Publications

Large Language Models (LLMs) are capable of performing zero-shot closed-book question answering tasks, based on their internal knowl-edge stored in parameters during pre-training. However, such internalized knowledge might be insufficient and incorrect, which could lead LLMs to generate factually wrong answers. Furthermore, fine-tuning LLMs to update their knowledge is expensive. To this end, we pro-pose

Recent NLP literature pays little attention to the robustness of toxicity language predictors, while these systems are most likely to be used in adversarial contexts. This paper presents a novel adversarial attack, ToxicTrap, introducing small word-level perturbations to fool SOTA text classifiers to predict toxic text samples as benign. ToxicTrap exploits greedy based search strategies to enable fast and

Pre-trained encoder-only and sequence-to-sequence (seq2seq) models each have advantages; however, training both model types from scratch is computationally expensive. We explore recipes to improve pre-training efficiency by initializing one model from the other. (1) Extracting the encoder from a seq2seq model, we show it underperforms a Masked Language Modeling (MLM) encoder, particularly on sequence labeling

This paper describes the speech translation system submitted as part of the IWSLT 2023 shared task on low resource speech translation. The low resource task aids in building models for language pairs where the training corpus is limited. In this paper, we focus on two language pairs, namely, Tamasheq-French (Tmh→Fra) and Marathi-Hindi (Mr→Hi) and implement a speech translation system that is unconstrained

Voice digital assistants must keep up with trending search queries. We rely on a speech recognition model using contextual biasing with a rapidly updated set of entities, instead of frequent model retraining, to keep up with trends. There are several challenges with this approach: (1) the entity set must be frequently reconstructed, (2) the entity set is of limited size due to latency and accuracy trade-offs

We evaluated the effectiveness of machine learning (ML) and natural language processing (NLP) for automatically scoring a simulation requiring audio-based constructed responses. We administered the simulation to 3,174 recent new professional-level hires working in a large multinational technology company. Human subject matter experts (SMEs) scored each response using behaviorally anchored rating scales

Missing information is a common issue of dialogue summarization where some information in the reference summaries is not covered in the generated summaries. To address this issue, we propose to utilize natural language inference (NLI) models to improve coverage while avoiding introducing factual inconsistencies. Specifically, we use NLI to compute fine-grained training signals to encourage the model to

The Open-Domain Question Answering (ODQA) task involves retrieving and subsequently generating answers from fine-grained relevant passages within a database. Current systems leverage Pre-trained Language Models (PLMs) to model the relationship between questions and passages. However, the diversity in surface form expressions can hinder the model’s ability to capture accurate correlations, especially within

Recent years have seen a surge of interest in learning high-level causal representations from low-level image pairs under interventions. Yet, existing efforts are largely limited to simple synthetic settings that are far away from real-world problems. In this paper, we present Causal Triplet, a causal representation learning benchmark featuring not only visually more complex scenes, but also two crucial

Causal discovery methods are intrinsically constrained by the set of assumptions needed to ensure structure identifiability. Moreover additional restrictions are often imposed in order to simplify the inference task: this is the case for the Gaussian noise assumption on additive nonlinear models, which is common to many causal discovery approaches. In this paper we show the shortcomings of inference under

Transformers have been recently utilized for vision and language tasks successfully. For example, recent image and language models with more than 200M parameters have been proposed to learn visual grounding in the pre-training step and show impressive results on downstream vision and language tasks. On the other hand, there exists a large amount of computational redundancy in these large models which skips

Large pre-trained language models (PLMs) have been shown to retain implicit knowledge within their parameters. To enhance this implicit knowledge, we propose Knowledge Injection into Language Models (KILM), a novel approach that injects entity-related knowledge into encoder-decoder PLMs, via a generative knowledge infilling objective through continued pre-training. This is done without architectural modifications

Automatic melody-to-lyric generation is a task in which song lyrics are generated to go with a given melody. It is of significant practical interest and more challenging than unconstrained lyric generation as the music imposes additional constraints onto the lyrics. The training data is limited as most songs are copyrighted, resulting in models that underfit the complicated cross-modal relationship between

Recent breakthroughs in synthetic data generation approaches made it possible to produce highly photorealistic images which are hardly distinguishable from real ones. Furthermore, synthetic generation pipelines have the potential to generate an unlimited number of images. The combination of high photorealism and scale turn synthetic data into a promising candidate for improving various machine learning

Deep Neural Networks (DNNs) are becoming integral components of real world services relied upon by millions of users. Unfortunately, architects of these systems can find it difficult to ensure reliable performance as irrelevant details like random initialization can unexpectedly change the outputs of a trained system with potentially disastrous consequences. We formulate the model stability problem by studying