Datasets:
Commit ·
19d0c5a
0
Parent(s):
Duplicate from SII-Monument-Valley/CiQi-VQA
Browse filesCo-authored-by: SII-Monument-Valley <SII-Monument-Valley@users.noreply.huggingface.co>
This view is limited to 50 files because it contains too many changes. See raw diff
- .gitattributes +61 -0
- README.md +148 -0
- images/00044b400078a592f3c2632074373091218087ef98c3163d54ee3effdddc24ca.jpg +3 -0
- images/00105ae503fb17e659e2488f22763af61776a6144a6216f92f5bcd8d98de6a46.jpg +3 -0
- images/00157cbe574f8576215d2dd6e08f99a497899069b2fcd7f6cf699c399bb5f99f.jpg +3 -0
- images/0017ae934a99486e126fc38a397f98540be0af40fd62357e3d5da66396583e5b.jpg +3 -0
- images/00241ef3178cf8c1efe3e95150660ed80d1096eac21dac97f040aa11d2b608ab.jpg +3 -0
- images/0027f491164f5d8b443e9d2634799e652cb3c4e34f7636fc72597ae16386d5a2.jpg +3 -0
- images/00284da8a3cae04d6a003f25f23e7600363a098cdd8f5343a26ba95de0ad0aa8.jpg +3 -0
- images/003486d634d77ac893f1b3a6b48eaf614f78b51382bf02ac8f705cc67e9d3cdd.jpg +3 -0
- images/003c962827f19761d473dd3b576882d96d7618576b7ee7254363af973b5093a7.jpg +3 -0
- images/004882fde19c8bb52f724c36e4f64ff9ce727697cedcde36048f66b90de8f894.jpg +3 -0
- images/004e44131719807f841219000e62da32c080e439ba55e2bd967171cd2d4c6618.jpg +3 -0
- images/00639001f91b0b3ce7425f47cd43f273d2532b7794e29a6fc2487318f5111d86.jpg +3 -0
- images/0088885b9a7799a4e9563462f64bdadce0a4e5dcbf85df2a1e12e84c1b95f717.jpg +3 -0
- images/0096389aa2a5e25f0544a194bd73ef8214f4f54ee2576bcc4772d07bb408fdc1.jpg +3 -0
- images/00a3cdb5c7895c771f1fe4febded829e91ba72e6a5dc77941e077f089282922f.jpg +3 -0
- images/00b2fa840a7a5da78e57dbde8a770126484585d38e740569258ae2a7466b33a4.jpg +3 -0
- images/00b5ad65d2be91743799fea2ff5742d64203efd4df43d7d368df8e9188ff3bbf.jpg +3 -0
- images/00ba15cc3eb47ef22a99c7e8060bea2194bad31d84f2399b3e7f7c49f8921f7c.jpg +3 -0
- images/00c7a24d425617def697324d1a9483ac1b3e32d510fe779e520dad6b6d4b84fa.jpg +3 -0
- images/00c7e4332668cc718b894d643fe3db4d8e448a17a47c9b1fc3bd9636db929485.jpg +3 -0
- images/00cb927e11d6320e67be1a37f54c1bda214f92121f580e4fb4295516970d811b.jpg +3 -0
- images/00d21a3bd46da646ac4c0db4bfa946b53dbd9dc84b9c8ec1f458e68e1832e125.jpg +3 -0
- images/00d4da7773334507e11107ab5382373dc3dc6f5d66d641287b4d4b679b590ada.jpg +3 -0
- images/00d73c3d736485f8756aa5528a293813521852ae5ab257d1a5f1df708ca1e65f.jpg +3 -0
- images/00e84435f7710022818693867bd60cddd00f2b97fe8014ce2145e9ba780e76a1.jpg +3 -0
- images/00f7705f1686bb0bd34ff790f11d2fd338a82ab693497531a4f26a29d43cf5b1.jpg +3 -0
- images/00f922578e2a03acde5dbbc86f1b2f17dd5022b58028a3e940d1805072ea84a7.jpg +3 -0
- images/00fa28efd5e221955b6d95cf3654ef4e9719effe01ae27a77b83730f415935a5.jpg +3 -0
