Recent advancements in integrating external tools with Large Language Models (LLMs) have opened new frontiers, with applications in mathematical reasoning, code generators, and smart assistants. However, existing methods, relying on simple one-time retrieval strategies, fall short on effectively and accurately shortlisting relevant tools. This paper introduces a novel PLUTO (Planning, Learning, and Understanding for TOols) approach, encompassing “Plan-and-Retrieve (P&R)” and “Edit-and-Ground (E&G)” paradigms. The P&R paradigm consists of a neural retrieval module for shortlisting relevant tools and an LLM-based query planner that decomposes complex queries into actionable tasks, enhancing the effectiveness of tool utilization. The E&G paradigm utilizes LLMs to enrich tool descriptions based on user scenarios, bridging the gap between user queries and tool functionalities. Experiment results demonstrate that these paradigms significantly improve the recall and NDCG in tool retrieval tasks, significantly surpassing current state-of-the-art models.

Natural Language Generation (NLG) typically involves evaluating the generated text in various aspects (e.g., consistency and naturalness) to obtain a comprehensive assessment. However, multi-aspect evaluation remains challenging as it may require the evaluator to generalize to any given evaluation aspect even if it’s absent during training. In this paper, we introduce X-Eval, a two-stage instruction tuning framework to evaluate text in both seen and unseen aspects customized by end users. X-Eval consists of two learning stages: the vanilla instruction tuning stage that improves the model’s ability to follow evaluation instructions, and an enhanced instruction tuning stage that exploits the connections between fine-grained evaluation aspects to better assess text quality. To support the training of X-Eval, we collect AspectInstruct, the first instruction tuning dataset tailored for multi-aspect NLG evaluation spanning 27 diverse evaluation aspects with 65 tasks. To enhance task diversity, we devise an augmentation strategy that converts human rating annotations into diverse forms of NLG evaluation tasks, including scoring, comparison, ranking, and Boolean question answering. Extensive experiments across three essential categories of NLG tasks: dialogue generation, summarization, and data-to-text coupled with 21 aspects in meta-evaluation, demonstrate that X-Eval enables even a lightweight language model to achieve a comparable if not higher correlation with human judgments compared to the state-of-the-art NLG evaluators like GPT-4.

We propose CHRT (Control HiddenRepresentation Transformation) – a con-trolled language generation framework thatsteers large language models to generatetext pertaining to certain attributes (such astoxicity). CHRT gains attribute control bymodifying the hidden representation of thebase model through learned transformations. We employ a contrastive-learning frameworkto learn these transformations that can becombined to gain multi-attribute control. Theeffectiveness of CHRT is experimentallyshown by comparing it with seven baselinesover three attributes. CHRT outperforms all thebaselines in the task of detoxification, positivesentiment steering, and text simplificationwhile minimizing the loss in linguistic qualities. Further, our approach has the lowest inferencelatency of only 0.01 seconds more than thebase model, making it the most suitable forhigh-performance production environments. We open-source our code and release two noveldatasets to further propel controlled languagegeneration research

Large Language Models have been successful in a wide variety of Natural Language Processing tasks by capturing the compositionality of the text representations. In spite of their great success, these vector representations fail to capture meaning of idiomatic multi-word expressions (MWEs). In this paper, we focus on the detection of idiomatic expressions by using binary classification. We use a dataset consisting of the literal and idiomatic usage of MWEs in English and Portuguese. Thereafter, we perform the classification in two different settings: zero shot and one shot, to determine if a given sentence contains an idiom or not. N shot classification for this task is defined by N number of common idioms between the training and testing sets. In this paper, we train multiple Large Language Models in both the settings and achieve an F1 score (macro) of 0.73 for the zero shot setting and an F1 score (macro) of 0.85 for the one shot setting. An implementation of our work can be found at https://github.com/ashwinpathak20/Idiomaticity_Detection_Using_Few_Shot_Learning.

