T5-REF-CORRUPT-EN

pipeline_tag: text-generation language: en library_name: transformers tags:

• t5
• grammar-correction
• text-generation

---

T5-REF-CORRUPT-EN: Automatic Error Correction of Academic Referencing According to Institutional Guidelines of the Center for Translation Studies (CTS) of University of Vienna

Objective: This model corrects errors in academic referencing. For example:

Input (wrong sentence): According to Smith & Peterson 2016 56, the translation reveals patterns that suggest underlying semantic shifts

Output (clean sentence): According to Smith and Peterson (2016: 56), the translation reveals patterns that suggest underlying semantic shifts.

Model Details:

• Model name: T5-REF-CORRUPT-EN
• Base model: T5-base
• Language: English
• Training data: Synthetically generated using LLMs and synthetically corrupted real student sentences.

Usage Cases: Error correction of academic references according to CTS guidelines.

Example

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

model_id = "elizaveta-dev/T5-REF-CORRUPT-EN"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForSeq2SeqLM.from_pretrained(model_id)

text = "According to Smith & Peterson 2016 56, the translation reveals patterns that suggest underlying semantic shifts."
inputs = tokenizer(text, return_tensors="pt")

outputs = model.generate(**inputs, max_new_tokens=128)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Use-Cases

The model can perform automatic corrections of various referencing errors, including:

1. Incorrect Citation Type (Parenthetical vs. Narrative)

Example of mistake: (Lopez 2018; Chen 2012) found that cultural context strongly influences translation strategies.

Example of correction: Lopez (2018) and Chen (2012) found that cultural context strongly influences translation strategies.

Example of mistake: This topic has been widely researched Baker (2006).

Example of correction: This topic has been widely researched (Baker 2006).

---

2. Incorrect Citation for Two Authors

Example of mistake: The concept of functional equivalence was analyzed by Baker & Green (2007).

Example of correction: The concept of functional equivalence was analyzed by Baker and Green (2007).

Example of mistake: Previous research (Müller, Schmidt 2001) highlights challenges in literary translation.

Example of correction: Previous research (Müller & Schmidt 2001) highlights challenges in literary translation.

---

3. Incorrect Placement of Citations

Example of mistake: According to Williams, translation theory continues to evolve (2011: 77).

Example of correction: According to Williams (2011: 77), translation theory continues to evolve.

---

4. Redundant Entities

Example of mistake: As Lee (2009) explains, equivalence is central in translation (Lee 2009).

Example of correction: As Lee (2009) explains, equivalence is central in translation.