- images/01031192b37aa7602d8806ba29977bd5ce14495330534386dc7dde84afbb033d.jpg +3 -0
- images/0107890f207220afcd735775587b9f11d714f8f171b6b322dd54659f328959f1.jpg +3 -0
- images/010de916bacc1e358acb1299760c623bd0db80cf4500a1cd409ec06a57697abb.jpg +3 -0
- images/011bd325d15c932a959d3b623073102b724bcfe8e05369d40ffadbe5b55e446a.jpg +3 -0
- images/0127f64539cea08c7b949eef1bcd97c0c79a485429e07a0931423712534d26ba.jpg +3 -0
- images/012e19a03595a771ec6dcc9b7fea6f6219d311070aa4dd5d6d1f4ec180408adf.jpg +3 -0
- images/012fb026a412675aaa8b517608541184cd93a238fc83c9d875aa6dfaa8bb10a8.jpg +3 -0
- images/01310dee64bc502c36589a76022035fa3d8a40e7c9b59ba43de2788cc8195642.jpg +3 -0
- images/0134a4592f1b48258458a144f9620c00bea470f81580a620e933edee5ef869ed.jpg +3 -0
- images/013557f404cb94e162e9c09bc2142af60e72d68d7c4a3ded0f0efde2e021bfbc.jpg +3 -0
- images/013bd69d391918bce3e0adb1bdcfa50a3b04ff3b1eafddb933bee021d1fde2b8.jpg +3 -0
- images/01433a38e4c2cc5b5b4385922e593ffb203ec80add0aab62b213be69ac7faa52.jpg +3 -0
- images/01438089c165a338e4c12f8e744ffee1fe64d6c6f6c85baebd71bab536beb210.jpg +3 -0
- images/014ba6bbf09579a2ef8b299e02213007bf1ddd50cc1e7762080aaeda4de5db59.jpg +3 -0
- images/014db6dfaa3d059fb03f56d283a1ec2dcf0dd49f7b271c426852545bbdae0087.jpg +3 -0
- images/014eb62430029103e398887d33a07a36a80239a43b04083dfc69ff7f63ef0611.jpg +3 -0
- images/01529ba8581934faa8a644b9b157cdccc2a91f481554804f7ddcd346b7bdd350.jpg +3 -0
- images/0153ddfbcd575c6b6190d0450864da16f352a5fd408faa97090fc5a0cf4c113d.jpg +3 -0
- images/01580f78a83286b937b9de54e57344283dcc78a894b146b705a05c59f77929e5.jpg +3 -0
- images/015d1b5546b7153c314e73b75d88e28a6bbcd69be1f512feb4a66db4eef0c9db.jpg +3 -0
.gitattributes
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.avro filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.lz4 filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.mds filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
# Audio files - uncompressed
|
| 40 |
+
*.pcm filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
*.sam filter=lfs diff=lfs merge=lfs -text
|
| 42 |
+
*.raw filter=lfs diff=lfs merge=lfs -text
|
| 43 |
+
# Audio files - compressed
|
| 44 |
+
*.aac filter=lfs diff=lfs merge=lfs -text
|
| 45 |
+
*.flac filter=lfs diff=lfs merge=lfs -text
|
| 46 |
+
*.mp3 filter=lfs diff=lfs merge=lfs -text
|
| 47 |
+
*.ogg filter=lfs diff=lfs merge=lfs -text
|
| 48 |
+
*.wav filter=lfs diff=lfs merge=lfs -text
|
| 49 |
+
# Image files - uncompressed
|
| 50 |
+
*.bmp filter=lfs diff=lfs merge=lfs -text
|
| 51 |
+
*.gif filter=lfs diff=lfs merge=lfs -text
|
| 52 |
+
*.png filter=lfs diff=lfs merge=lfs -text
|
| 53 |
+
*.tiff filter=lfs diff=lfs merge=lfs -text
