arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

Rahul Goel

79 papers · 2006–2026

2006: 12007: 02008: 02009: 02010: 02011: 32012: 02013: 12014: 02015: 12016: 12017: 02018: 102019: 92020: 52021: 82022: 142023: 72024: 82025: 62026: 5
2006peak 14 in 20222026

Frequent co-authors

Papers

showing 1–25 of 79 RSS

Gemini: A Family of Highly Capable Multimodal Models

Gemini Team, Rohan Anil, Sebastian Borgeaud +1348 more

cs.CL · · cited 816× · arXiv ↗ · more like this ·

This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the state of the art in every one of the 20 multimodal benchmarks we examined. We believe that the new capabilities of the Gemini family in cross-modal reasoning and language understanding will enable a wide variety of use cases. We discuss our approach toward post-training and deploying Gemini models…

Physical Activity Promotion and the United Nations Sustainable Development Goals: Building Synergies to Maximize Impact

Deborah Salvo, Leandro García, Rodrigo Siqueira Reis +6 more

Transportation · Journal of Physical Activity and Health · · cited 213× · DOI ↗ · more like this ·

BACKGROUND: Many of the known solutions to the physical inactivity pandemic operate across sectors relevant to the United Nations Sustainable Development Goals (SDGs). METHODS: The authors examined the contribution of physical activity promotion strategies toward achieving the SDGs through a conceptual linkage exercise, a scoping review, and an agent-based model. RESULTS: Possible benefits of physical activity promotion were identified for 15 of the 17 SDGs, with more robust evidence supporting benefits for SDGs 3 (good health and well-being), 9 (industry, innovation, and infrastructure), 11 (sustainable cities and communities), 13 (climate action), and 16 (peace, justice, and strong institutions). Current evidence supports prioritizing at-scale physical activity-promoting transport and urban design strategies and community-based programs. Expected physical activity gains are greater…

TableFormer: Robust Transformer Modeling for Table-Text Encoding

Jingfeng Yang, Aditya Gupta, Shyam Upadhyay +3 more

cs.CL · · cited 161× · arXiv ↗ · more like this ·

Understanding tables is an important aspect of natural language understanding. Existing models for table understanding require linearization of the table structure, where row or column order is encoded as an unwanted bias. Such spurious biases make the model vulnerable to row and column order perturbations. Additionally, prior work has not thoroughly modeled the table structures or table-text alignments, hindering the table-text understanding ability. In this work, we propose a robust and structurally aware table-text encoding architecture TableFormer, where tabular structural biases are incorporated completely through learnable attention biases. TableFormer is (1) strictly invariant to row and column orders, and, (2) could understand tables better due to its tabular inductive biases. Our evaluations showed that TableFormer outperforms strong baselines in all settings on SQA, WTQ and…

Gender differences in active travel in major cities across the world

Rahul Goel, Oyinlola Oyebode, Louise Foley +3 more

Transportation · Transportation · · cited 138× · DOI ↗ · more like this ·

There is lack of literature on international comparison of gender differences in the use of active travel modes. We used population-representative travel surveys for 19 major cities across 13 countries and 6 continents, representing a mix of cites from low-and-middle income (n = 8) and high-income countries (n = 11). In all the cities, females are more likely than males to walk and, in most cities, more likely to use public transport. This relationship reverses in cycling, with females often less likely users than males. In high cycling cities, both genders are equally likely to cycle. Active travel to access public transport contributes 30-50% of total active travel time. The gender differences in active travel metrics are age dependent. Among children (< 16 years), these metrics are often equal for girls and boys, while gender disparity increases with age. On average, active travel…

MultiWOZ 2.1: Multi-Domain Dialogue State Corrections and State Tracking Baselines

Mihail Eric, Rahul Goel, Shachi Paul +4 more

Artificial Intelligence · · cited 120× · OpenAlex ↗ · more like this ·

MultiWOZ is a recently-released multidomain dialogue dataset spanning 7 distinct domains and containing over 10000 dialogues, one of the largest resources of its kind to-date. Though an immensely useful resource, while building different classes of dialogue state tracking models using MultiWOZ, we detected substantial errors in the state annotations and dialogue utterances which negatively impacted the performance of our models. In order to alleviate this problem, we use crowdsourced workers to fix the state annotations and utterances in the original version of the data. Our correction process results in changes to over 32% of state annotations across 40% of the dialogue turns. In addition, we fix 146 dialogue utterances throughout the dataset focusing in particular on addressing slot value errors represented within the conversations. We then benchmark a number of state-of-the-art…

