Can Large Language Models Explain Themselves? A Study of LLM-Generated Self-Explanations

• URL: http://arxiv.org/abs/2310.11207v1
• Date: Tue, 17 Oct 2023 12:34:32 GMT
• Title: Can Large Language Models Explain Themselves? A Study of LLM-Generated Self-Explanations
• Authors: Shiyuan Huang, Siddarth Mamidanna, Shreedhar Jangam, Yilun Zhou, Leilani H. Gilpin
• Abstract summary: Large language models (LLMs) such as ChatGPT have demonstrated superior performance on a variety of natural language processing (NLP) tasks. Since these models are instruction-tuned on human conversations to produce "helpful" responses, they can and often will produce explanations along with the response.
• Score: 14.685170467182369
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) such as ChatGPT have demonstrated superior performance on a variety of natural language processing (NLP) tasks including sentiment analysis, mathematical reasoning and summarization. Furthermore, since these models are instruction-tuned on human conversations to produce "helpful" responses, they can and often will produce explanations along with the response, which we call self-explanations. For example, when analyzing the sentiment of a movie review, the model may output not only the positivity of the sentiment, but also an explanation (e.g., by listing the sentiment-laden words such as "fantastic" and "memorable" in the review). How good are these automatically generated self-explanations? In this paper, we investigate this question on the task of sentiment analysis and for feature attribution explanation, one of the most commonly studied settings in the interpretability literature (for pre-ChatGPT models). Specifically, we study different ways to elicit the self-explanations, evaluate their faithfulness on a set of evaluation metrics, and compare them to traditional explanation methods such as occlusion or LIME saliency maps. Through an extensive set of experiments, we find that ChatGPT's self-explanations perform on par with traditional ones, but are quite different from them according to various agreement metrics, meanwhile being much cheaper to produce (as they are generated along with the prediction). In addition, we identified several interesting characteristics of them, which prompt us to rethink many current model interpretability practices in the era of ChatGPT(-like) LLMs.

Related papers

• Evaluating the Reliability of Self-Explanations in Large Language Models [2.8894038270224867]
  We evaluate two kinds of such self-explanations - extractive and counterfactual. Our findings reveal, that, while these self-explanations can correlate with human judgement, they do not fully and accurately follow the model's decision process. We show that this gap can be bridged because prompting LLMs for counterfactual explanations can produce faithful, informative, and easy-to-verify results.
  arXiv  Detail & Related papers  (2024-07-19T17:41:08Z)
• Scenarios and Approaches for Situated Natural Language Explanations [18.022428746019582]
  We collect a benchmarking dataset, Situation-Based Explanation. This dataset contains 100 explanandums. For each "explanandum paired with an audience" situation, we include a human-written explanation. We examine three categories of prompting methods: rule-based prompting, meta-prompting, and in-context learning prompting.
  arXiv  Detail & Related papers  (2024-06-07T15:56:32Z)
• Evaluating Consistency and Reasoning Capabilities of Large Language Models [0.0]
  Large Language Models (LLMs) are extensively used today across various sectors, including academia, research, business, and finance. Despite their widespread adoption, these models often produce incorrect and misleading information, exhibiting a tendency to hallucinate. This paper aims to evaluate and compare the consistency and reasoning capabilities of both public and proprietary LLMs.
  arXiv  Detail & Related papers  (2024-04-25T10:03:14Z)
• Beware of Words: Evaluating the Lexical Richness of Conversational Large Language Models [3.0059120458540383]
  We consider the evaluation of the lexical richness of the text generated by conversational Large Language Models (LLMs) and how it depends on the model parameters. The results show how lexical richness depends on the version of ChatGPT and some of its parameters, such as the presence penalty, or on the role assigned to the model.
  arXiv  Detail & Related papers  (2024-02-11T13:41:17Z)
• "You Are An Expert Linguistic Annotator": Limits of LLMs as Analyzers of Abstract Meaning Representation [60.863629647985526]
  We examine the successes and limitations of the GPT-3, ChatGPT, and GPT-4 models in analysis of sentence meaning structure. We find that models can reliably reproduce the basic format of AMR, and can often capture core event, argument, and modifier structure. Overall, our findings indicate that these models out-of-the-box can capture aspects of semantic structure, but there remain key limitations in their ability to support fully accurate semantic analyses or parses.
  arXiv  Detail & Related papers  (2023-10-26T21:47:59Z)
• Towards a Mechanistic Interpretation of Multi-Step Reasoning Capabilities of Language Models [107.07851578154242]
  Language models (LMs) have strong multi-step (i.e., procedural) reasoning capabilities. It is unclear whether LMs perform tasks by cheating with answers memorized from pretraining corpus, or, via a multi-step reasoning mechanism. We show that MechanisticProbe is able to detect the information of the reasoning tree from the model's attentions for most examples.
  arXiv  Detail & Related papers  (2023-10-23T01:47:29Z)
• STREET: A Multi-Task Structured Reasoning and Explanation Benchmark [56.555662318619135]
  We introduce a unified multi-task and multi-domain natural language reasoning and explanation benchmark. We expect models to not only answer questions, but also produce step-by-step structured explanations describing how premises in the question are used to produce intermediate conclusions that can prove the correctness of a certain answer.
  arXiv  Detail & Related papers  (2023-02-13T22:34:02Z)
• The Goldilocks of Pragmatic Understanding: Fine-Tuning Strategy Matters for Implicature Resolution by LLMs [26.118193748582197]
  We evaluate four categories of widely used state-of-the-art models. We find that, despite only evaluating on utterances that require a binary inference, models in three of these categories perform close to random. These results suggest that certain fine-tuning strategies are far better at inducing pragmatic understanding in models.
  arXiv  Detail & Related papers  (2022-10-26T19:04:23Z)
• Interpreting Language Models with Contrastive Explanations [99.7035899290924]
  Language models must consider various features to predict a token, such as its part of speech, number, tense, or semantics. Existing explanation methods conflate evidence for all these features into a single explanation, which is less interpretable for human understanding. We show that contrastive explanations are quantifiably better than non-contrastive explanations in verifying major grammatical phenomena.
  arXiv  Detail & Related papers  (2022-02-21T18:32:24Z)
• Prompting Contrastive Explanations for Commonsense Reasoning Tasks [74.7346558082693]
  Large pretrained language models (PLMs) can achieve near-human performance on commonsense reasoning tasks. We show how to use these same models to generate human-interpretable evidence.
  arXiv  Detail & Related papers  (2021-06-12T17:06:13Z)
• Evaluating Explanations: How much do explanations from the teacher aid students? [103.05037537415811]
  We formalize the value of explanations using a student-teacher paradigm that measures the extent to which explanations improve student models in learning. Unlike many prior proposals to evaluate explanations, our approach cannot be easily gamed, enabling principled, scalable, and automatic evaluation of attributions.
  arXiv  Detail & Related papers  (2020-12-01T23:40:21Z)

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.