In this work, we examine the problems associated with neural dialog models under the common theme of compositionality. Specifically, we investigate three manifestations of compositionality: (1) Productivity, (2) Substitutivity, and (3) Systematicity. These manifestations shed light on the generalization, syntactic robustness, and semantic capabilities of neural dialog models. We design probing experiments by perturbing the training data to study the above phenomenon. We make informative observations based on automated metrics and hope that this work increases research interest in understanding the capacity of these models.

Word embeddings are the standard model for semantic and syntactic representations of words. Unfortunately, these models have been shown to exhibit undesirable word associations resulting from gender, racial, and religious biases. Existing post-processing methods for debiasing word embeddings are unable to mitigate gender bias hidden in the spatial arrangement of word vectors. In this paper, we propose RAN-Debias, a novel gender debiasing methodology that not only eliminates the bias present in a word vector but also alters the spatial distribution of its neighboring vectors, achieving a bias-free setting while maintaining minimal semantic offset. We also propose a new bias evaluation metric, Gender-based Illicit Proximity Estimate (GIPE), which measures the extent of undue proximity in word vectors resulting from the presence of gender-based predilections. Experiments based on a suite of evaluation metrics show that RAN-Debias significantly outperforms the state-of-the-art in reducing proximity bias (GIPE) by at least 42.02%. It also reduces direct bias, adding minimal semantic disturbance, and achieves the best performance in a downstream application task (coreference resolution).

Word embeddings are the standard model for semantic and syntactic representations of words. Unfortunately, these models have been shown to exhibit undesirable word associations resulting from gender, racial, and religious biases. Existing post-processing methods for debiasing word embeddings are unable to mitigate gender bias hidden in the spatial arrangement of word vectors. In this paper, we propose RAN-Debias, a novel gender debiasing methodology that not only eliminates the bias present in a word vector but also alters the spatial distribution of its neighboring vectors, achieving a bias-free setting while maintaining minimal semantic offset. We also propose a new bias evaluation metric, Gender-based Illicit Proximity Estimate (GIPE), which measures the extent of undue proximity in word vectors resulting from the presence of gender-based predilections. Experiments based on a suite of evaluation metrics show that RAN-Debias significantly outperforms the state-of-the-art in reducing proximity bias (GIPE) by at least 42.02%. It also reduces direct bias, adding minimal semantic disturbance, and achieves the best performance in a downstream application task (coreference resolution).

Question answering and conversational systems are often baffled and need help clarifying certain ambiguities. However, limitations of existing datasets hinder the development of large-scale models capable of generating and utilising clarification questions. In order to overcome these limitations, we devise a novel bootstrapping framework (based on self-supervision) that assists in the creation of a diverse, large-scale dataset of clarification questions based on post-comment tuples extracted from stackexchange. The framework utilises a neural network based architecture for classifying clarification questions. It is a two-step method where the first aims to increase the precision of the classifier and second aims to increase its recall. We quantitatively demonstrate the utility of the newly created dataset by applying it to the downstream task of question-answering. The final dataset, ClarQ, consists of ~2M examples distributed across 173 domains of stackexchange. We release this dataset in order to foster research into the field of clarification question generation with the larger goal of enhancing dialog and question answering systems.

Non-contextual word embedding models have been shown to inherit human-like stereotypical biases of gender, race and religion from the training corpora. To counter this issue, a large body of research has emerged which aims to mitigate these biases while keeping the syntactic and semantic utility of embeddings intact. This paper describes Fair Embedding Engine (FEE), a library for analysing and mitigating gender bias in word embeddings. FEE combines various state of the art techniques for quantifying, visualising and mitigating gender bias in word embeddings under a standard abstraction. FEE will aid practitioners in fast track analysis of existing debiasing methods on their embedding models. Further, it will allow rapid prototyping of new methods by evaluating their performance on a suite of standard metrics.

Non-contextual word embedding models have been shown to inherit human-like stereotypical biases of gender, race and religion from the training corpora. To counter this issue, a large body of research has emerged which aims to mitigate these biases while keeping the syntactic and semantic utility of embeddings intact. This paper describes Fair Embedding Engine (FEE), a library for analysing and mitigating gender bias in word embeddings. FEE combines various state of the art techniques for quantifying, visualising and mitigating gender bias in word embeddings under a standard abstraction. FEE will aid practitioners in fast track analysis of existing debiasing methods on their embedding models. Further, it will allow rapid prototyping of new methods by evaluating their performance on a suite of standard metrics.

