Current large language models can perform reasonably well on complex tasks that require step-by-step reasoning with few-shot learning. Are these models applying reasoning skills they have learnt during pre-training and reason outside of their training context, or are they simply memorizing their training corpus at finer granularity and have learnt to better understand their context? To tease apart these possibilities, we introduce ALERT, a benchmark and suite of analyses for assessing language models' reasoning ability comparing pre-trained and finetuned models on complex tasks that require reasoning skills to solve. ALERT provides a test bed to asses any language model on fine-grained reasoning skills, which spans over 20 datasets and covers 10 different reasoning skills. We leverage ALERT to further investigate the role of finetuning. With extensive empirical analysis we find that language models learn more reasoning skills such as textual entailment, abductive reasoning, and analogical reasoning during finetuning stage compared to pretraining state. We also find that when language models are finetuned they tend to overfit to the prompt template, which hurts the robustness of models causing generalization problems.
Understanding novel situations in the traffic domain requires an intricate combination of domain-specific and causal commonsense knowledge. Prior work has provided sufficient perception-based modalities for traffic monitoring, in this paper, we focus on a complementary research aspect of Intelligent Transportation: traffic understanding. We scope our study to text-based methods and datasets given the abundant commonsense knowledge that can be extracted using language models from large corpus and knowledge graphs. We adopt three knowledge-driven approaches for zero-shot QA over traffic situations, based on prior natural language inference methods, commonsense models with knowledge graph self-supervision, and dense retriever-based models. We constructed two text-based multiple-choice question answering sets: BDD-QA for evaluating causal reasoning in the traffic domain and HDT-QA for measuring the possession of domain knowledge akin to human driving license tests. Among the methods, Unified-QA reaches the best performance on the BDD-QA dataset with the adaptation of multiple formats of question answers. Language models trained with inference information and commonsense knowledge are also good at predicting the cause and effect in the traffic domain but perform badly at answering human-driving QA sets. For such sets, DPR+Unified-QA performs the best due to its efficient knowledge extraction.
We introduce a linguistically enhanced combination of pre-training methods for transformers. The pre-training objectives include POS-tagging, synset prediction based on semantic knowledge graphs, and parent prediction based on dependency parse trees. Our approach achieves competitive results on the Natural Language Inference task, compared to the state of the art. Specifically for smaller models, the method results in a significant performance boost, emphasizing the fact that intelligent pre-training can make up for fewer parameters and help building more efficient models. Combining POS-tagging and synset prediction yields the overall best results.
Stance detection (SD) concerns automatically determining the viewpoint (i.e., in favour of, against, or neutral) of a text's author towards a target. SD has been applied to many research topics, among which the detection of stances behind political tweets is an important one. In this paper, we apply SD to a dataset of tweets from official party accounts in the Netherlands between 2017 and 2021, with a focus on stances towards traditional gender role division, a dividing issue between (some) Dutch political parties. To implement and improve SD of traditional gender role division, we propose to leverage an established survey instrument from social sciences, which has been validated for the purpose of measuring attitudes towards traditional gender role division. Based on our experiments, we show that using such a validated survey instrument helps to improve SD performance.
Data augmentation is a widely used technique in machine learning to improve model performance. However, existing data augmentation techniques in natural language understanding (NLU) may not fully capture the complexity of natural language variations, and they can be challenging to apply to large datasets. This paper proposes the Random Position Noise (RPN) algorithm, a novel data augmentation technique that operates at the word vector level. RPN modifies the word embeddings of the original text by introducing noise based on the existing values of selected word vectors, allowing for more fine-grained modifications and better capturing natural language variations. Unlike traditional data augmentation methods, RPN does not require gradients in the computational graph during virtual sample updates, making it simpler to apply to large datasets. Experimental results demonstrate that RPN consistently outperforms existing data augmentation techniques across various NLU tasks, including sentiment analysis, natural language inference, and paraphrase detection. Moreover, RPN performs well in low-resource settings and is applicable to any model featuring a word embeddings layer. The proposed RPN algorithm is a promising approach for enhancing NLU performance and addressing the challenges associated with traditional data augmentation techniques in large-scale NLU tasks. Our experimental results demonstrated that the RPN algorithm achieved state-of-the-art performance in all seven NLU tasks, thereby highlighting its effectiveness and potential for real-world NLU applications.
