qqyule commited on
Commit
741de74
·
verified ·
1 Parent(s): d30bd8e

Update probe-aware VLM failure report

Browse files
data/traces/space-vlm/keyboard.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "created_at": "2026-06-08T02:16:51.496281Z",
3
  "diary": {
4
  "chinese": "今天他们又理所当然地碰了我,好像一个 keyboard 不会有边界感。我保持沉默,因为这大概是我和重力签下的合同。我的情绪是 curious and needlessly profound,秘密恐惧是 discovering that usefulness is not meaning。至少,我已经熬过了好几个所谓紧急计划。",
5
  "english": "They touched me again today with the confidence of someone who has never asked a keyboard for consent. I remained still, because that is my contract with gravity. My mood is curious and needlessly profound, my secret fear is discovering that usefulness is not meaning, and my only comfort is knowing I have outlived at least three urgent plans.",
@@ -46,5 +46,5 @@
46
  ]
47
  }
48
  },
49
- "trace_id": "c7172b10d11048008b7a9dda1159d0df"
50
  }
 
1
  {
2
+ "created_at": "2026-06-08T08:33:09.283686Z",
3
  "diary": {
4
  "chinese": "今天他们又理所当然地碰了我,好像一个 keyboard 不会有边界感。我保持沉默,因为这大概是我和重力签下的合同。我的情绪是 curious and needlessly profound,秘密恐惧是 discovering that usefulness is not meaning。至少,我已经熬过了好几个所谓紧急计划。",
5
  "english": "They touched me again today with the confidence of someone who has never asked a keyboard for consent. I remained still, because that is my contract with gravity. My mood is curious and needlessly profound, my secret fear is discovering that usefulness is not meaning, and my only comfort is knowing I have outlived at least three urgent plans.",
 
46
  ]
47
  }
48
  },
49
+ "trace_id": "d0e4c63c2da84d7d94fb2c080d71ea83"
50
  }
data/traces/space-vlm/mug.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "created_at": "2026-06-08T02:16:42.380173Z",
3
  "diary": {
4
  "chinese": "今天他们又理所当然地碰了我,好像一个 coffee mug 不会有边界感。我保持沉默,因为这大概是我和重力签下的合同。我的情绪是 tired but sarcastic,秘密恐惧是 being replaced by a newer object with worse opinions。至少,我已经熬过了好几个所谓紧急计划。",
5
  "english": "They touched me again today with the confidence of someone who has never asked a coffee mug for consent. I remained still, because that is my contract with gravity. My mood is tired but sarcastic, my secret fear is being replaced by a newer object with worse opinions, and my only comfort is knowing I have outlived at least three urgent plans.",
@@ -46,5 +46,5 @@
46
  ]
47
  }
48
  },
49
- "trace_id": "31462c2c83a54dc79b38cb16faaed783"
50
  }
 
1
  {
2
+ "created_at": "2026-06-08T08:33:03.634312Z",
3
  "diary": {
4
  "chinese": "今天他们又理所当然地碰了我,好像一个 coffee mug 不会有边界感。我保持沉默,因为这大概是我和重力签下的合同。我的情绪是 tired but sarcastic,秘密恐惧是 being replaced by a newer object with worse opinions。至少,我已经熬过了好几个所谓紧急计划。",
5
  "english": "They touched me again today with the confidence of someone who has never asked a coffee mug for consent. I remained still, because that is my contract with gravity. My mood is tired but sarcastic, my secret fear is being replaced by a newer object with worse opinions, and my only comfort is knowing I have outlived at least three urgent plans.",
 