|
| 54 |
+
# Image files - compressed
|
| 55 |
+
*.jpg filter=lfs diff=lfs merge=lfs -text
|
| 56 |
+
*.jpeg filter=lfs diff=lfs merge=lfs -text
|
| 57 |
+
*.webp filter=lfs diff=lfs merge=lfs -text
|
| 58 |
+
# Video files - compressed
|
| 59 |
+
*.mp4 filter=lfs diff=lfs merge=lfs -text
|
| 60 |
+
*.webm filter=lfs diff=lfs merge=lfs -text
|
| 61 |
+
train.jsonl filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,148 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
- zh
|
| 5 |
+
license: cc-by-nc-4.0
|
| 6 |
+
size_categories:
|
| 7 |
+
- 10K<n<100K
|
| 8 |
+
task_categories:
|
| 9 |
+
- question-answering
|
| 10 |
+
- image-text-to-text
|
| 11 |
+
tags:
|
| 12 |
+
- art
|
| 13 |
+
- agent
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
# CiQi-Agent
|
| 17 |
+
|
| 18 |
+
📖 [Paper](https://arxiv.org/abs/2603.28474) | 📊 [CiQi-VQA Dataset](https://huggingface.co/datasets/SII-Monument-Valley/CiQi-VQA) | 🧪 [CiQi-Bench Benchmark](https://huggingface.co/datasets/SII-Monument-Valley/CiQi-Bench) | 🤗 [CiQi-Agent Model](https://huggingface.co/SII-Monument-Valley/ciqi-agent)
|
| 19 |
+
|
| 20 |
+
---
|
| 21 |
+
|
| 22 |
+
**CiQi-Agent: Aligning Vision, Tools and Aesthetics in Multimodal Agent for Cultural Reasoning on Chinese Porcelains**
|
| 23 |
+
|
| 24 |
+
---
|
| 25 |
+
|
| 26 |
+
[Paper Link](https://arxiv.org/abs/2603.28474)
|
| 27 |
+
|
| 28 |
+
## 🎯 Overview
|
| 29 |
+
|
| 30 |
+
We present **CiQi-Agent**, a domain-specific multimodal agent for **antique Chinese porcelain connoisseurship**. The project is designed to combine **fine-grained visual perception**, **tool-augmented reasoning**, and **cultural-heritage knowledge grounding** for explainable porcelain analysis.
|
| 31 |
+
|
| 32 |
+
CiQi-Agent is built for **tool-augmented multimodal reasoning** on antique Chinese porcelains. During inference, it can inspect local visual evidence with an **image zoom-in tool**, retrieve visually similar examples with **image retrieval**, and access relevant domain knowledge with **text retrieval**, enabling more grounded and interpretable connoisseurship analysis.
|
| 33 |
+
|
| 34 |
+
Alongside the model, we release:
|
| 35 |
+
|
| 36 |
+
- **CiQi-VQA**, covering **29,596 porcelain specimens**, **51,553 images**, and **557,943 VQA pairs**
|
| 37 |
+
- **CiQi-Bench**, built from **775 porcelain specimens**, **878 images**, and **5,425 multiple-choice questions**
|
| 38 |
+
|
| 39 |
+
On CiQi-Bench, CiQi-Agent achieves **81.5%** average accuracy on multiple-choice evaluation and **66.7%** average score on free-form evaluation, outperforming strong open-source and closed-source multimodal baselines.