Access–egress and other travel characteristics of metro users in Delhi and its satellite cities

Rahul Goel, Geetam Tiwari

Transportation · IATSS Research · · cited 100× · DOI ↗ · more like this ·

We present details of access–egress as well as other travel characteristics of metro users in Delhi, and its satellite cities. For this study, we conducted an on-board survey of metro commuters ( n = 1112 ) in 2011. Survey respondents reported use of seven different modes to access metro stations, with 55% using non-motorized modes. The alternative modes reported by the respondents indicate that, in the absence of metro, a majority of commuters will use bus, and up to 40% will use private motorized modes. Up to 18% of the respondents may not have made the trips if metro was not available, indicating a significant proportion of induced trips. We used multinomial logistic regression models to understand the factors associated with the choice of access–egress modes, as well as alternative mode, for the current trip. Trip length, vehicle ownership, location of metro station (classified as…

On Evaluating and Comparing Open Domain Dialog Systems

Anu Venkatesh, Chandra Khatri, Ashwin Ram +10 more

cs.CL · · cited 86× · arXiv ↗ · more like this ·

Conversational agents are exploding in popularity. However, much work remains in the area of non goal-oriented conversations, despite significant growth in research interest over recent years. To advance the state of the art in conversational AI, Amazon launched the Alexa Prize, a 2.5-million dollar university competition where sixteen selected university teams built conversational agents to deliver the best social conversational experience. Alexa Prize provided the academic community with the unique opportunity to perform research with a live system used by millions of users. The subjectivity associated with evaluating conversations is key element underlying the challenge of building non-goal oriented dialogue systems. In this paper, we propose a comprehensive evaluation strategy with multiple metrics designed to reduce subjectivity by selecting metrics which correlate well with human…

Interactive Segmentation of Radiance Fields

Rahul Goel, Dhawal Sirikonda, Saurabh Saini +1 more

cs.CV · · cited 78× · arXiv ↗ · more like this ·

Radiance Fields (RF) are popular to represent casually-captured scenes for new view synthesis and several applications beyond it. Mixed reality on personal spaces needs understanding and manipulating scenes represented as RFs, with semantic segmentation of objects as an important step. Prior segmentation efforts show promise but don't scale to complex objects with diverse appearance. We present the ISRF method to interactively segment objects with fine structure and appearance. Nearest neighbor feature matching using distilled semantic features identifies high-confidence seed regions. Bilateral search in a joint spatio-semantic space grows the region to recover accurate segmentation. We show state-of-the-art results of segmenting objects from RFs and compositing them to another scene, changing appearance, etc., and an interactive segmentation tool that others can use. Project Page…

HyST: A Hybrid Approach for Flexible and Accurate Dialogue State Tracking

Rahul Goel, Shachi Paul, Dilek Hakkani-Tür

cs.CL · · cited 73× · arXiv ↗ · more like this ·

Recent works on end-to-end trainable neural network based approaches have demonstrated state-of-the-art results on dialogue state tracking. The best performing approaches estimate a probability distribution over all possible slot values. However, these approaches do not scale for large value sets commonly present in real-life applications and are not ideal for tracking slot values that were not observed in the training set. To tackle these issues, candidate-generation-based approaches have been proposed. These approaches estimate a set of values that are possible at each turn based on the conversation history and/or language understanding outputs, and hence enable state tracking over unseen values and large value sets however, they fall short in terms of performance in comparison to the first group. In this work, we analyze the performance of these two alternative dialogue state…

Online Embedding Compression for Text Classification using Low Rank Matrix Factorization

Anish Acharya, Rahul Goel, Angeliki Metallinou +1 more

cs.LG · · cited 69× · arXiv ↗ · more like this ·

Deep learning models have become state of the art for natural language processing (NLP) tasks, however deploying these models in production system poses significant memory constraints. Existing compression methods are either lossy or introduce significant latency. We propose a compression method that leverages low rank matrix factorization during training,to compress the word embedding layer which represents the size bottleneck for most NLP models. Our models are trained, compressed and then further re-trained on the downstream task to recover accuracy while maintaining the reduced size. Empirically, we show that the proposed method can achieve 90% compression with minimal impact in accuracy for sentence classification tasks, and outperforms alternative methods like fixed-point quantization or offline word embedding compression. We also analyze the inference time and storage space for…

