Datasets:
File size: 4,276 Bytes
77b4625 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 |
# QE4PE Post-task
The goal of the `posttask` is to evaluate translators' editing speed and quality after they acquired proficiency with the GroTE interface on a fixed set of additional texts from the same distribution of `pretask` and `main`, using the `no_highlight` modality for all translators to estimate individual baseline speed. Refer to the [translators' guidelines](./posttask_guidelines.pdf) for additional details about the task.
## Folder Structure
```shell
posttask/
├── inputs/
│ ├── eng-ita/
│ │ ├── posttask_eng-ita_doc1_input.txt
│ │ ├── posttask_eng-ita_doc2_input.txt
│ │ └── ... # GroTE input files with tags and ||| source-target separator
│ └── eng-nld/
│ │ ├── posttask_eng-nld_doc1_input.txt
│ │ ├── posttask_eng-nld_doc2_input.txt
│ │ └── ... # GroTE input files with tags and ||| source-target separator
├── outputs/
│ ├── eng-ita/
│ │ ├── logs/
│ │ │ ├── posttask_eng-ita_oracle_t1_logs.csv
│ │ │ └── ... # GroTE logs for every translator (e.g. oracle_t1, uses main task IDs)
│ │ ├── metrics/
│ │ │ ├── posttask_eng-ita_oracle_t1_metrics.csv
│ │ │ └── ... # Metrics for every translator (e.g. oracle_t1, uses main task IDs)
│ │ ├── posttask_eng-ita_doc1_oracle_t1_output.txt
│ │ └── ... # GroTE output files (one edited target per line)
│ └── eng-nld/
│ ├── logs/
│ │ ├── posttask_eng-nld_oracle_t1_logs.csv
│ │ └── ... # GroTE logs for every translator (e.g. oracle_t1, uses main task IDs)
│ ├── metrics/
│ │ ├── posttask_eng-nld_oracle_t1_metrics.csv
│ │ └── ... # Metrics for every translator (e.g. oracle_t1, uses main task IDs)
│ ├── example_eng-nld_doc1_oracle_t1_output.txt
│ └── ... # GroTE output files (one post-edited segment per line)
├── doc_id_map.json # Source and doc name maps
├── posttask_guidelines.pdf # Task guidelines for translators
└── README.md
```
## Important Notes about Data Issues
Translator `unsupervised_t1` for the `eng-ita` direction could not complete the `posttask`, and is hence not present in the data.
## Inputs
The `posttask` uses eight documents containing between 3-8 contiguous segments from the same `wmt23` collection of the `main` task that were matching all the `main` task requirements, but were not selected for the main collection. [doc_id_map.json](./doc_id_map.json) shows the document assignments from the original collection.
Documents were edited in the `no_highlight` setting by all translators to provide a baseline for speed/quality measurements in the main task. Input files for the task have names using the format:
```python
"{{TASK_ID}}_{{TRANSLATION_DIRECTION}}_{{DOC_ID}}_input.txt"
```
## Outputs
Files in `outputs/eng-ita` and `outputs/eng-nld` contain post-edited outputs (one per line, matching inputs) using format:
```python
"{{TASK_ID}}_{{TRANSLATION_DIRECTION}}_{{DOC_ID}}_{{TRANSLATOR_ID}}_output.txt"
```
> **Note**: `{{TRANSLATOR_ID}}` includes the `{{HIGHLIGHT_MODALITY}}` information matching the ID from the `main` task. Despite this, **all editing for `posttask` was performed in the `no_highlight` setting**.
The contents of `outputs/{{TRANSLATION_DIRECTION}}/logs` can be used to analyze editing behavior at a granular scale. Each log file has format:
```python
"{{TASK_ID}}_{{TRANSLATION_DIRECTION}}_{{TRANSLATOR_ID}}_logs.csv"
```
Refer to [GroTE](https://github.com/gsarti/grote) documentation for more information about the logs.
The contents of `outputs/{{TRANSLATION_DIRECTION}}/metrics` contain aggregated metrics for each translator, with name format:
```python
"{{TASK_ID}}_{{TRANSLATION_DIRECTION}}_{{TRANSLATOR_ID}}_metrics.csv"
```
Metrics include:
1. max/min/mean/std of BLEU, chrF, TER and COMET scores computed between the MT or the PE texts and all (other, in case of PE) post-edited variants of the same segment for `{{TRANSLATION_DIRECTION}}`.
2. XCOMET-XXL segment-level QE scores and errors for the MT and PE texts.
|