Datasets:

Modalities:
Text
Formats:
json
Languages:
English
ArXiv:
Libraries:
Datasets
Dask
License:
lim142857 commited on
Commit
85fff14
Β·
1 Parent(s): 3126e6b

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +75 -12
README.md CHANGED
@@ -42,21 +42,52 @@ configs:
42
  path: "qrels/test/*.txt"
43
  ---
44
 
45
- ## UniIR: Training and Benchmarking Universal Multimodal Information Retrievers
 
46
 
47
- Project Page: [https://tiger-ai-lab.github.io/UniIR/](https://tiger-ai-lab.github.io/UniIR/)
48
 
49
- Paper: [https://arxiv.org/pdf/2311.17136.pdf](https://arxiv.org/pdf/2311.17136.pdf)
50
 
51
- Code: [https://github.com/TIGER-AI-Lab/UniIR](https://github.com/TIGER-AI-Lab/UniIR)
 
 
52
 
53
 
54
  ## **Dataset Structure Overview**
55
- M-BEIR dataset comprises two main components: Query Data and Candidate Pool.
56
- Each of these sections consists of structured entries in JSONL format (JSON Lines), meaning each line is a valid JSON object. Below is a detailed breakdown of the components and their respective fields:
57
 
58
- Query Data (JSONL File)
59
- Each line in the Query Data file represents a unique query. The structure of each query JSON object is as follows::
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
60
  ```json
61
  {
62
  "qid": "A unique identifier formatted as {dataset_id}:{query_id}",
@@ -79,8 +110,30 @@ Each line in the Query Data file represents a unique query. The structure of eac
79
  }
80
  ```
81
 
82
- Candidate Pool (JSONL File)
83
- The Candidate Pool contains potential matching documents for the queries. The structure of each candidate JSON object in this file is as follows::
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
84
  ```json
85
  {
86
  "did": "A unique identifier for the document, formatted as {dataset_id}:{doc_id}",
@@ -91,9 +144,16 @@ The Candidate Pool contains potential matching documents for the queries. The st
91
  }
92
  ```
93
 
 
 
 
 
 
 
94
  ## **How to Use**
95
  ### Downloading the M-BEIR Dataset
96
- Download the dataset files directly from the page.
 
97
 
98
  ### Decompressing M-BEIR Images
99
  After downloading, you will need to decompress the image files. Follow these steps in your terminal:
@@ -107,9 +167,12 @@ sh -c 'cat mbeir_images.tar.gz.part-00 mbeir_images.tar.gz.part-01 mbeir_images.
107
  # Extract the images from the tar.gz file
108
  tar -xzf mbeir_images.tar.gz
109
  ```
 
110
 
111
- ## **Citation**
 
112
 
 
113
  Please cite our paper if you use our data, model or code.
114
 
115
  ```
 
42
  path: "qrels/test/*.txt"
43
  ---
44
 
45
+ ### **UniIR: Training and Benchmarking Universal Multimodal Information Retrievers**
46
+ [**🌐 Homepage**](https://tiger-ai-lab.github.io/UniIR/) | [**πŸ€— Paper**](https://huggingface.co/papers/2311.17136) | [**πŸ“– arXiv**](https://arxiv.org/pdf/2311.17136.pdf) | [**GitHub**](https://github.com/TIGER-AI-Lab/UniIR)
47
 
 
48
 
49
+ ## **Dataset Summary**
50
 
51
+ M-BEIR, the **M**ultimodal **BE**nchmark for **I**nstructed **R**etrieval, is a comprehensive large-scale retrieval benchmark designed to train and evaluate unified multimodal retrieval models (**UniIR models**).
52
+ The M-BEIR benchmark comprises eight multimodal retrieval tasks and ten datasets from a variety of domains and sources.
53
+ Each task is accompanied by human-authored instructions, encompassing 1.5 million queries and a pool of 5.6 million retrieval candidates in total.
54
 
55
 
56
  ## **Dataset Structure Overview**
57
+ The M-BEIR dataset is structured into five primary components: Query Data, Candidate Pool, Instructions, Qrels, and Images.
 
