paper-with-me

홈 › Papers

Say It Another Way: A Framework for User-Grounded Paraphrasing

2025-05-06 · Cléa Chataigner, Rebecca Ma, Prakhar Ganesh, Afaf Taïk, Elliot Creager, Golnoosh Farnadi

Small changes in how a prompt is worded can lead to meaningful differences in the behavior of large language models (LLMs), raising concerns about the stability and reliability of their evaluations. While prior work has explored simple formatting changes, these rarely capture the kinds of natural variation seen in real-world language use. We propose a controlled paraphrasing framework based on a taxonomy of minimal linguistic transformations to systematically generate natural prompt variations. Using the BBQ dataset, we validate our method with both human annotations and automated checks, then use it to study how LLMs respond to paraphrased prompts in stereotype evaluation tasks. Our analysis shows that even subtle prompt modifications can lead to substantial changes in model behavior. These results highlight the need for robust, paraphrase-aware evaluation protocols.

📄 PDF Abstract BibTeX arXiv:2505.03563

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

iParaphrasing: Extracting Visually Grounded Paraphrases via an Image

2018-06-12 · COLING 2018 8 · Chenhui Chu, Mayu Otani, Yuta Nakashima

A paraphrase is a restatement of the meaning of a text in other words. Paraphrases have been studied to enhance the performance of many natural language processing tasks. In this paper, we propose a novel task iParaphras…

Image CaptioningQuestion AnsweringVisual Question AnsweringVisual Question Answering (VQA)

WET: Overcoming Paraphrasing Vulnerabilities in Embeddings-as-a-Service with Linear Transformation Watermarks

2024-08-29 · Anudeex Shetty, Qiongkai Xu, Jey Han Lau

Embeddings-as-a-Service (EaaS) is a service offered by large language model (LLM) developers to supply embeddings generated by LLMs. Previous research suggests that EaaS is prone to imitation attacks -- attacks that clon…

Language ModelingLanguage ModellingLarge Language Model

CookDial: A dataset for task-oriented dialogs grounded in procedural documents

2022-06-17 · Yiwei Jiang, Klim Zaporojets, Johannes Deleu, Thomas Demeester 외

This work presents a new dialog dataset, CookDial, that facilitates research on task-oriented dialog systems with procedural knowledge understanding. The corpus contains 260 human-to-human task-oriented dialogs in which …

Decision MakingResponse Generation

Navigating Connected Memories with a Task-oriented Dialog System

2022-11-15 · Seungwhan Moon, Satwik Kottur, Alborz Geramifard, Babak Damavandi

Recent years have seen an increasing trend in the volume of personal media captured by users, thanks to the advent of smartphones and smart glasses, resulting in large media collections. Despite conversation being an int…

Retrieval

Navigating Connected Memories with a Task-oriented Dialog System

2022-01-16 · ACL ARR January 2022 1 · Anonymous

Recent years have seen an increasing trend in the volume of personal media captured by users, thanks to the advent of smartphones and smart glasses, resulting in large media collections. Despite conversation being an in…

Retrieval