paper-with-me

홈 › Papers

HILDIF: Interactive Debugging of NLI Models Using Influence Functions

2021-08-01 · ACL (InterNLP) 2021 8 · Hugo Zylberajch, Piyawat Lertvittayakumjorn, Francesca Toni

Biases and artifacts in training data can cause unwelcome behavior in text classifiers (such as shallow pattern matching), leading to lack of generalizability. One solution to this problem is to include users in the loop and leverage their feedback to improve models. We propose a novel explanatory debugging pipeline called HILDIF, enabling humans to improve deep text classifiers using influence functions as an explanation method. We experiment on the Natural Language Inference (NLI) task, showing that HILDIF can effectively alleviate artifact problems in fine-tuned BERT models and result in increased model generalizability.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Natural Language Inference

Similar Papers 제목 키워드 기반

Complaint-driven Training Data Debugging for Query 2.0

2020-04-12 · Weiyuan Wu, Lampros Flokas, Eugene Wu, Jiannan Wang

As the need for machine learning (ML) increases rapidly across all industry sectors, there is a significant interest among commercial database providers to support "Query 2.0", which integrates model inference into SQL q…

Understanding Black-box Predictions via Influence Functions

2017-03-14 · ICML 2017 8 · Pang Wei Koh, Percy Liang

How can we explain the predictions of a black-box model? In this paper, we use influence functions -- a classic technique from robust statistics -- to trace a model's prediction through the learning algorithm and back to…

Interactive Debugging and Steering of Multi-Agent AI Systems

2025-03-03 · Will Epperson, Gagan Bansal, Victor Dibia, Adam Fourney 외

Fully autonomous teams of LLM-powered AI agents are emerging that collaborate to perform complex tasks for users. What challenges do developers face when trying to build and debug these AI agent teams? In formative inter…

AI Agent

debug-gym: A Text-Based Environment for Interactive Debugging

2025-03-27 · Xingdi Yuan, Morgane M Moss, Charbel El Feghali, Chinmay Singh 외

Large Language Models (LLMs) are increasingly relied upon for coding tasks, yet in most scenarios it is assumed that all relevant information can be either accessed in context or matches their training data. We posit tha…

Interactive Visualization for Debugging RL

2020-08-14 · Shuby Deshpande, Benjamin Eysenbach, Jeff Schneider

Visualization tools for supervised learning allow users to interpret, introspect, and gain an intuition for the successes and failures of their models. While reinforcement learning practitioners ask many of the same ques…