JamPatoisNLI provides the first dataset for natural language inference in a creole language, Jamaican Patois. Many of the most-spoken low-resource languages are creoles. These languages commonly have a lexicon derived from a major world language and a distinctive grammar reflecting the languages of the original speakers and the process of language birth by creolization. This gives them a distinctive place in exploring the effectiveness of transfer from large monolingual or multilingual pretrained models. While our work, along with previous work, shows that transfer from these models to low-resource languages that are unrelated to languages in their training set is not very effective, we would expect stronger results from transfer to creoles. Indeed, our experiments show considerably better results from few-shot learning of JamPatoisNLI than for such unrelated languages, and help us begin to understand how the unique relationship between creoles and their high-resource base languages affect cross-lingual transfer. JamPatoisNLI, which consists of naturally-occurring premises and expert-written hypotheses, is a step towards steering research into a traditionally underserved language and a useful benchmark for understanding cross-lingual NLP.
Natural language inference has trended toward studying contexts beyond the sentence level. An important application area is law: past cases often do not foretell how they apply to new situations and implications must be inferred. This paper introduces LawngNLI, constructed from U.S. legal opinions with automatic labels with high human-validated accuracy. Premises are long and multigranular. Experiments show two use cases. First, LawngNLI can benchmark for in-domain generalization from short to long contexts. It has remained unclear if large-scale long-premise NLI datasets actually need to be constructed: near-top performance on long premises could be achievable by fine-tuning using short premises. Without multigranularity, benchmarks cannot distinguish lack of fine-tuning on long premises versus domain shift between short and long datasets. In contrast, our long and short premises share the same examples and domain. Models fine-tuned using several past NLI datasets and/or our short premises fall short of top performance on our long premises. So for at least certain domains (such as ours), large-scale long-premise datasets are needed. Second, LawngNLI can benchmark for implication-based retrieval. Queries are entailed or contradicted by target documents, allowing users to move between arguments and evidence. Leading retrieval models perform reasonably zero shot on a LawngNLI-derived retrieval task. We compare different systems for re-ranking, including lexical overlap and cross-encoders fine-tuned using a modified LawngNLI or past NLI datasets. LawngNLI can train and test systems for implication-based case retrieval and argumentation.
Advancement in large pretrained language models has significantly improved their performance for conditional language generation tasks including summarization albeit with hallucinations. To reduce hallucinations, conventional methods proposed improving beam search or using a fact checker as a postprocessing step. In this paper, we investigate the use of the Natural Language Inference (NLI) entailment metric to detect and prevent hallucinations in summary generation. We propose an NLI-assisted beam re-ranking mechanism by computing entailment probability scores between the input context and summarization model-generated beams during saliency-enhanced greedy decoding. Moreover, a diversity metric is introduced to compare its effectiveness against vanilla beam search. Our proposed algorithm significantly outperforms vanilla beam decoding on XSum and CNN/DM datasets.
This technical report briefly describes our JDExplore d-team's Vega v2 submission on the SuperGLUE leaderboard. SuperGLUE is more challenging than the widely used general language understanding evaluation (GLUE) benchmark, containing eight difficult language understanding tasks, including question answering, natural language inference, word sense disambiguation, coreference resolution, and reasoning. [Method] Instead of arbitrarily increasing the size of a pretrained language model (PLM), our aim is to 1) fully extract knowledge from the input pretraining data given a certain parameter budget, e.g., 6B, and 2) effectively transfer this knowledge to downstream tasks. To achieve goal 1), we propose self-evolution learning for PLMs to wisely predict the informative tokens that should be masked, and supervise the masked language modeling (MLM) process with rectified smooth labels. For goal 2), we leverage the prompt transfer technique to improve the low-resource tasks by transferring the knowledge from the foundation model and related downstream tasks to the target task. [Results] According to our submission record (Oct. 2022), with our optimized pretraining and fine-tuning strategies, our 6B Vega method achieved new state-of-the-art performance on 4/8 tasks, sitting atop the SuperGLUE leaderboard on Oct. 8, 2022, with an average score of 91.3.
Many NLP tasks can be regarded as a selection problem from a set of options, such as classification tasks, multi-choice question answering, etc. Textual entailment (TE) has been shown as the state-of-the-art (SOTA) approach to dealing with those selection problems. TE treats input texts as premises (P), options as hypotheses (H), then handles the selection problem by modeling (P, H) pairwise. Two limitations: first, the pairwise modeling is unaware of other options, which is less intuitive since humans often determine the best options by comparing competing candidates; second, the inference process of pairwise TE is time-consuming, especially when the option space is large. To deal with the two issues, this work first proposes a contextualized TE model (Context-TE) by appending other k options as the context of the current (P, H) modeling. Context-TE is able to learn more reliable decision for the H since it considers various context. Second, we speed up Context-TE by coming up with Parallel-TE, which learns the decisions of multiple options simultaneously. Parallel-TE significantly improves the inference speed while keeping comparable performance with Context-TE. Our methods are evaluated on three tasks (ultra-fine entity typing, intent detection and multi-choice QA) that are typical selection problems with different sizes of options. Experiments show our models set new SOTA performance; particularly, Parallel-TE is faster than the pairwise TE by k times in inference. Our code is publicly available at https://github.com/jiangshdd/LearningToSelect.
