wushengqiong
commited on
Commit
Β·
fa15ad2
1
Parent(s):
f541f38
update readme
Browse files
README.md
CHANGED
@@ -36,15 +36,10 @@ Please download the dataset accordingly.
|
|
36 |
|
37 |
## π Table of Contents
|
38 |
|
39 |
-
- [β¨
|
40 |
-
- [
|
41 |
-
- [
|
42 |
-
|
43 |
-
- [πΌοΈ Image Task Taxonomy](#imagetaxonomy)
|
44 |
-
- [π½οΈ Video Task Taxonomy](#videotaxonomy)
|
45 |
-
- [π Audio Task Taxonomy](#audiotaxonomy)
|
46 |
-
- [π 3D Task Taxonomy](#3dtaxonomy)
|
47 |
-
- [π Language Task Taxonomy](#languagetaxonomy)
|
48 |
|
49 |
|
50 |
|
@@ -52,68 +47,22 @@ Please download the dataset accordingly.
|
|
52 |
|
53 |
---
|
54 |
|
55 |
-
<span id='
|
56 |
|
57 |
-
# β¨β¨β¨ **
|
58 |
|
59 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
60 |
|
61 |
```
|
62 |
-
|
63 |
-
βββ Image
|
64 |
-
β βββ comprehension
|
65 |
-
β β βββ Bird-Detection
|
66 |
-
β β β βββ annotation.json
|
67 |
-
β β β βββ images
|
68 |
-
β β β βββ Acadian_Flycatcher_0070_29150.jpg
|
69 |
-
β β βββ Bottle-Anomaly-Detection
|
70 |
-
β β β βββ annotation.json
|
71 |
-
β β β βββ images
|
72 |
-
β β βββ ...
|
73 |
-
β βββ generation
|
74 |
-
β βββ Layout-to-Face-Image-Generation
|
75 |
-
β βββ annotation.json
|
76 |
-
β βββ images
|
77 |
-
β βββ ...
|
78 |
-
βββ Video
|
79 |
-
β βββ comprehension
|
80 |
-
β β βββ Human-Object-Interaction-Video-Captioning
|
81 |
-
β β βββ annotation.json
|
82 |
-
β β βββ videos
|
83 |
-
β β βββ ...
|
84 |
-
β βββ generation
|
85 |
-
β βββ Scene-Image-to-Video-Generation
|
86 |
-
β βββ annotation.json
|
87 |
-
β βββ videos
|
88 |
-
β βββ ...
|
89 |
-
βββ 3d
|
90 |
-
β βββ comprehension
|
91 |
-
β β βββ 3D-Furniture-Classification
|
92 |
-
β β βββ annotation.json
|
93 |
-
β β βββ pointclouds
|
94 |
-
β β βββ ...
|
95 |
-
β βββ generation
|
96 |
-
β βββ Text-to-3D-Living-and-Arts-Point-Cloud-Generation
|
97 |
-
β βββ annotation.json
|
98 |
-
β βββ pointclouds
|
99 |
-
β βββ ...
|
100 |
-
βββ Audio
|
101 |
-
β βββ comprehension
|
102 |
-
β β βββ Accent-Classification
|
103 |
-
β β βββ annotation.json
|
104 |
-
β β βββ audios
|
105 |
-
β β βββ ...
|
106 |
-
β βββ generation
|
107 |
-
β βββ Video-To-Audio
|
108 |
-
β βββ annotation.json
|
109 |
-
β βββ audios
|
110 |
-
β βββ ...
|
111 |
-
βββ NLP
|
112 |
-
β βββ History-Question-Answering
|
113 |
-
β β βββ annotation.json
|
114 |
-
β βββ Abstractive-Summarization
|
115 |
-
β β βββ annotation.json
|
116 |
-
β βββ ...
|
117 |
|
118 |
```
|
119 |
|
@@ -124,13 +73,6 @@ An illustrative example of file formats:
|
|
124 |

|
125 |
|
126 |
|
127 |
-
<span id='usage'/>
|
128 |
-
|
129 |
-
## πππ Usage
|
130 |
-
|
131 |
-
Please download all the files in this repository. We also provide overview.json, which is an example of the format of our dataset.
|
132 |
-
|
133 |
-
For more instructions, please go to the [document page](https://generalist.top/document).
