Rationale-Augmented Ensembles in Language Models
- URL: http://arxiv.org/abs/2207.00747v1
- Date: Sat, 2 Jul 2022 06:20:57 GMT
- Title: Rationale-Augmented Ensembles in Language Models
- Authors: Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc Le, Ed Chi, Denny Zhou
- Abstract summary: We reconsider rationale-augmented prompting for few-shot in-context learning.
We identify rationale sampling in the output space as the key component to robustly improve performance.
We demonstrate that rationale-augmented ensembles achieve more accurate and interpretable results than existing prompting approaches.
- Score: 53.45015291520658
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Recent research has shown that rationales, or step-by-step chains of thought,
can be used to improve performance in multi-step reasoning tasks. We reconsider
rationale-augmented prompting for few-shot in-context learning, where (input ->
output) prompts are expanded to (input, rationale -> output) prompts. For
rationale-augmented prompting we demonstrate how existing approaches, which
rely on manual prompt engineering, are subject to sub-optimal rationales that
may harm performance. To mitigate this brittleness, we propose a unified
framework of rationale-augmented ensembles, where we identify rationale
sampling in the output space as the key component to robustly improve
performance. This framework is general and can easily be extended to common
natural language processing tasks, even those that do not traditionally
leverage intermediate steps, such as question answering, word sense
disambiguation, and sentiment analysis. We demonstrate that rationale-augmented
ensembles achieve more accurate and interpretable results than existing
prompting approaches--including standard prompting without rationales and
rationale-based chain-of-thought prompting--while simultaneously improving
interpretability of model predictions through the associated rationales.
Related papers
- Characterizing Large Language Models as Rationalizers of
Knowledge-intensive Tasks [6.51301154858045]
Large language models (LLMs) are proficient at generating fluent text with minimal task-specific supervision.
We consider the task of generating knowledge-guided rationalization in natural language by using expert-written examples in a few-shot manner.
Surprisingly, crowd-workers preferred knowledge-grounded rationales over crowdsourced rationalizations, citing their factuality, sufficiency, and comprehensive refutations.
arXiv Detail & Related papers (2023-11-09T01:04:44Z) - Towards Trustworthy Explanation: On Causal Rationalization [9.48539398357156]
We propose a new model of rationalization based on two causal desiderata, non-spuriousness and efficiency.
The superior performance of the proposed causal rationalization is demonstrated on real-world review and medical datasets.
arXiv Detail & Related papers (2023-06-25T03:34:06Z) - HOP, UNION, GENERATE: Explainable Multi-hop Reasoning without Rationale
Supervision [118.0818807474809]
This work proposes a principled, probabilistic approach for training explainable multi-hop QA systems without rationale supervision.
Our approach performs multi-hop reasoning by explicitly modeling rationales as sets, enabling the model to capture interactions between documents and sentences within a document.
arXiv Detail & Related papers (2023-05-23T16:53:49Z) - LogiGAN: Learning Logical Reasoning via Adversarial Pre-training [58.11043285534766]
We present LogiGAN, an unsupervised adversarial pre-training framework for improving logical reasoning abilities of language models.
Inspired by the facilitation effect of reflective thinking in human learning, we simulate the learning-thinking process with an adversarial Generator-Verifier architecture.
Both base and large size language models pre-trained with LogiGAN demonstrate obvious performance improvement on 12 datasets.
arXiv Detail & Related papers (2022-05-18T08:46:49Z) - Interlock-Free Multi-Aspect Rationalization for Text Classification [33.33452117387646]
We show that we address the interlocking problem in the multi-aspect setting.
We propose a multi-stage training method incorporating an additional self-supervised contrastive loss.
Empirical results on the beer review dataset show that our method improves significantly the rationalization performance.
arXiv Detail & Related papers (2022-05-13T16:38:38Z) - Can Rationalization Improve Robustness? [39.741059642044874]
We investigate whether neural NLP models can provide robustness to adversarial attacks in addition to their interpretable nature.
We generate various types of 'AddText' attacks for both token and sentence-level rationalization tasks.
Our experiments reveal that the rationale models show the promise to improve robustness, while they struggle in certain scenarios.
arXiv Detail & Related papers (2022-04-25T17:02:42Z) - Probing as Quantifying the Inductive Bias of Pre-trained Representations [99.93552997506438]
We present a novel framework for probing where the goal is to evaluate the inductive bias of representations for a particular task.
We apply our framework to a series of token-, arc-, and sentence-level tasks.
arXiv Detail & Related papers (2021-10-15T22:01:16Z) - SPECTRA: Sparse Structured Text Rationalization [0.0]
We present a unified framework for deterministic extraction of structured explanations via constrained inference on a factor graph.
Our approach greatly eases training and rationale regularization, generally outperforming previous work on plausibility extracted explanations.
arXiv Detail & Related papers (2021-09-09T20:39:56Z) - Measuring Association Between Labels and Free-Text Rationales [60.58672852655487]
In interpretable NLP, we require faithful rationales that reflect the model's decision-making process for an explained instance.
We demonstrate that pipelines, existing models for faithful extractive rationalization on information-extraction style tasks, do not extend as reliably to "reasoning" tasks requiring free-text rationales.
We turn to models that jointly predict and rationalize, a class of widely used high-performance models for free-text rationalization whose faithfulness is not yet established.
arXiv Detail & Related papers (2020-10-24T03:40:56Z) - Invariant Rationalization [84.1861516092232]
A typical rationalization criterion, i.e. maximum mutual information (MMI), finds the rationale that maximizes the prediction performance based only on the rationale.
We introduce a game-theoretic invariant rationalization criterion where the rationales are constrained to enable the same predictor to be optimal across different environments.
We show both theoretically and empirically that the proposed rationales can rule out spurious correlations, generalize better to different test scenarios, and align better with human judgments.
arXiv Detail & Related papers (2020-03-22T00:50:27Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.