Surveying the Landscape of Image Captioning Evaluation: A Comprehensive Taxonomy and Novel Ensemble Method

The task of image captioning has recently been gaining popularity, and with it the complex task of evaluating the quality of image captioning models. In this work, we present the first survey and taxonomy of over 70 different image captioning metrics and their usage in hundreds of papers. We find that despite the diversity of proposed metrics, the vast majority of studies rely on only five popular metrics, which we show to be weakly correlated with human judgements. Instead, we propose EnsembEval -- an ensemble of evaluation methods achieving the highest reported correlation with human judgements across 5 image captioning datasets, showing there is a lot of room for improvement by leveraging a diverse set of metrics.

SEAM: A Stochastic Benchmark for Multi-Document Tasks

Various tasks, such as summarization, multi-hop question answering, or coreference resolution, are naturally phrased over collections of real-world documents. Such tasks present a unique set of challenges, revolving around the lack of coherent narrative structure across documents, which often leads to contradiction, omission, or repetition of information. Despite their real-world application and challenging properties, there is currently no benchmark which specifically measures the abilities of large language models (LLMs) on multi-document tasks. To bridge this gap, we present SEAM (a Stochastic Evaluation Approach for Multi-document tasks), a conglomerate benchmark over a diverse set of multi-document datasets, setting conventional evaluation criteria, input-output formats, and evaluation protocols. In particular, SEAM addresses the sensitivity of LLMs to minor prompt variations through repeated evaluations, where in each evaluation we sample uniformly at random the values of arbitrary factors (e.g., the order of documents). We evaluate different LLMs on SEAM finding that multi-document tasks pose a significant challenge for LLMs, even for state-of-the-art models with 70B parameters. In addition, we show that the stochastic approach uncovers underlying statistical trends which cannot be observed in a static benchmark. We hope that SEAM will spur progress via consistent and meaningful evaluation of multi-document tasks.

In-Context Learning on a Budget: A Case Study in Named Entity Recognition

Few shot in-context learning (ICL) typically assumes access to large annotated training sets. However, in many real world scenarios, such as domain adaptation, there is only a limited budget to annotate a small number of samples, with the goal of maximizing downstream performance. We study various methods for selecting samples to annotate within a predefined budget, specifically focusing on the named entity recognition (NER) task, which has real-world applications, is expensive to annotate, and is relatively less studied in ICL setups. Across different models and datasets, we find that a relatively small pool of annotated samples can achieve results comparable to using the entire training set. Moreover, we discover that random selection of samples for annotation yields surprisingly good performance. Finally, we observe that a diverse annotation pool is correlated with improved performance. We hope that future work adopts our realistic paradigm which takes annotation budget into account.

Applying Intrinsic Debiasing on Downstream Tasks: Challenges and Considerations for Machine Translation

Most works on gender bias focus on intrinsic bias -- removing traces of information about a protected group from the model's internal representation. However, these works are often disconnected from the impact of such debiasing on downstream applications, which is the main motivation for debiasing in the first place. In this work, we systematically test how methods for intrinsic debiasing affect neural machine translation models, by measuring the extrinsic bias of such systems under different design choices. We highlight three challenges and mismatches between the debiasing techniques and their end-goal usage, including the choice of embeddings to debias, the mismatch between words and sub-word tokens debiasing, and the effect on different target languages. We find that these considerations have a significant impact on downstream performance and the success of debiasing.

A Nurse is Blue and Elephant is Rugby: Cross Domain Alignment in Large Language Models Reveal Human-like Patterns

Cross-domain alignment refers to the task of mapping a concept from one domain to another. For example, ``If a \textit{doctor} were a \textit{color}, what color would it be?''. This seemingly peculiar task is designed to investigate how people represent concrete and abstract concepts through their mappings between categories and their reasoning processes over those mappings. In this paper, we adapt this task from cognitive science to evaluate the conceptualization and reasoning abilities of large language models (LLMs) through a behavioral study. We examine several LLMs by prompting them with a cross-domain mapping task and analyzing their responses at both the population and individual levels. Additionally, we assess the models' ability to reason about their predictions by analyzing and categorizing their explanations for these mappings. The results reveal several similarities between humans' and models' mappings and explanations, suggesting that models represent concepts similarly to humans. This similarity is evident not only in the model representation but also in their behavior. Furthermore, the models mostly provide valid explanations and deploy reasoning paths that are similar to those of humans.