|
| 40 |
+
|
| 41 |
+
## 🤖 Model
|
| 42 |
+
|
| 43 |
+
- Built on **Qwen2.5-VL-7B-Instruct**
|
| 44 |
+
- Uses **tool-augmented reasoning** with zoom-in, image retrieval, and text retrieval
|
| 45 |
+
- Trained with a **two-phase supervised fine-tuning + reinforcement learning pipeline**
|
| 46 |
+
- Evaluated on **CiQi-Bench** with both **multiple-choice** and **free-form** protocols
|
| 47 |
+
- Achieves **81.5%** average accuracy on multiple-choice evaluation, exceeding **GPT-5 by 5.7 points** and the strongest listed open-source baseline **GLM-4.5V (72.6%)** by **8.9 points**
|
| 48 |
+
- Achieves **66.7%** average score on free-form evaluation, exceeding **GPT-5 by 18.7 points** and **Qwen2.5-VL-72B-Instruct by 23.7 points**
|
| 49 |
+
|
| 50 |
+
## 📊 Model Performance
|
| 51 |
+
|
| 52 |
+
### Multiple-Choice Accuracy (%) on CiQi-Bench
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
| Model | Dynasty | Reign | Kiln | Color | Motif | Shape | Naming | Average |
|
| 56 |
+
| --------------------------- | -------- | -------- | -------- | -------- | -------- | -------- | -------- | -------- |
|
| 57 |
+
| GPT-5 | 65.7 | 61.4 | 79.6 | 86.5 | 69.3 | 83.8 | 84.3 | 75.8 |
|
| 58 |
+
| GPT-4.1 | 59.3 | 68.3 | 71.1 | 85.0 | 62.2 | 81.8 | 77.9 | 72.2 |
|
| 59 |
+
| GPT-4o | 59.1 | 60.4 | 68.6 | 89.2 | 70.1 | 84.2 | 82.1 | 73.4 |
|
| 60 |
+
| o3 | 57.6 | 57.4 | 72.2 | 82.6 | 62.4 | 76.8 | 76.6 | 69.4 |
|
| 61 |
+
| Qwen2.5-VL-72B-Instruct | 57.6 | 34.7 | 69.2 | 86.7 | 71.7 | 84.1 | 80.3 | 69.2 |
|
| 62 |
+
| GLM-4.5V (106B) | 58.3 | 59.4 | 75.8 | 82.3 | 70.4 | 81.8 | 80.6 | 72.6 |
|
| 63 |
+
| InternVL3.5-241B-A28B-Flash | 57.1 | 38.6 | 59.5 | 82.1 | 64.8 | 73.9 | 68.5 | 63.5 |
|
| 64 |
+
| Kimi-VL-A3B-Instruct (16B) | 59.3 | 22.8 | 48.8 | 84.8 | 59.8 | 77.9 | 70.3 | 60.5 |
|
| 65 |
+
| **CiQi-Agent (Ours, 7B)** | **77.6** | **70.3** | **81.8** | **91.4** | **75.7** | **88.1** | **85.2** | **81.5** |
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
### Free-Form Score (%) on CiQi-Bench
|
| 69 |
+
|
| 70 |
+
|
| 71 |
+
| Model | Dynasty | Reign | Kiln | Color | Motif | Shape | Average |
|
| 72 |
+
| --------------------------- | -------- | -------- | -------- | -------- | -------- | -------- | -------- |
|
| 73 |
+
| GPT-5 | 39.4 | 32.8 | 42.6 | 74.4 | 35.3 | 63.9 | 48.0 |
|
| 74 |
+
| GPT-4.1 | 36.7 | 27.2 | 29.0 | 67.5 | 27.6 | 60.1 | 41.3 |
|
| 75 |
+
| GPT-4o | 26.9 | 13.4 | 15.1 | 53.9 | 21.1 | 47.6 | 29.7 |
|
| 76 |
+
| o3 | 42.7 | 36.6 | 44.4 | 74.2 | 33.1 | 62.1 | 48.8 |
|
| 77 |
+
| Qwen2.5-VL-72B-Instruct | 29.5 | 31.2 | 27.7 | 75.8 | 31.0 | 62.6 | 43.0 |
|
| 78 |
+
| GLM-4.5V (106B) | 31.0 | 14.3 | 32.8 | 65.4 | 31.1 | 65.2 | 39.9 |
|
| 79 |
+
| InternVL3.5-241B-A28B-Flash | 42.4 | 31.6 | 36.9 | 52.6 | 19.6 | 41.5 | 37.4 |
|
| 80 |
+
| Kimi-VL-A3B-Instruct (16B) | 17.3 | 23.7 | 16.2 | 69.5 | 26.5 | 61.3 | 35.7 |
|
| 81 |
+
| **CiQi-Agent (Ours, 7B)** | **71.3** | **49.1** | **69.8** | **85.4** | **49.7** | **75.0** | **66.7** |