|
134 |
|
135 |
|
136 |
---
|
@@ -139,105 +81,166 @@ For more instructions, please go to the [document page](https://generalist.top/d
|
|
139 |
|
140 |
|
141 |
|
142 |
-
<span id='
|
143 |
|
144 |
|
145 |
|
146 |
-
# πππ
|
147 |
|
148 |
|
149 |
|
150 |
|
151 |
-
|
152 |
-
|
153 |
-
<div align="center">
|
154 |
-
<img src='https://cdn-uploads.huggingface.co/production/uploads/647773a1168cb428e00e9a8f/d4TIWw3rlWuxpBCEpHYJB.jpeg'>
|
155 |
-
<p> Overview of General-Bench, which covers 145 skills for more than 700 tasks with over 325,800 samples under
|
156 |
-
comprehension and generation categories in various modalities</p>
|
157 |
-
</div>
|
158 |
|
|
|
|
|
|
|
159 |
|
160 |
-
<span id='distribution'/>
|
161 |
|
162 |
-
|
|
|
|
|
163 |
|
164 |
-
<div align="center">
|
165 |
-
<img src='https://cdn-uploads.huggingface.co/production/uploads/64c139d867eff857ea51caa8/fF3iH95B3QEBvJYwqzZVG.png'>
|
166 |
-
<p> Distribution of various capabilities evaluated in General-Bench.</p>
|
167 |
-
</div>
|
168 |
|
|
|
|
|
169 |
|
170 |
-
|
171 |
-
|
172 |
-
|
173 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
174 |
|
|
|
175 |
|
176 |
|
177 |
|
|
|
178 |
|
179 |
-
<span id='imagetaxonomy'/>
|
180 |
|
181 |
-
# πΌοΈ Image Task Taxonomy
|
182 |
-
<div align="center">
|
183 |
-
<img src='https://cdn-uploads.huggingface.co/production/uploads/64c139d867eff857ea51caa8/2QYihQRhZ5C9K5IbukY7R.png'>
|
184 |
-
<p>Taxonomy and hierarchy of data in terms of Image modality.</p>
|
185 |
-
</div>
|
186 |
|
187 |
|
188 |
|
|
|
189 |
|
190 |
-
|
191 |
|
192 |
-
|
193 |
|
194 |
-
|
195 |
-
|
196 |
-
|
197 |
-
</div>
|
198 |
|
199 |
|
|
|
|
|
|
|
200 |
|
201 |
|
202 |
|
203 |
|
|
|
|
|
204 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
205 |
|
|
|
206 |
|
207 |
-
|
208 |
|
209 |
-
# π Audio Task Taxonomy
|
210 |
|
|
|
211 |
|
|
|
212 |
|
213 |
-
|
214 |
-
<img src='https://cdn-uploads.huggingface.co/production/uploads/64c139d867eff857ea51caa8/e-QBvBjeZy8vmcBjAB0PE.png'>
|
215 |
-
<p>Taxonomy and hierarchy of data in terms of Audio modality.</p>
|
216 |
-
</div>
|
217 |
|
|
|
|
|
|
|
218 |
|
219 |
|
220 |
-
|
|
|
|
|
|
|
221 |
|
222 |
-
# π 3D Task Taxonomy
|
223 |
|
224 |
|
225 |
-
<div align="center">
|
226 |
-
<img src='https://cdn-uploads.huggingface.co/production/uploads/64c139d867eff857ea51caa8/EBXb-wyve14ExoLCgrpDK.png'>
|
227 |
-
<p>Taxonomy and hierarchy of data in terms of 3D modality.</p>
|
228 |
-
</div>
|
229 |
|
230 |
|
|
|
|
|
|
|
|
|
231 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
232 |
|
233 |
-
|
234 |
|
235 |
-
# π Language Task Taxonomy
|
236 |
|
237 |
-
<div align="center">
|
238 |
-
<img src='https://cdn-uploads.huggingface.co/production/uploads/64c139d867eff857ea51caa8/FLfk3QGdYb2sgorKTj_LT.png'>
|
239 |
-
<p>Taxonomy and hierarchy of data in terms of Language modality.</p>
|
240 |
-
</div>
|
241 |
|
242 |
|
243 |
|
|
|
36 |
|
37 |
## π Table of Contents
|
38 |
|
39 |
+
- [β¨ Scope-A](#scope_a)
|
40 |
+
- [π Scope-B](#scope_b)
|
41 |
+
- [πΌοΈ Scope-C](#scope_c)
|
42 |
+
- [π½οΈ Scope-D](#scope_d)
|
|
|
|
|
|
|
|
|
|
|
43 |
|
44 |
|
45 |
|
|
|
47 |
|
48 |
---
|
49 |
|
50 |
+
<span id='scope_a'/>
|
51 |
|
52 |
+
# β¨β¨β¨ **Scope-A**
|
53 |
|
54 |
+
Full-spectrum leaderboard covering all modalities and tasks under General-Level, for highly capable, general-purpose multimodal models.