58
 
59
+ ### Query Data
60
+
61
+ Below is the directory structure for the query data:
62
+ ```
63
+ query/
64
+ β”‚
65
+ β”œβ”€β”€ train/
66
+ β”‚ β”œβ”€β”€ mbeir_cirr_train.jsonl
67
+ β”‚ β”œβ”€β”€ mbeir_edis_train.jsonl
68
+ β”‚ ...
69
+ β”œβ”€β”€ union_train/
70
+ β”‚ └── mbeir_union_up_train.jsonl
71
+ β”œβ”€β”€ val/
72
+ β”‚ β”œβ”€β”€ mbeir_visualnews_task0_val.jsonl
73
+ β”‚ β”œβ”€β”€ mbeir_visualnews_task3_val.jsonl
74
+ β”‚ ...
75
+ └── test/
76
+ β”œβ”€β”€ mbeir_visualnews_task0_test.jsonl
77
+ β”œβ”€β”€ mbeir_visualnews_task3_test.jsonl
78
+ ...
79
+ ```
80
+ `train`: Contains all the training data from 8 different datasets formatted in the M-BEIR style.
81
+
82
+ `mbeir_union_up_train.jsonl`: This file is the default training data for in-batch contrastive training specifically designed for UniIR models.
83
+ It aggregates all the data from the train directory and datasets with relatively smaller sizes have been upsampled to balance the training process.
84
+
85
+ `val`: Contains separate files for validation queries, organized by task.
86
+
87
+ `test`: Contains separate files for test queries, organized by task.
88
+
89
+ Every M-BEIR query instance has at least one positive candidate data and possibly no negative candidate data
90
+ Each line in a Query Data file represents a unique query. The structure of each query JSON object is as follows::
91
  ```json
92
  {
93
  "qid": "A unique identifier formatted as {dataset_id}:{query_id}",
 
110
  }
111
  ```
112
 
113
+ ### Candidate Pool
114
+ The Candidate Pool contains potential matching documents for the queries.
115
+
116
+ #### M-BEIR_5.6M
117
+ Within the global directory, the default retrieval setting requires models to retrieve positive candidates from a heterogeneous pool encompassing various modalities and domains.
118
+ The M-BEIR's global candidate pool, comprising 5.6 million candidates, includes the retrieval corpus from all tasks and datasets.
119
+
120
+ #### M-BEIR_local
121
+ Within the local directory, we provide dataset-task-specific pool as M-BEIR_local. Dataset-task-specific pool contains homogeneous candidates that originate from by the original dataset.
122
+
123
+ Below is the directory structure for the candidate pool:
124
+ ```
125
+ cand_pool/
126
+ β”‚
127
+ β”œβ”€β”€ global/
128
+ β”‚ β”œβ”€β”€ mbeir_union_val_cand_pool.jsonl
129
+ β”‚ └──mbeir_union_test_cand_pool.jsonl
130
+ β”‚
131
+ └── local/
132
+ β”œβ”€β”€ mbeir_visualnews_task0_cand_pool.jsonl
133
+ β”œβ”€β”€ mbeir_visualnews_task3_cand_pool.jsonl
134
+ ...
135
+ ```
136
+ The structure of each candidate JSON object in cand_pool file is as follows::
137
  ```json
138
  {
139
  "did": "A unique identifier for the document, formatted as {dataset_id}:{doc_id}",
 
144
  }
145
  ```
146
 
147
+ ### Instructions
148
+ `query_instructions.tsv` contains human-authorized instructions within the UniIR framework. Each task is accompanied by four human-authored instructions.
149
+
150
+ ### Qrels
151
+ Within the `qrels` directory, you will find qrels for both the validation and test sets. These files serve the purpose of evaluating UniIR models.
152
+
153
  ## **How to Use**
154
  ### Downloading the M-BEIR Dataset
155
+ Clone the M-BEIR repo from the current Page.
156
+ Ensure that Git LFS (Large File Storage) is installed on your system, as it will download the required data files.
157
 
158
  ### Decompressing M-BEIR Images
159
  After downloading, you will need to decompress the image files. Follow these steps in your terminal:
 
167
  # Extract the images from the tar.gz file
168
  tar -xzf mbeir_images.tar.gz
169
  ```
170
+ Now, you are ready to use the M-BEIR benchmark.
171
 
172
+ ### Dataloader and Evaluation Pipeline
173
+ We offer a dedicated dataloader and evaluation pipeline for the M-BEIR benchmark. Please refer to [**GitHub Repo**](https://github.com/TIGER-AI-Lab/UniIR) for detailed information.
174
 
175
+ ## **Citation**
176
  Please cite our paper if you use our data, model or code.
177
 
178
  ```