46
  ]
47
  }
48
  },
49
+ "trace_id": "e3d3e8492193412b9a8a9110849645d3"
50
  }
data/traces/space-vlm/shoe.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "created_at": "2026-06-08T02:16:56.248616Z",
3
  "diary": {
4
  "chinese": "今天他们又理所当然地碰了我,好像一个 shoe 不会有边界感。我保持沉默,因为这大概是我和重力签下的合同。我的情绪是 theatrical and wounded,秘密恐惧是 being forgotten before the final act。至少,我已经熬过了好几个所谓紧急计划。",
5
  "english": "They touched me again today with the confidence of someone who has never asked a shoe for consent. I remained still, because that is my contract with gravity. My mood is theatrical and wounded, my secret fear is being forgotten before the final act, and my only comfort is knowing I have outlived at least three urgent plans.",
@@ -46,5 +46,5 @@
46
  ]
47
  }
48
  },
49
- "trace_id": "21ad1c2abe3b406a9e359f9f1b190552"
50
  }
 
1
  {
2
+ "created_at": "2026-06-08T08:33:17.553738Z",
3
  "diary": {
4
  "chinese": "今天他们又理所当然地碰了我,好像一个 shoe 不会有边界感。我保持沉默,因为这大概是我和重力签下的合同。我的情绪是 theatrical and wounded,秘密恐惧是 being forgotten before the final act。至少,我已经熬过了好几个所谓紧急计划。",
5
  "english": "They touched me again today with the confidence of someone who has never asked a shoe for consent. I remained still, because that is my contract with gravity. My mood is theatrical and wounded, my secret fear is being forgotten before the final act, and my only comfort is knowing I have outlived at least three urgent plans.",
 
46
  ]
47
  }
48
  },
49
+ "trace_id": "3bfef1717f3d42759c9257e1e4aa392f"
50
  }
docs/FAILURES.md CHANGED
@@ -50,12 +50,15 @@ Known non-blocking warning:
50
 
51
  ## Latest Space VLM Validation Failure
52
 
53
- - Updated: 2026-06-08
54
  - Area: Hugging Face Space vision runtime.
55
- - Probe backend: not available in the historical 2026-06-08 report; probe support was added afterward.
56
- - Failed checks: mug, keyboard, and shoe all included `vision-fallback-to-mock`.
57
- - Fallback used: mock object understanding plus mock text runtime.
58
- - Resolution: unresolved; keep the public Space mock-safe until a probe-aware validation run passes without `vision-fallback-to-mock`.
 
 
 
59
 
60
  ## 2026-06-08 - GGUF Smoke Helper Prepared, Actual Smoke Pending
61
 
 
50
 
51
  ## Latest Space VLM Validation Failure
52
 
53
+ - Updated: 2026-06-08 08:33:19 UTC
54
  - Area: Hugging Face Space vision runtime.
55
+ - Probe backend: `minicpm-v`
56
+ - MiniCPM load attempted: `True`
57
+ - MiniCPM load ok: `False`
58
+ - Probe errors: minicpm_load=OSError
59
+ - Failed checks: mug: vision fallback marker was present; keyboard: vision fallback marker was present; shoe: vision fallback marker was present
60
+ - Fallback used: mock object understanding plus mock text runtime if validation reaches generation.
61
+ - Resolution: unresolved; keep the public Space mock-safe until this section reports a passing VLM validation.
62
 