|
55 |
+
|
56 |
+
- π Details:
|
57 |
+
- βοΈ Covers all General-Level tasks and modalities.
|
58 |
+
- βοΈ Most challenging track; requires high model capacity and resource commitment.
|
59 |
+
|
60 |
+
- π Highlights:
|
61 |
+
- βοΈ Evaluates holistic generalization and cross-modal synergy.
|
62 |
+
- βοΈ Suitable for near-AGI or foundation-level multimodal generalists.
|
63 |
|
64 |
```
|
65 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
66 |
|
67 |
```
|
68 |
|
|
|
73 |

|
74 |
|
75 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
76 |
|
77 |
|
78 |
---
|
|
|
81 |
|
82 |
|
83 |
|
84 |
+
<span id='scope_b'/>
|
85 |
|
86 |
|
87 |
|
88 |
+
# πππ Scope-B
|
89 |
|
90 |
|
91 |
|
92 |
|
93 |
+
Modality-specific leaderboards focusing on single modality or partially joint modality (e.g., image, video, audio, 3D) for modality-wise generalists.
|
|
|
|
|
|
|
|
|
|
|
|
|
94 |
|
95 |
+
- π Details:
|
96 |
+
- βοΈ 7 separate leaderboards (4 single modality + 3 combined modality).
|
97 |
+
- βοΈ Focuses on mastering diverse tasks within a single modality.
|
98 |
|
|
|
99 |
|
100 |
+
- π Highlights:
|
101 |
+
- βοΈ Measures within-modality generalization.
|
102 |
+
- βοΈ Suited for intermediate-level models with cross-task transferability.
|
103 |
|
|
|
|
|
|
|
|
|
104 |
|
105 |
+
In [Scope-B](https://huggingface.co/datasets/General-Level/General-Bench-Closeset-Scoped/tree/main/Scope-B), we provide the subset of data corresponding to each sub-leaderboard. Each task is represented by a separate JSON file, which specifies the dataset associated with that particular sub-leaderboard, including the relevant file names.
|
106 |
+
All referenced data files can be found in the [π **`Close Set`**](https://huggingface.co/datasets/General-Level/General-Bench-Closeset) repository.
|
107 |
|
108 |
+
```
|
109 |
+
{
|
110 |
+
## paradigm
|
111 |
+
"comprehension": {
|
112 |
+
## skill name
|
113 |
+
"Speech Accent Understanding":
|
114 |
+
[
|
115 |
+
{
|
116 |
+
## task nameοΌdata file name in Closeset
|
117 |
+
"Accent Classification": "AccentClassification"
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"Accent Sex Classification": "AccentSexClassification"
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"Speaker Identification": "SpeakerIdentification"
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"Vocal Sound Classification": "VocalSoundClassification"
|
127 |
+
}
|
128 |
+
],
|
129 |
+
...
|
130 |
+
}
|
131 |
+
}
|
132 |
|
133 |
+
```
|
134 |
|
135 |
|
136 |
|
137 |
+
----
|
138 |
|
|
|
139 |
|
|
|
|
|
|
|
|
|
|
|
140 |
|
141 |
|
142 |
|
143 |
+
<span id='scope_c'/>
|
144 |
|
145 |
+
# πΌοΈπΌοΈπΌοΈ Scope-C
|
146 |
|
147 |
+
Leaderboards categorized by comprehension vs. generation tasks within each modality. Lower entry barrier for early-stage or lightweight models.
|
148 |
|
149 |
+
- π Details:
|
150 |
+
- βοΈ 8 leaderboards: 2 Γ 4 for multimodal comprehension/generation under different modalities.
|
151 |
+
- βοΈ Supports entry-level model evaluation or teams with limited resources.
|
|
|
152 |
|
153 |
|
154 |
+
- π Highlights:
|
155 |
+
- βοΈ Assesses task-type specialization: understanding or generation.