This paper presents a highly effective pipeline for passage retrieval in a conversational search setting. The pipeline comprises of two components: Conversational Term Selection (CTS) and Multi-View Reranking (MVR). CTS is responsible for performing the first-stage of passage retrieval. Given an input question, it uses a BERT-based classifier (trained with weak supervision) to de-contextualize the input by selecting relevant terms from the dialog history. Using the question and the selected terms, it issues a query to a search engine to perform the first-stage of passage retrieval. On the other hand, MVR is responsible for contextualized passage reranking. It first constructs multiple views of the information need embedded within an input question. The views are based on the dialog history and the top documents obtained in the first-stage of retrieval. It then uses each view to rerank passages using BERT (fine-tuned for passage ranking). Finally, MVR performs a fusion over the rankings produced by the individual views. Experiments show that the above combination improves first-state retrieval as well as the overall accuracy in a reranking pipeline. On the key metric of NDCG@3, the proposed combination achieves a relative performance improvement of 14.8% over the state-of-the-art baseline and is also able to surpass the Oracle.

Word embeddings have become a staple of several natural language processing tasks, yet much remains to be understood about their properties. In this work, we analyze word embeddings in terms of their principal components and arrive at a number of novel and counterintuitive observations. In particular, we characterize the utility of variance explained by the principal components as a proxy for downstream performance. Furthermore, through syntactic probing of the principal embedding space, we show that the syntactic information captured by a principal component does not correlate with the amount of variance it explains. Consequently, we investigate the limitations of variance based embedding post-processing algorithms and demonstrate that such post-processing is counter-productive in sentence classification and machine translation tasks. Finally, we offer a few precautionary guidelines on applying variance based embedding post-processing and explain why non-isotropic geometry might be integral to word embedding performance.

Code-mixing is the phenomenon of mixing the vocabulary and syntax of multiple languages in the same sentence. It is an increasingly common occurrence in today’s multilingual society and poses a big challenge when encountered in different downstream tasks. In this paper, we present a hybrid architecture for the task of Sentiment Analysis of English-Hindi code-mixed data. Our method consists of three components, each seeking to alleviate different issues. We first generate subword level representations for the sentences using a CNN architecture. The generated representations are used as inputs to a Dual Encoder Network which consists of two different BiLSTMs - the Collective and Specific Encoder. The Collective Encoder captures the overall sentiment of the sentence, while the Specific Encoder utilizes an attention mechanism in order to focus on individual sentiment-bearing sub-words. This, combined with a Feature Network consisting of orthographic features and specially trained word embeddings, achieves state-of-the-art results - 83.54% accuracy and 0.827 F1 score - on a benchmark dataset.

Code-mixing, use of two or more languages in a single sentence, is ubiquitous; generated by multi-lingual speakers across the world. The phenomenon presents itself prominently in social media discourse. Consequently, there is a growing need for translating code-mixed hybrid language into standard languages. However, due to the lack of gold parallel data, existing machine translation systems fail to properly translate code-mixed text. In an effort to initiate the task of machine translation of code-mixed content, we present a newly created parallel corpus of code-mixed English-Hindi and English. We selected previously available English-Hindi code-mixed data as a starting point for the creation of our parallel corpus. We then chose 4 human translators, fluent in both English and Hindi, for translating the 6088 code-mixed English-Hindi sentences to English. With the help of the created parallel corpus, we analyzed the structure of English-Hindi code-mixed data and present a technique to augment run-of-the-mill machine translation (MT) approaches that can help achieve superior translations without the need for specially designed translation systems. We present an augmentation pipeline for existing MT approaches, like Phrase Based MT (Moses) and Neural MT, to improve the translation of code-mixed text. The augmentation pipeline is presented as a pre-processing step and can be plugged with any existing MT system, which we demonstrate by improving translations done by systems like Moses, Google Neural Machine Translation System (NMTS) and Bing Translator for English-Hindi code-mixed content.