Understanding Understanding: A Pragmatic Framework Motivated by Large Language Models

• URL: http://arxiv.org/abs/2406.10937v2
• Date: Wed, 19 Jun 2024 08:34:21 GMT
• Title: Understanding Understanding: A Pragmatic Framework Motivated by Large Language Models
• Authors: Kevin Leyton-Brown, Yoav Shoham,
• Abstract summary: In Turing-test fashion, the framework is based solely on the agent's performance, and specifically on how well it answers questions. We show how high confidence can be achieved via random sampling and the application of probabilistic confidence bounds.
• Score: 13.279760256875127
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Motivated by the rapid ascent of Large Language Models (LLMs) and debates about the extent to which they possess human-level qualities, we propose a framework for testing whether any agent (be it a machine or a human) understands a subject matter. In Turing-test fashion, the framework is based solely on the agent's performance, and specifically on how well it answers questions. Elements of the framework include circumscribing the set of questions (the "scope of understanding"), requiring general competence ("passing grade"), avoiding "ridiculous answers", but still allowing wrong and "I don't know" answers to some questions. Reaching certainty about these conditions requires exhaustive testing of the questions which is impossible for nontrivial scopes, but we show how high confidence can be achieved via random sampling and the application of probabilistic confidence bounds. We also show that accompanying answers with explanations can improve the sample complexity required to achieve acceptable bounds, because an explanation of an answer implies the ability to answer many similar questions. According to our framework, current LLMs cannot be said to understand nontrivial domains, but as the framework provides a practical recipe for testing understanding, it thus also constitutes a tool for building AI agents that do understand.

Related papers

• Crafting Interpretable Embeddings by Asking LLMs Questions [89.49960984640363]
  Large language models (LLMs) have rapidly improved text embeddings for a growing array of natural-language processing tasks. We introduce question-answering embeddings (QA-Emb), embeddings where each feature represents an answer to a yes/no question asked to an LLM. We use QA-Emb to flexibly generate interpretable models for predicting fMRI voxel responses to language stimuli.
  arXiv  Detail & Related papers  (2024-05-26T22:30:29Z)
• Distilling Reasoning Ability from Large Language Models with Adaptive Thinking [54.047761094420174]
  Chain of thought finetuning (cot-finetuning) aims to endow small language models (SLM) with reasoning ability to improve their performance towards specific tasks. Most existing cot-finetuning methods adopt a pre-thinking mechanism, allowing the SLM to generate a rationale before providing an answer. This mechanism enables SLM to analyze and think about complex questions, but it also makes answer correctness highly sensitive to minor errors in rationale. We propose a robust post-thinking mechanism to generate answers before rationale.
  arXiv  Detail & Related papers  (2024-04-14T07:19:27Z)
• Clarify When Necessary: Resolving Ambiguity Through Interaction with LMs [58.620269228776294]
  We propose a task-agnostic framework for resolving ambiguity by asking users clarifying questions. We evaluate systems across three NLP applications: question answering, machine translation and natural language inference. We find that intent-sim is robust, demonstrating improvements across a wide range of NLP tasks and LMs.
  arXiv  Detail & Related papers  (2023-11-16T00:18:50Z)
• Open-ended Commonsense Reasoning with Unrestricted Answer Scope [47.14397700770702]
  Open-ended Commonsense Reasoning is defined as solving a commonsense question without providing 1) a short list of answer candidates and 2) a pre-defined answer scope. In this work, we leverage pre-trained language models to iteratively retrieve reasoning paths on the external knowledge base. The reasoning paths can help to identify the most precise answer to the commonsense question.
  arXiv  Detail & Related papers  (2023-10-18T02:45:54Z)
• Reasoning over Hierarchical Question Decomposition Tree for Explainable Question Answering [83.74210749046551]
  We propose to leverage question decomposing for heterogeneous knowledge integration. We propose a novel two-stage XQA framework, Reasoning over Hierarchical Question Decomposition Tree (RoHT) Experiments on complex QA datasets KQA Pro and Musique show that our framework outperforms SOTA methods significantly.
  arXiv  Detail & Related papers  (2023-05-24T11:45:59Z)
• Keeping the Questions Conversational: Using Structured Representations to Resolve Dependency in Conversational Question Answering [26.997542897342164]
  We propose a novel framework, CONVSR (CONVQA using Structured Representations) for capturing and generating intermediate representations as conversational cues. We test our model on the QuAC and CANARD datasets and illustrate by experimental results that our proposed framework achieves a better F1 score than the standard question rewriting model.
  arXiv  Detail & Related papers  (2023-04-14T13:42:32Z)
• Successive Prompting for Decomposing Complex Questions [50.00659445976735]
  Recent works leverage the capabilities of large language models (LMs) to perform complex question answering in a few-shot setting. We introduce Successive Prompting'', where we iteratively break down a complex task into a simple task, solve it, and then repeat the process until we get the final solution. Our best model (with successive prompting) achieves an improvement of 5% absolute F1 on a few-shot version of the DROP dataset.
  arXiv  Detail & Related papers  (2022-12-08T06:03:38Z)
• WikiWhy: Answering and Explaining Cause-and-Effect Questions [62.60993594814305]
  We introduce WikiWhy, a QA dataset built around explaining why an answer is true in natural language. WikiWhy contains over 9,000 "why" question-answer-rationale triples, grounded on Wikipedia facts across a diverse set of topics. GPT-3 baselines achieve only 38.7% human-evaluated correctness in the end-to-end answer & explain condition.
  arXiv  Detail & Related papers  (2022-10-21T17:59:03Z)
• Interrogating the Black Box: Transparency through Information-Seeking Dialogues [9.281671380673306]
  We propose to construct an investigator agent to query a learning agent to investigate its adherence to an ethical policy. This formal dialogue framework is the main contribution of this paper. We argue that the introduced formal dialogue framework opens many avenues both in the area of compliance checking and in the analysis of properties of opaque systems.
  arXiv  Detail & Related papers  (2021-02-09T09:14:04Z)

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.