paper-with-me

홈 › Papers

Measuring Causal Effects of Data Statistics on Language Model's `Factual' Predictions

2022-07-28 · Yanai Elazar, Nora Kassner, Shauli Ravfogel, Amir Feder, Abhilasha Ravichander, Marius Mosbach, Yonatan Belinkov, Hinrich Schütze, Yoav Goldberg

Large amounts of training data are one of the major reasons for the high performance of state-of-the-art NLP models. But what exactly in the training data causes a model to make a certain prediction? We seek to answer this question by providing a language for describing how training data influences predictions, through a causal framework. Importantly, our framework bypasses the need to retrain expensive models and allows us to estimate causal effects based on observational data alone. Addressing the problem of extracting factual knowledge from pretrained language models (PLMs), we focus on simple data statistics such as co-occurrence counts and show that these statistics do influence the predictions of PLMs, suggesting that such models rely on shallow heuristics. Our causal framework and our results demonstrate the importance of studying datasets and the benefits of causality for understanding NLP models.

📄 PDF Abstract BibTeX arXiv:2207.14251

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Detecting and Measuring Confounding Using Causal Mechanism Shifts

2024-09-26 · Abbavaram Gowtham Reddy, Vineeth N Balasubramanian

Detecting and measuring confounding effects from data is a key challenge in causal inference. Existing methods frequently assume causal sufficiency, disregarding the presence of unobserved confounding variables. Causal s…

Causal DiscoveryCausal Inference

Doubly Robust Kernel Statistics for Testing Distributional Treatment Effects

2022-12-09 · Jake Fawkes, Robert Hu, Robin J. Evans, Dino Sejdinovic

With the widespread application of causal inference, it is increasingly important to have tools which can test for the presence of causal effects in a diverse array of circumstances. In this vein we focus on the problem …

Causal InferencecounterfactualOff-policy evaluation

Federated Causal Inference from Observational Data

2023-08-24 · Thanh Vinh Vo, Young Lee, Tze-Yun Leong

Decentralized data sources are prevalent in real-world applications, posing a formidable challenge for causal inference. These sources cannot be consolidated into a single entity owing to privacy constraints. The presenc…

Causal InferenceFederated LearningGaussian ProcessesMissing Values+1

A Bayesian Solution to the M-Bias Problem

2019-06-17 · David Rohde

It is common practice in using regression type models for inferring causal effects, that inferring the correct causal relationship requires extra covariates are included or ``adjusted for''. Without performing this adjus…

Causal Inference

Improving Steering Vectors by Targeting Sparse Autoencoder Features

2024-11-04 · Sviatoslav Chalnev, Matthew Siu, Arthur Conmy

To control the behavior of language models, steering methods attempt to ensure that outputs of the model satisfy specific pre-defined properties. Adding steering vectors to the model is a promising method of model contro…