Image captioning has conventionally relied on reference-based automatic evaluations, where machine captions are compared against captions written by humans. This is in contrast to the reference-free manner in which humans assess caption quality. In t
his paper, we report the surprising empirical finding that CLIP (Radford et al., 2021), a cross-modal model pretrained on 400M image+caption pairs from the web, can be used for robust automatic evaluation of image captioning without the need for references. Experiments spanning several corpora demonstrate that our new reference-free metric, CLIPScore, achieves the highest correlation with human judgements, outperforming existing reference-based metrics like CIDEr and SPICE. Information gain experiments demonstrate that CLIPScore, with its tight focus on image-text compatibility, is complementary to existing reference-based metrics that emphasize text-text similarities. Thus, we also present a reference-augmented version, RefCLIPScore, which achieves even higher correlation. Beyond literal description tasks, several case studies reveal domains where CLIPScore performs well (clip-art images, alt-text rating), but also where it is relatively weaker in comparison to reference-based metrics, e.g., news captions that require richer contextual knowledge.
Due to complex cognitive and inferential efforts involved in the manual generation of one caption per image/video input, the human annotation resources are very limited for captioning tasks. We define language resource efficient as reaching the same
performance with fewer annotated captions per input. We first study the performance degradation of caption models in different language resource settings. Our analysis of caption models with SC loss shows that the performance degradation is caused by the increasingly noisy estimation of reward and baseline with fewer language resources. To mitigate this issue, we propose to reduce the variance of noise in the baseline by generalizing the single pairwise comparison in SC loss and using multiple generalized pairwise comparisons. The generalized pairwise comparison (GPC) measures the difference between the evaluation scores of two captions with respect to an input. Empirically, we show that the model trained with the proposed GPC loss is efficient on language resource and achieves similar performance with the state-of-the-art models on MSCOCO by using only half of the language resources. Furthermore, our model significantly outperforms the state-of-the-art models on a video caption dataset that has only one labeled caption per input in the training set.
Pre-training (PT) and back-translation (BT) are two simple and powerful methods to utilize monolingual data for improving the model performance of neural machine translation (NMT). This paper takes the first step to investigate the complementarity be
tween PT and BT. We introduce two probing tasks for PT and BT respectively and find that PT mainly contributes to the encoder module while BT brings more benefits to the decoder. Experimental results show that PT and BT are nicely complementary to each other, establishing state-of-the-art performances on the WMT16 English-Romanian and English-Russian benchmarks. Through extensive analyses on sentence originality and word frequency, we also demonstrate that combining Tagged BT with PT is more helpful to their complementarity, leading to better translation quality. Source code is freely available at https://github.com/SunbowLiu/PTvsBT.
This paper presents several challenges faced when annotating Turkish treebanks in accordance with the Universal Dependencies (UD) guidelines and proposes solutions to address them. Most of these challenges stem from the lack of adequate support in th
e UD framework to accurately represent null morphemes and complex derivations, which results in a significant loss of information for Turkish. This loss negatively impacts the tools that are developed based on these treebanks. We raised and discussed these issues within the community on the official UD portal. This paper presents these issues and our proposals to more accurately represent morphosyntactic information for Turkish while adhering to guidelines of UD. This work aims to contribute to the representation of Turkish and other agglutinative languages in UD-based treebanks, which in turn aids to develop more accurately annotated datasets for such languages.
We address the annotation data bottleneck for sequence classification. Specifically we ask the question: if one has a budget of N annotations, which samples should we select for annotation? The solution we propose looks for diversity in the selected
sample, by maximizing the amount of information that is useful for the learning algorithm, or equivalently by minimizing the redundancy of samples in the selection. This is formulated in the context of spectral learning of recurrent functions for sequence classification. Our method represents unlabeled data in the form of a Hankel matrix, and uses the notion of spectral max-volume to find a compact sub-block from which annotation samples are drawn. Experiments on sequence classification confirm that our spectral sampling strategy is in fact efficient and yields good models.
