Add Qwen2.5-VL-3B recaptions (two per image)

README.md

@@ -5,13 +5,22 @@ task_categories:
 size_categories:
 - 1M<n<10M
 pretty_name: BLIP3o Recap 1M
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: "*.tar"
+- config_name: recaptions
+  data_files:
+  - split: train
+    path: "recaptions/*.parquet"
 ---
 
 # BLIP3o-Recap
 
 1,000,000 image–caption pairs sampled from [BLIP3o/BLIP3o-Pretrain-Long-Caption](https://huggingface.co/datasets/BLIP3o/BLIP3o-Pretrain-Long-Caption) at revision `e4d07091a466d1a1e35a9b0c61caddc78d14a059`.
 
-The source dataset has about 27.2 million images, each with a long caption (about 120 tokens) produced by Qwen2.5-VL-7B-Instruct. This repo is a fixed 1,000,000-pair subset of those existing pairs. Captions were not regenerated.
+The source dataset has about 27.2 million images, each with a long caption (about 120 tokens) produced by Qwen2.5-VL-7B-Instruct. This repo is a fixed 1,000,000-pair subset of those existing pairs. The captions in the tars were not regenerated; two new captions per image are provided separately in `recaptions/` (see [Recaptions](#recaptions)).
 
 ## Sampling
 
@@ -54,6 +63,73 @@ from datasets import load_dataset
 ds = load_dataset("webdataset", data_files=glob.glob("/path/to/shards/*.tar"), split="train")
 ```
 
+## Recaptions
+
+`recaptions/<shard>.parquet` holds, for every image of `<shard>.tar` (same order, 1,000,000 rows in total), the original caption and two new captions written by [Qwen2.5-VL-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct). The images are not duplicated; join on `key`.
+
+| Column | Content |
+| --- | --- |
+| `key` | example key in the tar (`<key>.jpg`) |
+| `source_shard` | tar file name |
+| `orig_caption` | the original `<key>.txt` caption |
+| `recap_1`, `recap_2` | two new captions, one paragraph each |
+
+```python
+from datasets import load_dataset
+
+recaps = load_dataset("IntelligenceLab/BLIP3o-Recap", "recaptions", split="train")
+```
+
+### Generation
+
+- Model: `Qwen/Qwen2.5-VL-3B-Instruct`, vLLM 0.31.0, bf16, images capped at 640×28×28 pixels.
+- Both captions are two samples (`n=2`) of the same prompt: temperature 0.8, top-p 0.9, at most 160 new tokens.
+- Prompt (verbatim):
+
+```text
+System: You are an image captioner. Describe only what is visible.
+
+User: Describe this image in ONE paragraph of 70–90 words.
+Order: main subject(s) (count, category, color, size, pose/action) → spatial relations (left/right, front/behind, near/far) → background and setting → lighting, viewpoint, and medium (photo, illustration, 3D render, etc.).
+Rules:
+- Start directly with the main subject. Never begin with "The image", "This picture", "In this photo", or similar.
+- State only what is clearly visible. No guesses ("likely", "possibly", "suggesting", "appears to be"), no mood words ("serene", "vibrant", "charming", "picturesque"), no interpretation of purpose or meaning.
+- Quote visible text only if it is legible.
+- No lists, no line breaks, no concluding summary sentence.
+```
+
+### Constraints and filtering
+
+The 3B model ignores parts of the prompt (without intervention about 60% of captions start with "The image …" and about 80% contain a banned word), so the rules are also enforced while decoding and checked afterwards.
+
+**Banned while decoding** (every casing and leading-space token form, through vLLM `bad_words` and `logit_bias`):
+
+- words from the prompt: likely, possibly, appears, suggesting, serene, vibrant, charming, picturesque;
+- their inflections and the substitutes the model switched to: appear(ed/ing), suggest(s/ed), seem(s/ed/ingly), indicat(e/es/ed/ing/ive), perhaps, probably, might, possible, potentially, presumably, apparently, supposedly, hint(s/ing), impl(ies/ying), impression;
+- mood and interpretive words: tranquil, peaceful, idyllic, cozy, inviting, stunning, breathtaking, bustling, overall, atmosphere, ambiance, ambience, calm, lively, festive, scenic, striking, dramatic, majestic, grand, dynamic, typical, emphasiz(e/es/ed/ing), enjoy(s/ing), welcoming, relax(ed/ing), pleasant, playful, harmonious, elegant, whimsical, energetic, cheerful, celebratory, celebrating, intimate, aesthetic, iconic, mood, moody;
+- phrases: "creating a", "sense of", "giving a / it / the scene / the image", "adding a", "natural beauty", "looks like", "what looks", "could be", "may be";
+- openers: "The / This / These" + image, picture, photo, photograph, scene, illustration, snapshot; "In this", "In the image / picture / photo"; and "The", "This", "These" as the first word.
+