|
| 82 |
+
|
| 83 |
+
|
| 84 |
+
## 📦 Dataset & Benchmark
|
| 85 |
+
|
| 86 |
+
### 📊 CiQi-VQA
|
| 87 |
+
|
| 88 |
+
**CiQi-VQA** is a large-scale dataset for porcelain-centered multimodal training.
|
| 89 |
+
|
| 90 |
+
- 29,596 porcelain specimens
|
| 91 |
+
- 51,553 images
|
| 92 |
+
- 557,943 VQA pairs
|
| 93 |
+
- 38 dynasties
|
| 94 |
+
- 42 reign periods
|
| 95 |
+
- 246 glaze color categories
|
| 96 |
+
- 248 decorative motif categories
|
| 97 |
+
- 158 vessel shape categories
|
| 98 |
+
|
| 99 |
+
Link: [https://huggingface.co/datasets/SII-Monument-Valley/CiQi-VQA](https://huggingface.co/datasets/SII-Monument-Valley/CiQi-VQA)
|
| 100 |
+
|
| 101 |
+
### 🧪 CiQi-Bench
|
| 102 |
+
|
| 103 |
+
**CiQi-Bench** is an expert-aligned benchmark for evaluating porcelain connoisseurship ability.
|
| 104 |
+
|
| 105 |
+
- 775 porcelain specimens
|
| 106 |
+
- 878 images
|
| 107 |
+
- 5,425 multiple-choice questions
|
| 108 |
+
- Free-form evaluation with attribute-wise scoring
|
| 109 |
+
|
| 110 |
+
Link: [https://huggingface.co/datasets/SII-Monument-Valley/CiQi-Bench](https://huggingface.co/datasets/SII-Monument-Valley/CiQi-Bench)
|
| 111 |
+
|
| 112 |
+
### 📈 Dataset and Benchmark Statistics
|
| 113 |
+
|
| 114 |
+
|
| 115 |
+
| Split / Resource | Porcelains | Images | VQA Questions | Multiple-Choice Questions | Attributes |
|
| 116 |
+
| --------------------- | ---------- | ------ | ------------- | ------------------------- | ----------------------------------------- |
|
| 117 |
+
| CiQi-VQA SFT | 28,821 | 50,675 | 557,168 | --- | dynasty, reign, kiln, color, motif, shape |
|
| 118 |
+
| CiQi-VQA RL subset | 10,275 | 10,275 | 10,275 | --- | dynasty, reign, kiln, color, motif, shape |
|
| 119 |
+
| CiQi-Bench Evaluation | 775 | 878 | 775 | 5,425 | dynasty, reign, kiln, color, motif, shape |
|
| 120 |
+
| Total | 29,596 | 51,553 | 557,943 | 5,425 | dynasty, reign, kiln, color, motif, shape |
|
| 121 |
+
|
| 122 |
+
|
| 123 |
+
## 📜 License
|
| 124 |
+
|
| 125 |
+
Licensing will be specified separately for the model and the dataset.
|
| 126 |
+
|
| 127 |
+
- **Model license**: TBD
|
| 128 |
+
- **Dataset license**: CC BY-NC 4.0
|
| 129 |
+
|
| 130 |
+
## 🤝 Acknowledgement
|
| 131 |
+
|
| 132 |
+
We thank **[Verl](https://github.com/volcengine/verl)** for providing an open-source reinforcement learning framework that supports this line of research.
|
| 133 |
+
|
| 134 |
+
We also thank **[DeepEyes](https://github.com/Visual-Agent/DeepEyes)** for inspiring and informing our exploration of tool-augmented multimodal reasoning.
