# It was the training data pruning too!

Source: [It was the training data pruning too!](<https://www.sota2.com/research/paper/68fa642b1e209131cea03177>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T20:58:11.540Z

Authors: Pramod Kaushik Mudrakarta, Ankur Taly, Mukund Sundararajan, Kedar Dhamdhere

Published: 2018-03-12T23:59:37.000Z

## Abstract

We study the current best model (KDG) for question answering on tabular data evaluated over the WikiTableQuestions dataset. Previous ablation studies performed against this model attributed the model's performance to certain aspects of its architecture. In this paper, we find that the model's performance also crucially depends on a certain pruning of the data used to train the model. Disabling the pruning step drops the accuracy of the model from 43.3% to 36.3%. The large impact on the performance of the KDG model suggests that the pruning may be a useful pre-processing step in training other semantic parsers as well.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1803.04579>)

[PDF](<https://arxiv.org/pdf/1803.04579.pdf>)
