Stacker |
+| eb5a1d5d | CompressGrid | Compress |
 | ed36ccf7 | Rotate_90 | Rotation |
 
-## Transform Usage (across all 146 solved pairs)
-
-| Transform | Pairs |
-|---|---|
-| tile_to_target | 34 |
-| CropToContent | 19 |
-| MirrorTile4Way | 11 |
-| Rotate_90 | 9 |
-| MirrorTileH | 8 |
-| Rotate_180 | 7 |
-| MirrorTileV | 7 |
-| Transpose | 7 |
-| Upscale_2x | 6 |
-| ShiftedTile | 6 |
-| Upscale_3x | 6 |
-| KroneckerSelfSimilar | 5 |
-| GravityUp | 5 |
-| GravityDown | 4 |
-| Reflect_h | 4 |
-| KroneckerSelfSimilarInv | 3 |
-| Reflect_v | 3 |
-| InvertColors | 1 |
-| Rotate_270 | 1 |
-
-**Every single new transform contributed to at least one solve.**
-
-## Unsolved σ Distribution
+## Architecture
 
-| σ Range | Pairs |
-|---|---|
-| (0, 5] | 56 |
-| (5, 10] | 85 |
-| (10, 20] | 155 |
-| (20, 50] | 341 |
-| (50, 100] | 230 |
-| (100, 500] | 263 |
-| (500+) | 26 |
+### 33 Atomic Transforms
+- **Tiling**: tile_to_target, ShiftedTile, FillEnclosedHarmonic
+- **Self-similar**: KroneckerSelfSimilar, KroneckerSelfSimilarInv
+- **Mirror**: MirrorTileH, MirrorTileV, MirrorTile4Way
+- **Upscale**: Upscale_2x, Upscale_3x
+- **Stack**: StackH_3, StackV_3
+- **Structural**: Transpose, CropToContent
+- **Object**: ExtractLargest/Smallest/Unique/MostCommon, KeepLargest/Smallest, SortBySize
+- **Fill/Connect**: FillInterior, ConnectSameColorH/V
+- **Compress**: CompressGrid, RemoveBlackLines
+- **Spatial**: ColorByProximity, DrawBorder
+- **Symmetry**: Rotate_90/180/270, Reflect_h/v
+- **Gravity**: GravityDown, GravityUp (optional)
+- **Color**: InvertColors (optional)
 
-Median σ for unsolved pairs: 44.0
+### Greedy Stacker
+After beam search, tries `overlay(T1(input), T2(input))` for the top-N depth-1 pieces. This allows composition of two independent transforms — critical for tasks where the output is a combination of two views.
 
-## Almost-Solved Tasks (28)
-Tasks where ≥1 pair reaches σ=0 but not all. These likely need a **composition** of two transforms or a new primitive not yet in the library.
+### Object Layer (`object_layer.py`)
+Connected component extraction (4-conn, 8-conn), color splitting, list reducers (largest/smallest/most_common/unique), spatial queries (bbox, center, height, width), and composition (paint/underpaint/overlay/cover/canvas).
 
 ## Parameters
 ```json