paper-with-me

Papers

PPTC-R benchmark: Towards Evaluating the Robustness of Large Language Models for PowerPoint Task Completion

2024-03-06 · Zekai Zhang, Yiduo Guo, Yaobo Liang, Dongyan Zhao, Nan Duan

The growing dependence on Large Language Models (LLMs) for finishing user instructions necessitates a comprehensive understanding of their robustness to complex task completion in real-world situations. To address this critical need, we propose the PowerPoint Task Completion Robustness benchmark (PPTC-R) to measure LLMs' robustness to the user PPT task instruction and software version. Specifically, we construct adversarial user instructions by attacking user instructions at sentence, semantic, and multi-language levels. To assess the robustness of Language Models to software versions, we vary the number of provided APIs to simulate both the newest version and earlier version settings. Subsequently, we test 3 closed-source and 4 open-source LLMs using a benchmark that incorporates these robustness settings, aiming to evaluate how deviations impact LLMs' API calls for task completion. We find that GPT-4 exhibits the highest performance and strong robustness in our benchmark, particularly in the version update and the multilingual settings. However, we find that all LLMs lose their robustness when confronted with multiple challenges (e.g., multi-turn) simultaneously, leading to significant performance drops. We further analyze the robustness behavior and error reasons of LLMs in our benchmark, which provide valuable insights for researchers to understand the LLM's robustness in task completion and develop more robust LLMs and agents. We release the code and data at \url{https://github.com/ZekaiGalaxy/PPTCR}.

📄 PDF Abstract BibTeX arXiv:2403.03788

Code (1)

zekaigalaxy/pptcr 공식 구현

Tasks

Sentence

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Adam 설명 없음

Similar Papers 제목 키워드 기반

PPTC Benchmark: Evaluating Large Language Models for PowerPoint Task Completion

2023-11-03 · Yiduo Guo, Zekai Zhang, Yaobo Liang, Dongyan Zhao 외

Recent evaluations of Large Language Models (LLMs) have centered around testing their zero-shot/few-shot capabilities for basic natural language tasks and their ability to translate instructions into tool APIs. However, …

MVI-Bench: A Comprehensive Benchmark for Evaluating Robustness to Misleading Visual Inputs in LVLMs

2025-11-18 · Huiyi Chen, Jiawei Peng, Dehai Min, Changchang Sun 외 arxiv

Evaluating the robustness of Large Vision-Language Models (LVLMs) is essential for their continued development and responsible deployment in real-world applications. However, existing robustness benchmarks typically focu…

GLUE-X: Evaluating Natural Language Understanding Models from an Out-of-distribution Generalization Perspective

2022-11-15 · Linyi Yang, Shuibai Zhang, Libo Qin, Yafu Li 외

Pre-trained language models (PLMs) are known to improve the generalization performance of natural language understanding models by leveraging large amounts of data during the pre-training phase. However, the out-of-distr…

Natural Language UnderstandingOut-of-Distribution Generalization

Evaluating Robustness of Large Language Models in Enterprise Applications: Benchmarks for Perturbation Consistency Across Formats and Languages

2026-01-09 · Tara Bogavelli, Oluwanifemi Bamgbose, Gabrielle Gauthier Melançon, Fanny Riols 외 arxiv

Enterprise LLM applications require consistently high quality and reliable performance across diverse scenarios, demanding robustness to minor variations. Existing research shows that even small prompt changes can lead t…

SelfPrompt: Autonomously Evaluating LLM Robustness via Domain-Constrained Knowledge Guidelines and Refined Adversarial Prompts

2024-12-01 · Aihua Pei, Zehua Yang, Shunan Zhu, Ruoxi Cheng 외

Traditional methods for evaluating the robustness of large language models (LLMs) often rely on standardized benchmarks, which can escalate costs and limit evaluations across varied domains. This paper introduces a novel…

DescriptiveKnowledge Graphs