Precise information of word boundary can alleviate the problem of lexical ambiguity to improve the performance of natural language processing (NLP) tasks. Thus, Chinese word segmentation (CWS) is a fundamental task in NLP. Due to the development of p
re-trained language models (PLM), pre-trained knowledge can help neural methods solve the main problems of the CWS in significant measure. Existing methods have already achieved high performance on several benchmarks (e.g., Bakeoff-2005). However, recent outstanding studies are limited by the small-scale annotated corpus. To further improve the performance of CWS methods based on fine-tuning the PLMs, we propose a novel neural framework, LBGCN, which incorporates a lexicon-based graph convolutional network into the Transformer encoder. Experimental results on five benchmarks and four cross-domain datasets show the lexicon-based graph convolutional network successfully captures the information of candidate words and helps to improve performance on the benchmarks (Bakeoff-2005 and CTB6) and the cross-domain datasets (SIGHAN-2010). Further experiments and analyses demonstrate that our proposed framework effectively models the lexicon to enhance the ability of basic neural frameworks and strengthens the robustness in the cross-domain scenario.
Acquisition of multilingual training data continues to be a challenge in word sense disambiguation (WSD). To address this problem, unsupervised approaches have been proposed to automatically generate sense annotations for training supervised WSD syst
ems. We present three new methods for creating sense-annotated corpora which leverage translations, parallel bitexts, lexical resources, as well as contextual and synset embeddings. Our semi-supervised method applies machine translation to transfer existing sense annotations to other languages. Our two unsupervised methods refine sense annotations produced by a knowledge-based WSD system via lexical translations in a parallel corpus. We obtain state-of-the-art results on standard WSD benchmarks.
Speech translation (ST) has lately received growing interest for the generation of subtitles without the need for an intermediate source language transcription and timing (i.e. captions). However, the joint generation of source captions and target su
btitles does not only bring potential output quality advantages when the two decoding processes inform each other, but it is also often required in multilingual scenarios. In this work, we focus on ST models which generate consistent captions-subtitles in terms of structure and lexical content. We further introduce new metrics for evaluating subtitling consistency. Our findings show that joint decoding leads to increased performance and consistency between the generated captions and subtitles while still allowing for sufficient flexibility to produce subtitles conforming to language-specific needs and norms.
FrameNet and the Multilingual FrameNet project have produced multilingual semantic annotations of parallel texts that yield extremely fine-grained typological insights. Moreover, frame semantic annotation of a wide cross-section of languages would pr
ovide information on the limits of Frame Semantics (Fillmore 1982, Fillmore1985). Multilingual semantic annotation offers critical input for research on linguistic diversity and recurrent patterns in computational typology. Drawing on results from FrameNet annotation of parallel texts, this paper proposes frame semantic annotation as a new component to complement the state of the art in computational semantic typology.
Biomaterials are synthetic or natural materials used for constructing artificial organs, fabricating prostheses, or replacing tissues. The last century saw the development of thousands of novel biomaterials and, as a result, an exponential increase i
n scientific publications in the field. Large-scale analysis of biomaterials and their performance could enable data-driven material selection and implant design. However, such analysis requires identification and organization of concepts, such as materials and structures, from published texts. To facilitate future information extraction and the application of machine-learning techniques, we developed a semantic annotator specifically tailored for the biomaterials literature. The Biomaterials Annotator has been implemented following a modular organization using software containers for the different components and orchestrated using Nextflow as workflow manager. Natural language processing (NLP) components are mainly developed in Java. This set-up has allowed named entity recognition of seventeen classes relevant to the biomaterials domain. Here we detail the development, evaluation and performance of the system, as well as the release of the first collection of annotated biomaterials abstracts. We make both the corpus and system available to the community to promote future efforts in the field and contribute towards its sustainability.
