REVIEW 6 cited by
On the Robustness of ChatGPT: An Adversarial and Out-of-distribution Perspective
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
ChatGPT is a recent chatbot service released by OpenAI and is receiving increasing attention over the past few months. While evaluations of various aspects of ChatGPT have been done, its robustness, i.e., the performance to unexpected inputs, is still unclear to the public. Robustness is of particular concern in responsible AI, especially for safety-critical applications. In this paper, we conduct a thorough evaluation of the robustness of ChatGPT from the adversarial and out-of-distribution (OOD) perspective. To do so, we employ the AdvGLUE and ANLI benchmarks to assess adversarial robustness and the Flipkart review and DDXPlus medical diagnosis datasets for OOD evaluation. We select several popular foundation models as baselines. Results show that ChatGPT shows consistent advantages on most adversarial and OOD classification and translation tasks. However, the absolute performance is far from perfection, which suggests that adversarial and OOD robustness remains a significant threat to foundation models. Moreover, ChatGPT shows astounding performance in understanding dialogue-related texts and we find that it tends to provide informal suggestions for medical tasks instead of definitive answers. Finally, we present in-depth discussions of possible research directions.
Forward citations
Cited by 6 Pith papers
-
Fisher Random Walk: Automatic Debiasing Contextual Preference Inference for Large Language Model Evaluation
A Fisher random walk weighted residual estimator achieves semiparametric efficient confidence intervals for contextual Bradley-Terry-Luce preference comparisons with flexible score estimators.
-
Would You Walk to the Car Wash? Revealing the Salience Bias of Large Language Models in Commonsense Reasoning
LLMs' failures on physically impossible, number-laden tasks are mostly knowledge suppression by salient distractors, not missing commons sense, and light prompting largely fixes them.
-
ASSURE: Metamorphic Testing for AI-powered Browser Extensions
A modular metamorphic testing framework for LLM-based browser extensions reports 531 automatically detected issues across six real-world extensions.
-
Evaluating LLMs Robustness in Less Resourced Languages with Proxy Models
Important words chosen by a small proxy model, when perturbed with typos or spacing errors, push Bielik, Mistral-7B, and Llama-3.1-8B to wrong answers on Polish classification tasks more often than random edits.
-
Evaluating and Improving Robustness in Large Language Models: A Survey and Future Directions
LLM robustness research is organized into adversarial robustness, out-of-distribution robustness, and evaluation, with an accompanying GitHub collection of papers.
-
Building Task Bots with Self-learning for Enhanced Adaptability, Extensibility, and Factuality
A thesis that combines self-learning from dialog logs, schema-guided prompting, and self-aligned factuality to build task bots with minimal human intervention.
Discussion (0). Sign in to comment.