no code implementations • 7 Dec 2023 • Wei Liu, Haozhao Wang, Jun Wang, Zhiying Deng, Yuankai Zhang, Cheng Wang, Ruixuan Li
Rationalization empowers deep learning models with self-explaining capabilities through a cooperative game, where a generator selects a semantically consistent subset of the input as a rationale, and a subsequent predictor makes predictions based on the selected rationale.
1 code implementation • NeurIPS 2023 • Wei Liu, Jun Wang, Haozhao Wang, Ruixuan Li, Zhiying Deng, Yuankai Zhang, Yang Qiu
Instead of attempting to rectify the issues of the MMI criterion, we propose a novel criterion to uncover the causal rationale, termed the Minimum Conditional Dependence (MCD) criterion, which is grounded on our finding that the non-causal features and the target label are \emph{d-separated} by the causal rationale.
1 code implementation • 23 May 2023 • Wei Liu, Jun Wang, Haozhao Wang, Ruixuan Li, Yang Qiu, Yuankai Zhang, Jie Han, Yixiong Zou
However, such a cooperative game may incur the degeneration problem where the predictor overfits to the uninformative pieces generated by a not yet well-trained generator and in turn, leads the generator to converge to a sub-optimal model that tends to select senseless pieces.
1 code implementation • 8 May 2023 • Wei Liu, Haozhao Wang, Jun Wang, Ruixuan Li, Xinyang Li, Yuankai Zhang, Yang Qiu
Rationalization is to employ a generator and a predictor to construct a self-explaining NLP model in which the generator selects a subset of human-intelligible pieces of the input text to the following predictor.
1 code implementation • 17 Sep 2022 • Wei Liu, Haozhao Wang, Jun Wang, Ruixuan Li, Chao Yue, Yuankai Zhang
Conventional works generally employ a two-phase model in which a generator selects the most important pieces, followed by a predictor that makes predictions based on the selected pieces.