|
| 135 |
+
|
| 136 |
+
## 📜 Citation
|
| 137 |
+
|
| 138 |
+
```bibtex
|
| 139 |
+
@misc{wang2026ciqiagentaligningvisiontools,
|
| 140 |
+
title={CiQi-Agent: Aligning Vision, Tools and Aesthetics in Multimodal Agent for Cultural Reasoning on Chinese Porcelains},
|
| 141 |
+
author={Wenhan Wang and Zhixiang Zhou and Zhongtian Ma and Yanzhu Chen and Ziyu Lin and Hao Sheng and Pengfei Liu and Honglin Ma and Wenqi Shao and Qiaosheng Zhang and Yu Qiao},
|
| 142 |
+
year={2026},
|
| 143 |
+
eprint={2603.28474},
|
| 144 |
+
archivePrefix={arXiv},
|
| 145 |
+
primaryClass={cs.CV},
|
| 146 |
+
url={https://arxiv.org/abs/2603.28474},
|
| 147 |
+
}
|
| 148 |
+
```
|
images/00044b400078a592f3c2632074373091218087ef98c3163d54ee3effdddc24ca.jpg
ADDED
|
Git LFS Details
|
images/00105ae503fb17e659e2488f22763af61776a6144a6216f92f5bcd8d98de6a46.jpg
ADDED
|
Git LFS Details
|
images/00157cbe574f8576215d2dd6e08f99a497899069b2fcd7f6cf699c399bb5f99f.jpg
ADDED
|
Git LFS Details
|
images/0017ae934a99486e126fc38a397f98540be0af40fd62357e3d5da66396583e5b.jpg
ADDED
|
Git LFS Details
|
images/00241ef3178cf8c1efe3e95150660ed80d1096eac21dac97f040aa11d2b608ab.jpg
ADDED
|
Git LFS Details
|
images/0027f491164f5d8b443e9d2634799e652cb3c4e34f7636fc72597ae16386d5a2.jpg
ADDED
|
Git LFS Details
|
images/00284da8a3cae04d6a003f25f23e7600363a098cdd8f5343a26ba95de0ad0aa8.jpg
ADDED
|
Git LFS Details
|
images/003486d634d77ac893f1b3a6b48eaf614f78b51382bf02ac8f705cc67e9d3cdd.jpg
ADDED
|
Git LFS Details
|
images/003c962827f19761d473dd3b576882d96d7618576b7ee7254363af973b5093a7.jpg
ADDED
|
Git LFS Details
|
images/004882fde19c8bb52f724c36e4f64ff9ce727697cedcde36048f66b90de8f894.jpg
ADDED
|
Git LFS Details
|
images/004e44131719807f841219000e62da32c080e439ba55e2bd967171cd2d4c6618.jpg
ADDED
|
Git LFS Details
|
images/00639001f91b0b3ce7425f47cd43f273d2532b7794e29a6fc2487318f5111d86.jpg
ADDED
|
Git LFS Details
|
images/0088885b9a7799a4e9563462f64bdadce0a4e5dcbf85df2a1e12e84c1b95f717.jpg
ADDED
|
Git LFS Details
|
images/0096389aa2a5e25f0544a194bd73ef8214f4f54ee2576bcc4772d07bb408fdc1.jpg
ADDED
|
Git LFS Details
|
images/00a3cdb5c7895c771f1fe4febded829e91ba72e6a5dc77941e077f089282922f.jpg
ADDED
|
Git LFS Details
|
images/00b2fa840a7a5da78e57dbde8a770126484585d38e740569258ae2a7466b33a4.jpg
ADDED
|
Git LFS Details
|
images/00b5ad65d2be91743799fea2ff5742d64203efd4df43d7d368df8e9188ff3bbf.jpg
ADDED
|
Git LFS Details
|
images/00ba15cc3eb47ef22a99c7e8060bea2194bad31d84f2399b3e7f7c49f8921f7c.jpg
ADDED
|
Git LFS Details
|
images/00c7a24d425617def697324d1a9483ac1b3e32d510fe779e520dad6b6d4b84fa.jpg
ADDED
|
Git LFS Details
|
images/00c7e4332668cc718b894d643fe3db4d8e448a17a47c9b1fc3bd9636db929485.jpg
ADDED
|
Git LFS Details
|
images/00cb927e11d6320e67be1a37f54c1bda214f92121f580e4fb4295516970d811b.jpg