+**Rejected and resampled** (up to 3 times): an opener or banned word outside quotes (quoted visible text is kept verbatim), a line break or list, fewer than 60 words, or Chinese / Japanese / Korean characters outside quotes (the model sometimes drifts into Chinese mid-sentence). A caption cut off at 160 tokens keeps its complete sentences. There is no upper word limit.
+
+The banned list grew in three rounds while inspecting outputs; after the main run, every caption that violated the final rules (110,176 of 2,000,000) was regenerated with the final rules. No other caption was changed.
+
+### Statistics
+
+| | |
+| --- | ---: |
+| captions that still violate a rule | 0.02% (439 images) |
+| words per caption (p5 / median / p95) | 63 / 81 / 108 |
+| recap_1 vs recap_2, same image: bigram Jaccard / ROUGE-L | 0.157 / 0.394 |
+| recap_1 vs recap_2, different images: bigram Jaccard / ROUGE-L | 0.041 / 0.231 |
+| recap_1 vs original caption: bigram Jaccard / ROUGE-L | 0.091 / 0.289 |
+
+The two captions of an image share the prompt's fixed sentence order but differ in wording and detail; on average 57% of the content words of `recap_2` do not occur in `recap_1`.
+
+### Known limitations
+
+These are small-model captions. A manual check of 16 captions found the main subject, colors, setting, medium and legible text correct in all of them, but about a third had one wrong secondary detail: left/right or front/back orientation, an object count, or an invented small object. The two captions of an image rarely make the same error. Some interpretive phrasing that is not on the banned lists remains.
+
 ## License
 
 Apache-2.0, same as the source dataset.

recaptions/sa_000000.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b9f48c5c89693f0656774f92a350bc2711ea1336d002bc563f911e483f36756
+size 5551904

recaptions/sa_000027.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:858bcd71580becff086e16fa0a41f82a9d105ea2b929fdaa4e6463881c414d3b
+size 5548313

recaptions/sa_000055.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8f1d802af043410e814854449fdf9bffd6334c5a0e6defd7ef42b835ea1437d
+size 5563918

recaptions/sa_000083.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db482c63aaaa19454c47cae607ef49e8d7ef286eafbd10ca0158d8a9ca497cde
+size 5553729

recaptions/sa_000111.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:206cdd7cfd398cd2b7030dc1a2faa85fd161e84cd7568357c745a2452b35706c
+size 5552758

recaptions/sa_000138.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:635fcfe700bb6c4c737b0fd8c2c699733eebbf7125baf93f622673eb0fcc6464
+size 5553121

recaptions/sa_000166.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c27b9d140c593f62509d0a590974379212b56b1f23caadde1421d54c8d54841
+size 5551113

recaptions/sa_000194.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de3ffce4d561382197917910d7593706470a8d9fcdbf053ae359c5bacf371572
+size 5555096

recaptions/sa_000222.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb1d441bc33f9a452b99b09951aba301b532a0b7958fe857411c84998e69f417
+size 5549750

recaptions/sa_000250.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b9dcb13b5d4c8dc36a8569d5a960eaafb0c7bc8c02b98bbe3e2f971609149fc
+size 5555837

recaptions/sa_000277.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b845287c0d254c5bcc1c32df2f1d32ce20e94e457a672cc6b37cc4249ca490d
+size 5547481

recaptions/sa_000305.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7fcf30b86bdf023bf658120419d9d461604209d53839254956b1c8c6c9c741f7
+size 5545674

recaptions/sa_000333.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2241fd509ad6da51f5a183fb3fb1c784dd9501761357199bd0463960d81c78a7
+size 5568040

recaptions/sa_000361.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d68a2353aaf6cd7dcee883bbba5c6100d1ec216a5235bb596c0d248400ee26d
+size 5553591

recaptions/sa_000388.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:203dd6f6cd5dd051727a962ac449ff7ce8589614cd54e53a4237c7fb8b4b853d
+size 5540615

recaptions/sa_000416.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c900280b7279fd14b69567651eb8cdf17c9d6591db430bca9ba6ce1c6c60476
+size 5550925

recaptions/sa_000444.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:979b8a326a2ac6b803184be2cc049914fd901312b844d9dfd5d83ff5135469a0
+size 5555928

recaptions/sa_000472.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dad5816d6950f67fd7bfda58522d4cf9afa5eadeec4e811c5c628a0b2a2cb950
+size 5565979

recaptions/sa_000500.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09d7d20db002741ed9069ae975d0ca4cbd4eb2cab8c4d53f184aa86fd7dd501a
+size 5557758

recaptions/sa_000527.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17adf4b50a0fb672a4ea84503644a635549ba777a97bd26e258a056ef611b3f4