Do Zombies Understand? A Choose-Your-Own-Adventure Exploration of Machine Cognition

Recent advances in LLMs have sparked a debate on whether they understand text. In this position paper, we argue that opponents in this debate hold different definitions for understanding, and particularly differ in their view on the role of consciousness. To substantiate this claim, we propose a thought experiment involving an open-source chatbot $Z$ which excels on every possible benchmark, seemingly without subjective experience. We ask whether $Z$ is capable of understanding, and show that different schools of thought within seminal AI research seem to answer this question differently, uncovering their terminological disagreement. Moving forward, we propose two distinct working definitions for understanding which explicitly acknowledge the question of consciousness, and draw connections with a rich literature in philosophy, psychology and neuroscience.

Leveraging Collection-Wide Similarities for Unsupervised Document Structure Extraction

Document collections of various domains, e.g., legal, medical, or financial, often share some underlying collection-wide structure, which captures information that can aid both human users and structure-aware models. We propose to identify the typical structure of document within a collection, which requires to capture recurring topics across the collection, while abstracting over arbitrary header paraphrases, and ground each topic to respective document locations. These requirements pose several challenges: headers that mark recurring topics frequently differ in phrasing, certain section headers are unique to individual documents and do not reflect the typical structure, and the order of topics can vary between documents. Subsequently, we develop an unsupervised graph-based method which leverages both inter- and intra-document similarities, to extract the underlying collection-wide structure. Our evaluations on three diverse domains in both English and Hebrew indicate that our method extracts meaningful collection-wide structure, and we hope that future work will leverage our method for multi-document applications and structure-aware models.

K-QA: A Real-World Medical Q&A Benchmark

Ensuring the accuracy of responses provided by large language models (LLMs) is crucial, particularly in clinical settings where incorrect information may directly impact patient health. To address this challenge, we construct K-QA, a dataset containing 1,212 patient questions originating from real-world conversations held on K Health (an AI-driven clinical platform). We employ a panel of in-house physicians to answer and manually decompose a subset of K-QA into self-contained statements. Additionally, we formulate two NLI-based evaluation metrics approximating recall and precision: (1) comprehensiveness, measuring the percentage of essential clinical information in the generated answer and (2) hallucination rate, measuring the number of statements from the physician-curated response contradicted by the LLM answer. Finally, we use K-QA along with these metrics to evaluate several state-of-the-art models, as well as the effect of in-context learning and medically-oriented augmented retrieval schemes developed by the authors. Our findings indicate that in-context learning improves the comprehensiveness of the models, and augmented retrieval is effective in reducing hallucinations. We make K-QA available to to the community to spur research into medically accurate NLP applications.

State of What Art? A Call for Multi-Prompt LLM Evaluation

Recent advances in large language models (LLMs) have led to the development of various evaluation benchmarks. These benchmarks typically rely on a single instruction template for evaluating all LLMs on a specific task. In this paper, we comprehensively analyze the brittleness of results obtained via single-prompt evaluations across 6.5M instances, involving 20 different LLMs and 39 tasks from 3 benchmarks. To improve robustness of the analysis, we propose to evaluate LLMs with a set of diverse prompts instead. We discuss tailored evaluation metrics for specific use cases (e.g., LLM developers vs. developers interested in a specific downstream task), ensuring a more reliable and meaningful assessment of LLM capabilities. We then implement these criteria and conduct evaluations of multiple models, providing insights into the true strengths and limitations of current LLMs.

Exploring the Impact of Training Data Distribution and Subword Tokenization on Gender Bias in Machine Translation

We study the effect of tokenization on gender bias in machine translation, an aspect that has been largely overlooked in previous works. Specifically, we focus on the interactions between the frequency of gendered profession names in training data, their representation in the subword tokenizer's vocabulary, and gender bias. We observe that female and non-stereotypical gender inflections of profession names (e.g., Spanish "doctora" for "female doctor") tend to be split into multiple subword tokens. Our results indicate that the imbalance of gender forms in the model's training corpus is a major factor contributing to gender bias and has a greater impact than subword splitting. We show that analyzing subword splits provides good estimates of gender-form imbalance in the training data and can be used even when the corpus is not publicly available. We also demonstrate that fine-tuning just the token embedding layer can decrease the gap in gender prediction accuracy between female and male forms without impairing the translation quality.