Summary of Mast Kalandar at Semeval-2024 Task 8: on the Trail Of Textual Origins: Roberta-bilstm Approach to Detect Ai-generated Text, by Jainit Sushil Bafna et al.
Mast Kalandar at SemEval-2024 Task 8: On the Trail of Textual Origins: RoBERTa-BiLSTM Approach to Detect AI-Generated Text
by Jainit Sushil Bafna, Hardik Mittal, Suyash Sethia, Manish Shrivastava, Radhika Mamidi
First submitted to arxiv on: 3 Jul 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary Large Language Models (LLMs) have demonstrated impressive capabilities in generating fluent responses to diverse user queries, raising concerns about potential misuse in journalism, education, and academia. To address these concerns, SemEval 2024 introduces the Multigenerator, Multidomain, and Multilingual Black-Box Machine-Generated Text Detection task, aiming to develop automated systems for identifying machine-generated text and detecting potential misuse. This paper proposes a RoBERTa-BiLSTM based classifier to classify text into AI-generated or human categories and conducts a comparative study with baseline approaches to evaluate its effectiveness. The proposed architecture ranked 46th on the official leaderboard with an accuracy of 80.83 among 125, contributing to the advancement of automatic text detection systems in addressing machine-generated text misuse. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Large language models can create realistic texts that look like they were written by humans. This is great for some things, but it also raises concerns about using these fake texts in important places like news articles or school assignments. To help solve this problem, a new task was created to develop computer programs that can identify when text was generated by a machine rather than a human. In this paper, the authors propose a way to do this using a special type of artificial intelligence called RoBERTa-BiLSTM. They also compare their approach to other methods and show how well it works. |