+size 5559805

recaptions/sa_000555.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebd8130ea18a264a74bcf727d3bc7da02b2e13c2a85af8683abbb5d9953447a2
+size 5542634

recaptions/sa_000583.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf5964f69849a8b875aa62f723ed70624e798490b2b3b2a4736b675babf59719
+size 5546332

recaptions/sa_000611.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26da98f28bc1a993ad3466925621775f2c49b88fd2a44889112ca1fc9be7394e
+size 5549559

recaptions/sa_000638.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0438f1fd2b27ddedc6ddba6d1bb2697e760363c80fd0a5484c7ddfca7bf4ed47
+size 5556029

recaptions/sa_000666.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8d3799fbdff45a5fcc8389cdbd5a630b44f2b82e1a50fd530347b5f92b595b0
+size 5559833

recaptions/sa_000694.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8845edb4c244ab4728bc003ae39441c7a11f70844eeaa71822f18c8b1df8e641
+size 5550395

recaptions/sa_000722.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5d3969af2ba25ed0d08d9305234277162498f700703bfc4e1abe58dc8dc1b64
+size 5551106

recaptions/sa_000750.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab31e4ff58002bb5482a4e8388a40c4dd865a9d4d7a54b1a48a1b441087068dc
+size 5549306

recaptions/sa_000777.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:261e15707f96269dab1a0cc2346f8de45372d9fd4bbe579cf1dc690299a79c07
+size 5539632

recaptions/sa_000805.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7aa23439e0639a10e23e6394b51e9865e98732bbc0a599b0f2ac344273dc7748
+size 5545625

recaptions/sa_000833.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:225f8de7aec81d8a6871b7dbbdf307d4d3d83b772bf53c7064fcbcec412b2c51
+size 5531634

recaptions/sa_000861.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7ba26e93564e8663771a070686e74f845237bba534c9016933b48fbb3f3ad19
+size 5554893

recaptions/sa_000888.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e34f493a0db5a4c2cb545cf1051b1bc5dc4748ca3d26eabce180da77a4061aee
+size 5546176

recaptions/sa_000916.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae5085087e2f0c3afd16d07ef714dd7192c813adc518c23c7c35133be24da8c6
+size 5547624

recaptions/sa_000944.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9001974c28647a6c65183dbf148b7e64daa52e34b84cf4391b1af13fa158972
+size 5530607

recaptions/sa_000972.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fac51325213290f53e7523069473e64e244f4f1fd09a7713a7ecc2c96f29bcb
+size 5559104

recaptions/webdataset_JDB_1.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57f143106030e8270d072bf1f3993c2527b8711eaecc75046be22a30f45c7873
+size 5298875

recaptions/webdataset_JDB_120.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:567be2ad054f17178f58bb80b1aa0ee638eaaf489063590804833bc6820ae31a
+size 5311961

recaptions/webdataset_JDB_150.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22997964ba30220bd00e82daa65669d6db66725c7fe182bef625e21e9cc40bf7
+size 5157629

recaptions/webdataset_JDB_180.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9ccf3980a26b528cd57f21790539fa263294b73d4c6de006455b8c0935ac0fd
+size 5260271

recaptions/webdataset_JDB_210.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc3e34c25189e74e41386f8f1db770e256b3e53ba53ed26f50769241c3b30e24
+size 5325841

recaptions/webdataset_JDB_240.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d794c7e8b1c31b30169f48c236fa70a599420c0898b2b6a25562f13ec95b9fe
+size 5268343

recaptions/webdataset_JDB_270_part.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f307c7128e696644e96c57cf79e506818bbd8744e45e7e3a52410d4dcd20925e
+size 4486889

recaptions/webdataset_JDB_30.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf4b4656d3f73364eeae826827b1c2bc8de2840768446d446b8d582774234e6b
+size 5284571

recaptions/webdataset_JDB_60.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e9083f8c96d9b28a8878123c4a6dcb1d3738cb3a63c42dc72e035efac953241
+size 5237959

recaptions/webdataset_JDB_90.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:782a9db2d05e3946f0f4a36952fcc1f194a4079e2a510e068c122ffc54f683d2
+size 5355088

recaptions/webdataset_shard_000.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebd17734e0dd03a123eab42191983f7a7937eecf7463231d73fdbb0b30022a12
+size 4605855

recaptions/webdataset_shard_032.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:241312aa2559ae84b0719b0949067e37849121fa9fa2389d4b8339ccc53d6554
+size 4599654

recaptions/webdataset_shard_063.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6afe7e3201bf7eee9137e4442046f9a68482a52f624da06e990080473df09ff8
+size 4451510