|
156 |
+
- βοΈ Reflects generalization across task types.
|
157 |
|
158 |
|
159 |
|
160 |
|
161 |
+
In [Scope-C](https://huggingface.co/datasets/General-Level/General-Bench-Closeset-Scoped/tree/main/Scope-C), we provide the subset of data corresponding to each sub-leaderboard. Each task is represented by a separate JSON file, which specifies the dataset associated with that particular sub-leaderboard, including the relevant file names.
|
162 |
+
All referenced data files can be found in the [π **`Close Set`**](https://huggingface.co/datasets/General-Level/General-Bench-Closeset) repository.
|
163 |
|
164 |
+
```
|
165 |
+
{
|
166 |
+
## skill name
|
167 |
+
"3D Human-related Object Classification": [
|
168 |
+
{
|
169 |
+
## task nameοΌdata file name in Closeset
|
170 |
+
"3D Accessory Classification": "3d_classification/ModelNet40/accessory"
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"3D Appliance Classification": "3d_classification/ModelNet40/appliance"
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"3D Tableware Classification": "3d_classification/ModelNet40/tableware"
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"3D Musical Instrument Classification": "3d_classification/ModelNet40/musical_instrument"
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"3D Person Classification": "3d_classification/ModelNet40/person"
|
183 |
+
}
|
184 |
+
],
|
185 |
+
...
|
186 |
+
}
|
187 |
|
188 |
+
```
|
189 |
|
190 |
+
----
|
191 |
|
|
|
192 |
|
193 |
+
<span id='scope_d'/>
|
194 |
|
195 |
+
# π½οΈπ½οΈπ½οΈ Scope-D
|
196 |
|
197 |
+
Fine-grained leaderboards focused on specific task clusters (e.g., VQA, Captioning, Speech Recognition), ideal for partial generalists.
|
|
|
|
|
|
|
198 |
|
199 |
+
- π Details:
|
200 |
+
- βοΈ Large number of sub-leaderboards, each scoped to a skill set.
|
201 |
+
- βοΈ Easiest to participate; lowest cost.
|
202 |
|
203 |
|
204 |
+
- π Highlights:
|
205 |
+
- βοΈ Evaluates fine-grained skill performance.
|
206 |
+
- βοΈ Helps identify model strengths and specialization areas.
|
207 |
+
- βοΈ Encourages progressive development toward broader leaderboard participation.
|
208 |
|
|
|
209 |
|
210 |
|
|
|
|
|
|
|
|
|
211 |
|
212 |
|
213 |
+
In [Scope-D](https://huggingface.co/datasets/General-Level/General-Bench-Closeset-Scoped/tree/main/Scope-D), we provide subset datasets corresponding to each sub-leaderboard. Each task is represented by a JSON file named in the format:
|
214 |
+
`{modality name}ββ{comp/gen}_{clustered skill name}.json`.
|
215 |
+
Each JSON file specifies the dataset used for the corresponding sub-leaderboard task, including the list of relevant file names.
|
216 |
+
All referenced data files can be found in the [π **`Close Set`**](https://huggingface.co/datasets/General-Level/General-Bench-Closeset) repository.
|
217 |
|
218 |
+
```
|
219 |
+
{
|
220 |
+
## clusered skill name
|
221 |
+
"Classifcation": {
|
222 |
+
## skill name
|
223 |
+
"3D Human-related Object Classification": [
|
224 |
+
"3d_classification/ModelNet40/accessory", ## data file name in Closeset
|
225 |
+
"3d_classification/ModelNet40/appliance",
|
226 |
+
"3d_classification/ModelNet40/tableware",
|
227 |
+
"3d_classification/ModelNet40/musical_instrument",
|
228 |
+
"3d_classification/ModelNet40/person"
|
229 |
+
],
|
230 |
+
"3D Structure and Environment Classification": [
|
231 |
+
"3d_classification/ModelNet40/furniture",
|
232 |
+
"3d_classification/ModelNet40/structure"
|
233 |
+
],
|
234 |
+
"Transportation and Technology Object Classification": [
|
235 |
+
"3d_classification/ModelNet40/electronic",
|
236 |
+
"3d_classification/ModelNet40/vehicle"
|
237 |
+
]
|
238 |
+
}
|
239 |
+
}
|
240 |
|
241 |
+
```
|
242 |
|
|
|
243 |
|
|
|
|
|
|
|
|
|
244 |
|
245 |
|
246 |
|