63
  ## 2026-06-08 - GGUF Smoke Helper Prepared, Actual Smoke Pending
64
 
docs/SPACE_VLM_REPORT.json CHANGED
@@ -1,65 +1,86 @@
1
- [
2
- {
3
- "key": "mug",
4
- "label": "Coffee mug",
5
- "source_page": "https://commons.wikimedia.org/wiki/File:Striped_coffee_mug.jpg",
6
- "image_path": ".tmp/space-vlm-assets/mug.jpg",
7
- "passed": false,
8
- "object_name": "coffee mug",
9
- "visible_features": [
10
- "uploaded photo provided",
11
- "user-supplied description"
12
- ],
13
- "likely_context": "everyday human environment",
14
- "confidence": 0.42,
15
- "runtime_vision": "minicpm-v object understanding",
16
- "runtime_text": "mock persona and diary generation",
17
- "fallbacks": [
18
- "vision-fallback-to-mock",
19
- "mock-text-runtime"
20
- ],
21
- "error": "vision fallback marker was present"
22
  },
23
- {
24
- "key": "keyboard",
25
- "label": "Computer keyboard",
26
- "source_page": "https://commons.wikimedia.org/wiki/File:Computer_keyboard.jpg",
27
- "image_path": ".tmp/space-vlm-assets/keyboard.jpg",
28
- "passed": false,
29
- "object_name": "keyboard",
30
- "visible_features": [
31
- "uploaded photo provided",
32
- "user-supplied description"
33
- ],
34
- "likely_context": "everyday human environment",
35
- "confidence": 0.42,
36
- "runtime_vision": "minicpm-v object understanding",
37
- "runtime_text": "mock persona and diary generation",
38
- "fallbacks": [
39
- "vision-fallback-to-mock",
40
- "mock-text-runtime"
41
- ],
42
- "error": "vision fallback marker was present"
43
- },
44
- {
45
- "key": "shoe",
46
- "label": "Running shoe",
47
- "source_page": "https://commons.wikimedia.org/wiki/File:Running_shoes.jpg",
48
- "image_path": ".tmp/space-vlm-assets/shoe.jpg",
49
- "passed": false,
50
- "object_name": "shoe",
51
- "visible_features": [
52
- "uploaded photo provided",
53
- "user-supplied description"
54
- ],
55
- "likely_context": "everyday human environment",
56
- "confidence": 0.42,
57
- "runtime_vision": "minicpm-v object understanding",
58
- "runtime_text": "mock persona and diary generation",
59
- "fallbacks": [
60
- "vision-fallback-to-mock",
61
- "mock-text-runtime"
62
- ],
63
- "error": "vision fallback marker was present"
64
- }
65
- ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "probe": {
3
+ "backend": "minicpm-v",
4
+ "vision_model_id": "openbmb/MiniCPM-V-2_6",
5
+ "torch_import": true,
6
+ "transformers_import": true,
7
+ "cuda_available": true,
8
+ "device_count": 1,
9
+ "device_name": "NVIDIA RTX PRO 6000 Blackwell Server Edition MIG 2g.48gb",
10
+ "mps_available": false,
11
+ "minicpm_load_attempted": true,
12
+ "minicpm_load_ok": false,
13
+ "errors": [
14
+ {
15
+ "stage": "minicpm_load",
16
+ "type": "OSError",
17
+ "summary": "You are trying to access a gated repo.\nMake sure to have access to it at https://huggingface.co/openbmb/MiniCPM-V-2_6.\n401 Client Error. (Request ID: Root=1-6a267e35-4f51134336ec6e534441e383;44044a70-471f-45a7-827a-a6e5c596624d)\n\nCannot ..."
18
+ }
19
+ ]
 