Scoring the factuality of a generated summary involves measuring the degree to which a target text contains factual information using the input document as support. Given the similarities in the problem formulation, previous work has shown that Natural Language Inference models can be effectively repurposed to perform this task. As these models are trained to score entailment at a sentence level, several recent studies have shown that decomposing either the input document or the summary into sentences helps with factuality scoring. But is fine-grained decomposition always a winning strategy? In this paper we systematically compare different granularities of decomposition -- from document to sub-sentence level, and we show that the answer is no. Our results show that incorporating additional context can yield improvement, but that this does not necessarily apply to all datasets. We also show that small changes to previously proposed entailment-based scoring methods can result in better performance, highlighting the need for caution in model and methodology selection for downstream tasks.
There is no such thing as a perfect dataset. In some datasets, deep neural networks discover underlying heuristics that allow them to take shortcuts in the learning process, resulting in poor generalization capability. Instead of using standard cross-entropy, we explore whether a modulated version of cross-entropy called focal loss can constrain the model so as not to use heuristics and improve generalization performance. Our experiments in natural language inference show that focal loss has a regularizing impact on the learning process, increasing accuracy on out-of-distribution data, but slightly decreasing performance on in-distribution data. Despite the improved out-of-distribution performance, we demonstrate the shortcomings of focal loss and its inferiority in comparison to the performance of methods such as unbiased focal loss and self-debiasing ensembles.
Careful prompt design is critical to the use of large language models in zero-shot or few-shot learning. As a consequence, there is a growing interest in automated methods to design optimal prompts. In this work, we propose Test-time Prompt Editing using Reinforcement learning (TEMPERA). In contrast to prior prompt generation methods, TEMPERA can efficiently leverage prior knowledge, is adaptive to different queries and provides an interpretable prompt for every query. To achieve this, we design a novel action space that allows flexible editing of the initial prompts covering a wide set of commonly-used components like instructions, few-shot exemplars, and verbalizers. The proposed method achieves significant gains compared with recent SoTA approaches like prompt tuning, AutoPrompt, and RLPrompt, across a variety of tasks including sentiment analysis, topic classification, natural language inference, and reading comprehension. Our method achieves 5.33x on average improvement in sample efficiency when compared to the traditional fine-tuning methods.
While large pre-trained language models are powerful, their predictions often lack logical consistency across test inputs. For example, a state-of-the-art Macaw question-answering (QA) model answers 'Yes' to 'Is a sparrow a bird?' and 'Does a bird have feet?' but answers 'No' to 'Does a sparrow have feet?'. To address this failure mode, we propose a framework, Consistency Correction through Relation Detection, or ConCoRD, for boosting the consistency and accuracy of pre-trained NLP models using pre-trained natural language inference (NLI) models without fine-tuning or re-training. Given a batch of test inputs, ConCoRD samples several candidate outputs for each input and instantiates a factor graph that accounts for both the model's belief about the likelihood of each answer choice in isolation and the NLI model's beliefs about pair-wise answer choice compatibility. We show that a weighted MaxSAT solver can efficiently compute high-quality answer choices under this factor graph, improving over the raw model's predictions. Our experiments demonstrate that ConCoRD consistently boosts accuracy and consistency of off-the-shelf closed-book QA and VQA models using off-the-shelf NLI models, notably increasing accuracy of LXMERT on ConVQA by 5% absolute. See https://ericmitchell.ai/emnlp-2022-concord/ for code and data.
This work examines the use of contradiction in natural language inference (NLI) for question answering (QA). Typically, NLI systems help answer questions by determining if a potential answer is \emph{entailed} (supported) by some background context. But is it useful to also determine if an answer contradicts the context? We test this in two settings, multiple choice and extractive QA, and find that systems that incorporate contradiction can do slightly better than entailment-only systems on certain datasets. However, the best performances come from using contradiction, entailment, and QA model confidence scores together. This has implications for the deployment of QA systems in domains such as medicine and science where safety is an issue.
