We propose to use novel and classical audio and text signal-processing and otherwise techniques for "inexpensive" fast writer identification tasks of scanned hand-written documents "visually". The "inexpensive" refers to the efficiency of the identification process in terms of CPU cycles while preserving decent accuracy for preliminary identification. This is a comparative study of multiple algorithm combinations in a pattern recognition pipeline implemented in Java around an open-source Modular Audio Recognition Framework (MARF) that can do a lot more beyond audio. We present our preliminary experimental findings in such an identification task. We simulate "visual" identification by "looking" at the hand-written document as a whole rather than trying to extract fine-grained features out of it prior classification.
We present the architecture of the UNL-French deconverter, which "generates" from the UNL interlingua by first"localizing" the UNL form for French, within UNL, and then applying slightly adapted but classical transfer and generation techniques, implemented in GETA's Ariane-G5 environment, supplemented by some UNL-specific tools. Online interaction can be used during deconversion to enhance output quality and is now used for development purposes. We show how interaction could be delayed and embedded in the postedition phase, which would then interact not directly with the output text, but indirectly with several components of the deconverter. Interacting online or offline can improve the quality not only of the utterance at hand, but also of the utterances processed later, as various preferences may be automatically changed to let the deconverter "learn".
After presenting a novel O(n^3) parsing algorithm for dependency grammar, we develop three contrasting ways to stochasticize it. We propose (a) a lexical affinity model where words struggle to modify each other, (b) a sense tagging model where words fluctuate randomly in their selectional preferences, and (c) a generative model where the speaker fleshes out each word's syntactic and conceptual structure without regard to the implications for the hearer. We also give preliminary empirical results from evaluating the three models' parsing performance on annotated Wall Street Journal training text (derived from the Penn Treebank). In these results, the generative (i.e., top-down) model performs significantly better than the others, and does about equally well at assigning part-of-speech tags.
Question answering-based summarization evaluation metrics must automatically determine whether the QA model's prediction is correct or not, a task known as answer verification. In this work, we benchmark the lexical answer verification methods which have been used by current QA-based metrics as well as two more sophisticated text comparison methods, BERTScore and LERC. We find that LERC out-performs the other methods in some settings while remaining statistically indistinguishable from lexical overlap in others. However, our experiments reveal that improved verification performance does not necessarily translate to overall QA-based metric quality: In some scenarios, using a worse verification method -- or using none at all -- has comparable performance to using the best verification method, a result that we attribute to properties of the datasets.
This paper presents the shared task on Multilingual Idiomaticity Detection and Sentence Embedding, which consists of two subtasks: (a) a binary classification one aimed at identifying whether a sentence contains an idiomatic expression, and (b) a task based on semantic text similarity which requires the model to adequately represent potentially idiomatic expressions in context. Each subtask includes different settings regarding the amount of training data. Besides the task description, this paper introduces the datasets in English, Portuguese, and Galician and their annotation procedure, the evaluation metrics, and a summary of the participant systems and their results. The task had close to 100 registered participants organised into twenty five teams making over 650 and 150 submissions in the practice and evaluation phases respectively.
We study the effect of applying a language model (LM) on the output of Automatic Speech Recognition (ASR) systems for Indic languages. We fine-tune wav2vec $2.0$ models for $18$ Indic languages and adjust the results with language models trained on text derived from a variety of sources. Our findings demonstrate that the average Character Error Rate (CER) decreases by over $28$ \% and the average Word Error Rate (WER) decreases by about $36$ \% after decoding with LM. We show that a large LM may not provide a substantial improvement as compared to a diverse one. We also demonstrate that high quality transcriptions can be obtained on domain-specific data without retraining the ASR model and show results on biomedical domain.
We propose and demonstrate the task of giving natural language summaries of the actions of a robotic agent in a virtual environment. We explain why such a task is important, what makes it difficult, and discuss how it might be addressed. To encourage others to work on this, we show how a popular existing dataset that matches robot actions with natural language descriptions designed for an instruction following task can be repurposed to serve as a training ground for robot action summarization work. We propose and test several methods of learning to generate such summaries, starting from either egocentric video frames of the robot taking actions or intermediate text representations of the actions used by an automatic planner. We provide quantitative and qualitative evaluations of our results, which can serve as a baseline for future work.
Predicate entailment detection is a crucial task for question-answering from text, where previous work has explored unsupervised learning of entailment graphs from typed open relation triples. In this paper, we present the first pipeline for building Chinese entailment graphs, which involves a novel high-recall open relation extraction (ORE) method and the first Chinese fine-grained entity typing dataset under the FIGER type ontology. Through experiments on the Levy-Holt dataset, we verify the strength of our Chinese entailment graph, and reveal the cross-lingual complementarity: on the parallel Levy-Holt dataset, an ensemble of Chinese and English entailment graphs outperforms both monolingual graphs, and raises unsupervised SOTA by 4.7 AUC points.
In biomedical natural language processing, named entity recognition (NER) and named entity normalization (NEN) are key tasks that enable the automatic extraction of biomedical entities (e.g., diseases and chemicals) from the ever-growing biomedical literature. In this paper, we present BERN2 (Advanced Biomedical Entity Recognition and Normalization), a tool that improves the previous neural network-based NER tool (Kim et al., 2019) by employing a multi-task NER model and neural network-based NEN models to achieve much faster and more accurate inference. We hope that our tool can help annotate large-scale biomedical texts more accurately for various tasks such as biomedical knowledge graph construction.
Add to Chrome
Add to Firefox