ADDED
|
Git LFS Details
|
images/00d21a3bd46da646ac4c0db4bfa946b53dbd9dc84b9c8ec1f458e68e1832e125.jpg
ADDED
|
Git LFS Details
|
images/00d4da7773334507e11107ab5382373dc3dc6f5d66d641287b4d4b679b590ada.jpg
ADDED
|
Git LFS Details
|
images/00d73c3d736485f8756aa5528a293813521852ae5ab257d1a5f1df708ca1e65f.jpg
ADDED
|
Git LFS Details
|
images/00e84435f7710022818693867bd60cddd00f2b97fe8014ce2145e9ba780e76a1.jpg
ADDED
|
Git LFS Details
|
images/00f7705f1686bb0bd34ff790f11d2fd338a82ab693497531a4f26a29d43cf5b1.jpg
ADDED
|
Git LFS Details
|
images/00f922578e2a03acde5dbbc86f1b2f17dd5022b58028a3e940d1805072ea84a7.jpg
ADDED
|
Git LFS Details
|
images/00fa28efd5e221955b6d95cf3654ef4e9719effe01ae27a77b83730f415935a5.jpg
ADDED
|
Git LFS Details
|
images/01031192b37aa7602d8806ba29977bd5ce14495330534386dc7dde84afbb033d.jpg
ADDED
|
Git LFS Details
|
images/0107890f207220afcd735775587b9f11d714f8f171b6b322dd54659f328959f1.jpg
ADDED
|
Git LFS Details
|
images/010de916bacc1e358acb1299760c623bd0db80cf4500a1cd409ec06a57697abb.jpg
ADDED
|
Git LFS Details
|
images/011bd325d15c932a959d3b623073102b724bcfe8e05369d40ffadbe5b55e446a.jpg
ADDED
|
Git LFS Details
|
images/0127f64539cea08c7b949eef1bcd97c0c79a485429e07a0931423712534d26ba.jpg
ADDED
|
Git LFS Details
|
images/012e19a03595a771ec6dcc9b7fea6f6219d311070aa4dd5d6d1f4ec180408adf.jpg
ADDED
|
Git LFS Details
|
images/012fb026a412675aaa8b517608541184cd93a238fc83c9d875aa6dfaa8bb10a8.jpg
ADDED
|
Git LFS Details
|
images/01310dee64bc502c36589a76022035fa3d8a40e7c9b59ba43de2788cc8195642.jpg
ADDED
|
Git LFS Details
|
images/0134a4592f1b48258458a144f9620c00bea470f81580a620e933edee5ef869ed.jpg
ADDED
|
Git LFS Details
|
images/013557f404cb94e162e9c09bc2142af60e72d68d7c4a3ded0f0efde2e021bfbc.jpg
ADDED
|
Git LFS Details
|
images/013bd69d391918bce3e0adb1bdcfa50a3b04ff3b1eafddb933bee021d1fde2b8.jpg
ADDED
|
Git LFS Details
|
images/01433a38e4c2cc5b5b4385922e593ffb203ec80add0aab62b213be69ac7faa52.jpg
ADDED
|
Git LFS Details
|
images/01438089c165a338e4c12f8e744ffee1fe64d6c6f6c85baebd71bab536beb210.jpg
ADDED
|
Git LFS Details
|
images/014ba6bbf09579a2ef8b299e02213007bf1ddd50cc1e7762080aaeda4de5db59.jpg
ADDED
|
Git LFS Details
|
images/014db6dfaa3d059fb03f56d283a1ec2dcf0dd49f7b271c426852545bbdae0087.jpg
ADDED
|
Git LFS Details
|
images/014eb62430029103e398887d33a07a36a80239a43b04083dfc69ff7f63ef0611.jpg
ADDED
|
Git LFS Details
|
images/01529ba8581934faa8a644b9b157cdccc2a91f481554804f7ddcd346b7bdd350.jpg
ADDED
|
Git LFS Details
|
images/0153ddfbcd575c6b6190d0450864da16f352a5fd408faa97090fc5a0cf4c113d.jpg
ADDED
|
Git LFS Details
|
images/01580f78a83286b937b9de54e57344283dcc78a894b146b705a05c59f77929e5.jpg
ADDED
|
Git LFS Details
|
images/015d1b5546b7153c314e73b75d88e28a6bbcd69be1f512feb4a66db4eef0c9db.jpg
ADDED
|
Git LFS Details
|