Fairness and environmental impact are important research directions for the sustainable development of artificial intelligence. However, while each topic is an active research area in natural language processing (NLP), there is a surprising lack of research on the interplay between the two fields. This lacuna is highly problematic, since there is increasing evidence that an exclusive focus on fairness can actually hinder environmental sustainability, and vice versa. In this work, we shed light on this crucial intersection in NLP by (1) investigating the efficiency of current fairness approaches through surveying example methods for reducing unfair stereotypical bias from the literature, and (2) evaluating a common technique to reduce energy consumption (and thus environmental impact) of English NLP models, knowledge distillation (KD), for its impact on fairness. In this case study, we evaluate the effect of important KD factors, including layer and dimensionality reduction, with respect to: (a) performance on the distillation task (natural language inference and semantic similarity prediction), and (b) multiple measures and dimensions of stereotypical bias (e.g., gender bias measured via the Word Embedding Association Test). Our results lead us to clarify current assumptions regarding the effect of KD on unfair bias: contrary to other findings, we show that KD can actually decrease model fairness.
It has been shown that NLI models are usually biased with respect to the word-overlap between premise and hypothesis; they take this feature as a primary cue for predicting the entailment label. In this paper, we focus on an overlooked aspect of the overlap bias in NLI models: the reverse word-overlap bias. Our experimental results demonstrate that current NLI models are highly biased towards the non-entailment label on instances with low overlap, and the existing debiasing methods, which are reportedly successful on existing challenge datasets, are generally ineffective in addressing this category of bias. We investigate the reasons for the emergence of the overlap bias and the role of minority examples in its mitigation. For the former, we find that the word-overlap bias does not stem from pre-training, and for the latter, we observe that in contrast to the accepted assumption, eliminating minority examples does not affect the generalizability of debiasing methods with respect to the overlap bias.
Natural Language Inference (NLI) or Recognizing Textual Entailment (RTE) aims at predicting the relation between a pair of sentences (premise and hypothesis) as entailment, contradiction or semantic independence. Although deep learning models have shown promising performance for NLI in recent years, they rely on large scale expensive human-annotated datasets. Semi-supervised learning (SSL) is a popular technique for reducing the reliance on human annotation by leveraging unlabeled data for training. However, despite its substantial success on single sentence classification tasks where the challenge in making use of unlabeled data is to assign "good enough" pseudo-labels, for NLI tasks, the nature of unlabeled data is more complex: one of the sentences in the pair (usually the hypothesis) along with the class label are missing from the data and require human annotations, which makes SSL for NLI more challenging. In this paper, we propose a novel way to incorporate unlabeled data in SSL for NLI where we use a conditional language model, BART to generate the hypotheses for the unlabeled sentences (used as premises). Our experiments show that our SSL framework successfully exploits unlabeled data and substantially improves the performance of four NLI datasets in low-resource settings. We release our code at: https://github.com/msadat3/SSL_for_NLI.
Words of estimative probability (WEP) are expressions of a statement's plausibility (probably, maybe, likely, doubt, likely, unlikely, impossible...). Multiple surveys demonstrate the agreement of human evaluators when assigning numerical probability levels to WEP. For example, highly likely corresponds to a median chance of 0.90+-0.08 in Fagen-Ulmschneider (2015)'s survey. In this work, we measure the ability of neural language processing models to capture the consensual probability level associated to each WEP. Firstly, we use the UNLI dataset (Chen et al., 2020) which associates premises and hypotheses with their perceived joint probability p, to construct prompts, e.g. "[PREMISE]. [WEP], [HYPOTHESIS]." and assess whether language models can predict whether the WEP consensual probability level is close to p. Secondly, we construct a dataset of WEP-based probabilistic reasoning, to test whether language models can reason with WEP compositions. When prompted "[EVENTA] is likely. [EVENTB] is impossible.", a causal language model should not express that [EVENTA&B] is likely. We show that both tasks are unsolved by off-the-shelf English language models, but that fine-tuning leads to transferable improvement.
The explosion of e-commerce has caused the need for processing and analysis of product titles, like entity typing in product titles. However, the rapid activity in e-commerce has led to the rapid emergence of new entities, which is difficult to be solved by general entity typing. Besides, product titles in e-commerce have very different language styles from text data in general domain. In order to handle new entities in product titles and address the special language styles problem of product titles in e-commerce domain, we propose our textual entailment model with continuous prompt tuning based hypotheses and fusion embeddings for e-commerce entity typing. First, we reformulate the entity typing task into a textual entailment problem to handle new entities that are not present during training. Second, we design a model to automatically generate textual entailment hypotheses using a continuous prompt tuning method, which can generate better textual entailment hypotheses without manual design. Third, we utilize the fusion embeddings of BERT embedding and CharacterBERT embedding with a two-layer MLP classifier to solve the problem that the language styles of product titles in e-commerce are different from that of general domain. To analyze the effect of each contribution, we compare the performance of entity typing and textual entailment model, and conduct ablation studies on continuous prompt tuning and fusion embeddings. We also evaluate the impact of different prompt template initialization for the continuous prompt tuning. We show our proposed model improves the average F1 score by around 2% compared to the baseline BERT entity typing model.