An official website of the United States government
Computational models of verbal analogy and relational similarity judgments can employ different types of vector representations of word meanings (embeddings) generated by machine-learning algorithms. An important question is whether human-like relational processing depends on explicit representations of relations (i.e., representations separable from those of the concepts being related), or whether implicit relation representations suffice. Earlier machine-learning models produced static embeddings for individual words, identical across all contexts. However, more recent Large Language Models (LLMs), which use transformer architectures applied to much larger training corpora, are able to produce contextualized embeddings that have the potential to capture implicit knowledge of semantic relations. Here we compare multiple models based on different types of embeddings to human data concerning judgments of relational similarity and solutions of verbal analogy problems. For two datasets, a model that learns explicit representations of relations, Bayesian Analogy with Relational Transformations (BART), captured human performance more successfully than either a model using static embeddings (Word2vec) or models using contextualized embeddings created by LLMs (BERT, RoBERTa, and GPT-2). These findings support the proposal that human thinking depends on representations that separate relations from the concepts they relate.
more »
« less
Perspectives on Large Language Models for Relevance Judgment
When asked, large language models (LLMs) like ChatGPT claim that they can assist with relevance judgments but it is not clear whether automated judgments can reliably be used in evaluations of retrieval systems. In this perspectives paper, we discuss possible ways for LLMs to support relevance judgments along with concerns and issues that arise. We devise a human–machine collaboration spectrum that allows to categorize different relevance judgment strategies, based on how much humans rely on machines. For the extreme point of ‘fully automated judgments’, we further include a pilot experiment on whether LLM-based relevance judgments corre- late with judgments from trained human assessors. We conclude the paper by providing opposing perspectives for and against the use of LLMs for automatic relevance judgments, and a compromise per- spective, informed by our analyses of the literature, our preliminary experimental evidence, and our experience as IR researchers
more »
« less
- Award ID(s):
- 1846017
- PAR ID:
- 10473538
- Publisher / Repository:
- ACM
- Date Published:
- ISBN:
- 9798400700736
- Page Range / eLocation ID:
- 39 to 50
- Subject(s) / Keyword(s):
- large language models, relevance judgments, human–machine collaboration, automatic test collections
- Format(s):
- Medium: X
- Location:
- Taipei Taiwan
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
When reading narratives, human readers rely on their Theory of Mind (ToM) to infer not only what the characters know from their utterances, but also whether characters are likely to share common ground. As in human conversation, such decisions are not infallible but probabilistic, based on the evidence available in the narrative. By responding on a scale (rather than Yes/No), humans can indicate commitment to their inferences about what characters know (ToM). We use two prompting approaches to explore (i) how well LLM judgments align with human judgments, and (ii) how well LLMs infer the author’s intent from utterances intended to project knowledge in narratives.more » « less
-
Identifying anomalous human spatial trajectory patterns can indicate dynamic changes in mobility behavior with applications in domains like infectious disease monitoring and elderly care. Recent advancements in large language models (LLMs) have demonstrated their ability to reason in a manner akin to humans. This presents significant potential for analyzing temporal patterns in human mobility. In this paper, we conduct empirical studies to assess the capabilities of leading LLMs like GPT-4 and Claude-2 in detecting anomalous behaviors from mobility data, by comparing to specialized methods. Our key findings demonstrate that LLMs can attain reasonable anomaly detection performance even without any specific cues. In addition, providing contextual clues about potential irregularities could further enhances their prediction efficacy. Moreover, LLMs can provide reasonable explanations for their judgments, thereby improving transparency. Our work provides insights on the strengths and limitations of LLMs for human spatial trajectory analysis.more » « less
-
The paper explores the performance of LLMs in the context of multi-dimensional analytic writing assessments, i.e. their ability to provide both scores and comments based on multiple assessment criteria. Using a corpus of literature reviews written by L2 graduate students and assessed by human experts against 9 analytic criteria, we prompt several popular LLMs to perform the same task under various conditions. To evaluate the quality of feedback comments, we apply a novel feedback comment quality evaluation framework. This framework is interpretable, cost-efficient, scalable, and reproducible, compared to existing methods that rely on manual judgments. We find that LLMs can generate reasonably good and generally reliable multi-dimensional analytic assessments. We release our corpus and code\footnote{\url{https://github.com/jaaack-wang/multi-dimensional-analytic-writing-assessments}.} for reproducibility.more » « less
-
Dungeons & Dragons (D&D) is a tabletop roleplaying game with complex natural language interactions between players and hidden state information. Recent work has shown that large language models (LLMs) that have access to state information can generate higher quality game turns than LLMs that use dialog history alone. However, previous work used game state information that was heuristically created and was not a true gold standard game state. We present FIREBALL, a large dataset containing nearly 25,000 unique sessions from real D&D gameplay on Discord with true game state info. We recorded game play sessions of players who used the Avrae bot, which was developed to aid people in playing D&D online, capturing language, game commands and underlying game state information. We demonstrate that FIREBALL can improve natural language generation (NLG) by using Avrae state information, improving both automated metrics and human judgments of quality. Additionally, we show that LLMs can generate executable Avrae commands, particularly after finetuning.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1145/3578337.3605136