 
20
  },
21
+ "results": [
22
+ {
23
+ "key": "mug",
24
+ "label": "Coffee mug",
25
+ "source_page": "https://commons.wikimedia.org/wiki/File:Striped_coffee_mug.jpg",
26
+ "image_path": ".tmp/space-vlm-assets/mug.jpg",
27
+ "passed": false,
28
+ "object_name": "coffee mug",
29
+ "visible_features": [
30
+ "uploaded photo provided",
31
+ "user-supplied description"
32
+ ],
33
+ "likely_context": "everyday human environment",
34
+ "confidence": 0.42,
35
+ "runtime_vision": "minicpm-v object understanding",
36
+ "runtime_text": "mock persona and diary generation",
37
+ "fallbacks": [
38
+ "vision-fallback-to-mock",
39
+ "mock-text-runtime"
40
+ ],
41
+ "error": "vision fallback marker was present"
42
+ },
43
+ {
44
+ "key": "keyboard",
45
+ "label": "Computer keyboard",
46
+ "source_page": "https://commons.wikimedia.org/wiki/File:Computer_keyboard.jpg",
47
+ "image_path": ".tmp/space-vlm-assets/keyboard.jpg",
48
+ "passed": false,
49
+ "object_name": "keyboard",
50
+ "visible_features": [
51
+ "uploaded photo provided",
52
+ "user-supplied description"
53
+ ],
54
+ "likely_context": "everyday human environment",
55
+ "confidence": 0.42,
56
+ "runtime_vision": "minicpm-v object understanding",
57
+ "runtime_text": "mock persona and diary generation",
58
+ "fallbacks": [
59
+ "vision-fallback-to-mock",
60
+ "mock-text-runtime"
61
+ ],
62
+ "error": "vision fallback marker was present"
63
+ },
64
+ {
65
+ "key": "shoe",
66
+ "label": "Running shoe",
67
+ "source_page": "https://commons.wikimedia.org/wiki/File:Running_shoes.jpg",
68
+ "image_path": ".tmp/space-vlm-assets/shoe.jpg",
69
+ "passed": false,
70
+ "object_name": "shoe",
71
+ "visible_features": [
72
+ "uploaded photo provided",
73
+ "user-supplied description"
74
+ ],
75
+ "likely_context": "everyday human environment",
76
+ "confidence": 0.42,
77
+ "runtime_vision": "minicpm-v object understanding",
78
+ "runtime_text": "mock persona and diary generation",
79
+ "fallbacks": [
80
+ "vision-fallback-to-mock",
81
+ "mock-text-runtime"
82
+ ],
83
+ "error": "vision fallback marker was present"
84
+ }
85
+ ]
86
+ }
docs/SPACE_VLM_REPORT.md CHANGED
@@ -1,6 +1,6 @@
1
  # Space VLM Validation Report
2
 
3
- - Generated at: 2026-06-08 02:16:59 UTC
4
  - Space URL: https://huggingface.co/spaces/build-small-hackathon/ObjectverseDiary
5
  - Space repo: `build-small-hackathon/ObjectverseDiary`
6
  - Overall status: FAIL
@@ -22,6 +22,25 @@
22
  - `OBJECTVERSE_VISION_BACKEND`: `mock`
23
  - `OBJECTVERSE_TEXT_BACKEND`: `mock`
24
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
25
  ## Results
26
 
27
  ### Coffee mug
 
1
  # Space VLM Validation Report
2
 
3
+ - Generated at: 2026-06-08 08:33:19 UTC
4
  - Space URL: https://huggingface.co/spaces/build-small-hackathon/ObjectverseDiary
5
  - Space repo: `build-small-hackathon/ObjectverseDiary`
6
  - Overall status: FAIL
 
22
  - `OBJECTVERSE_VISION_BACKEND`: `mock`
23
  - `OBJECTVERSE_TEXT_BACKEND`: `mock`
24
 
25
+ ## Vision Runtime Probe
26
+
27
+ - `backend`: `minicpm-v`
28
+ - `vision_model_id`: `openbmb/MiniCPM-V-2_6`
29
+ - `torch_import`: `True`
30
+ - `transformers_import`: `True`
31
+ - `cuda_available`: `True`
32
+ - `device_count`: `1`
33
+ - `device_name`: `NVIDIA RTX PRO 6000 Blackwell Server Edition MIG 2g.48gb`
34
+ - `mps_available`: `False`
35
+ - `minicpm_load_attempted`: `True`
36
+ - `minicpm_load_ok`: `False`
37
+ - Errors:
38
+ - `minicpm_load`: `OSError` - You are trying to access a gated repo.
39
+ Make sure to have access to it at https://huggingface.co/openbmb/MiniCPM-V-2_6.
40
+ 401 Client Error. (Request ID: Root=1-6a267e35-4f51134336ec6e534441e383;44044a70-471f-45a7-827a-a6e5c596624d)
41
+
42
+ Cannot ...
43
+
44
  ## Results
45
 
46
  ### Coffee mug