Automatically generating a presentation from the text of a long document is a challenging and useful problem. In contrast to a flat summary, a presentation needs to have a better and non-linear narrative, i.e., the content of a slide can come from different and non-contiguous parts of the given document. However, it is difficult to incorporate such non-linear mapping of content to slides and ensure that the content is faithful to the document. LLMs are prone to hallucination and their performance degrades with the length of the input document. Towards this, we propose a novel graph based solution where we learn a graph from the input document and use a combination of graph neural network and LLM to generate a presentation with attribution of content for each slide. We conduct thorough experiments to show the merit of our approach compared to directly using LLMs for this task.

Generating presentation slides from a long document with multimodal elements such as text and images is an important task. This is time consuming and needs domain expertise if done manually. Existing approaches for generating a rich presentation from a document are often semi-automatic or only put a flat summary into the slides ignoring the importance of a good narrative. In this paper, we address this research gap by proposing a multi-staged end-to-end model which uses a combination of LLM and VLM. We have experimentally shown that compared to applying LLMs directly with state-of-the-art prompting, our proposed multi-staged solution is better in terms of automated metrics and human evaluation.

Repurposing existing content on-the-fly to suit author’s goals for creating initial drafts is crucial for document creation. We introduce the task of intent-guided and grounded document generation: given a user-specified intent (e.g., section title) and a few reference documents, the goal is to generate section-level multimodal documents spanning text and images, grounded on the given references, in a zero-shot setting. We present a data curation strategy to obtain general-domain samples from Wikipedia, and collect 1,000 Wikipedia sections consisting of textual and image content along with appropriate intent specifications and references. We propose a simple yet effective planning-based prompting strategy, Multimodal Plan-And-Write (MM-PAW), to prompt LLMs to generate an intermediate plan with text and image descriptions, to guide the subsequent generation. We compare the performances of MM-PAW and a text-only variant of it with those of zero-shot Chain-of-Thought (CoT) using recent close and open-domain LLMs. Both of them lead to significantly better performances in terms of content relevance, structure, and groundedness to the references, more so in the smaller models (upto 12.5 points increase in Rouge 1-F1) than in the larger ones (upto 4 points increase in R1-F1). They are particularly effective in improving relatively smaller models’ performances, to be on par or higher than those of their larger counterparts for this task.

Question generation methods based on pre-trained language models often suffer from factual inconsistencies and incorrect entities and are not answerable from the input paragraph. Domain shift – where the test data is from a different domain than the training data - further exacerbates the problem of hallucination. This is a critical issue for any natural language application doing question generation. In this work, we propose an effective data processing technique based on de-lexicalization for consistent question generation across domains. Unlike existing approaches for remedying hallucination, the proposed approach does not filter training data and is generic across question-generation models. Experimental results across six benchmark datasets show that our model is robust to domain shift and produces entity-level factually consistent questions without significant impact on traditional metrics.

Long documents like contracts, financial documents, etc., are often tedious to read through. Linearly consuming (via scrolling or navigation through default table of content) these documents is time-consuming and challenging. These documents are also authored to be consumed by varied entities (referred to as persona in the paper) interested in only certain parts of the document. In this work, we describe DynamicToC, a dynamic table of content-based navigator, to aid in the task of non-linear, persona-based document consumption. DynamicToC highlights sections of interest in the document as per the aspects relevant to different personas. DynamicToC is augmented with short questions to assist the users in understanding underlying content. This uses a novel deep-reinforcement learning technique to generate questions on these persona-clustered paragraphs. Human and automatic evaluations suggest the efficacy of both end-to-end pipeline and different components of DynamicToC.

This paper describes our system (IREL, reffered as himanshu.1007 on Codalab) for Shared Task on Empathy Detection, Emotion Classification, and Personality Detection at 12th Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis at ACL 2022. We participated in track 2 for predicting emotion at the essay level. We propose an ensemble approach that leverages the linguistic knowledge of the RoBERTa, BART-large, and RoBERTa model finetuned on the GoEmotions dataset. Each brings in its unique advantage, as we discuss in the paper. Our proposed system achieved a Macro F1 score of 0.585 and ranked one out of thirteen teams

This paper describes our system (IREL) for 3C-Citation Context Classification shared task of the Scholarly Document Processing Workshop at NAACL 2021. We participated in both subtask A and subtask B. Our best system achieved a Macro F1 score of 0.26973 on the private leaderboard for subtask A and was ranked one. For subtask B our best system achieved a Macro F1 score of 0.59071 on the private leaderboard and was ranked two. We used similar models for both the subtasks with some minor changes, as discussed in this paper. Our best performing model for both the subtask was a finetuned SciBert model followed by a linear layer. This paper provides a detailed description of all the approaches we tried and their results.