Towards Coherent and Engaging Spoken Dialog Response Generation Using Automatic Conversation Evaluators

Sanghyun Yi, Rahul Goel, Chandra Khatri +6 more

cs.CL · · cited 67× · arXiv ↗ · more like this ·

Encoder-decoder based neural architectures serve as the basis of state-of-the-art approaches in end-to-end open domain dialog systems. Since most of such systems are trained with a maximum likelihood~(MLE) objective they suffer from issues such as lack of generalizability and the generic response problem, i.e., a system response that can be an answer to a large number of user utterances, e.g., "Maybe, I don't know." Having explicit feedback on the relevance and interestingness of a system response at each turn can be a useful signal for mitigating such issues and improving system quality by selecting responses from different approaches. Towards this goal, we present a system that evaluates chatbot responses at each dialog turn for coherence and engagement. Our system provides explicit turn-level dialog quality feedback, which we show to be highly correlated with human evaluation. To…

Investigating the association between population density and travel patterns in Indian cities—An analysis of 2011 census data

Rahul Goel, Dinesh Mohan

Transportation · Cities · · cited 47× · DOI ↗ · more like this ·

Many transport planners consider urban population density to be a significant determinant of travel behaviour. Much of the evidence for this comes from research in low-density, high-income settings. The 2011 Census of India reported mode of travel to work and distance for the first time. We have used these data to investigate the effect of urban density on commute travel patterns at city-level for Indian cities. In addition, we investigated the relationship between travel behaviour and other city-level variables. Using regression, we found almost no independent effect of density on the mode share of walk, cycle, motorised two-wheelers, cars and public transport, after controlling for population and income levels for the cities. Further, it appears that once density levels are greater than ~80 persons per hectare (pph), other factors become more important in determining travel patterns…

Mobility Based SIR Model For Pandemics -- With Case Study Of COVID-19

Rahul Goel, Rajesh Sharma

cs.SI · · cited 39× · arXiv ↗ · more like this ·

In the last decade, humanity has faced many different pandemics such as SARS, H1N1, and presently novel coronavirus (COVID-19). On one side, scientists are focusing on vaccinations, and on the other side, there is a need to propose models that can help us in understanding the spread of these pandemics as it can help governmental and other concerned agencies to be well prepared, especially from pandemics, which spreads faster like COVID-19. The main reason for some epidemic turning into pandemics is the connectivity among different regions of the world, which makes it easier to affect a wider geographical area, often worldwide. In addition, the population distribution and social coherence in the different regions of the world is non-uniform. Thus, once the epidemic enters a region, then the local population distribution plays an important role. Inspired by these ideas, we proposed a…

Health impacts of changes in travel patterns in Greater Accra Metropolitan Area, Ghana

Leandro García, Rob Johnson, Alex Johnson +8 more

Transportation · Environment International · · cited 36× · DOI ↗ · more like this ·

BACKGROUND: Health impact assessments of alternative travel patterns are urgently needed to inform transport and urban planning in African cities, but none exists so far. OBJECTIVE: To quantify the health impacts of changes in travel patterns in the Greater Accra Metropolitan Area, Ghana. METHODS: We estimated changes to population exposures to physical activity, air pollution, and road traffic fatality risk and consequent health burden (deaths and years of life lost prematurely - YLL) in response to changes in transportation patterns. Five scenarios were defined in collaboration with international and local partners and stakeholders to reflect potential local policy actions. RESULTS: Swapping bus and walking trips for car trips can lead to more than 400 extra deaths and 20,500 YLL per year than travel patterns observed in 2009. If part of the rise in motorisation is from motorcycles…

Detecting Offensive Content in Open-domain Conversations using Two Stage Semi-supervision

Chandra Khatri, Behnam Hedayatnia, Rahul Goel +3 more

cs.CL · · cited 33× · arXiv ↗ · more like this ·

As open-ended human-chatbot interaction becomes commonplace, sensitive content detection gains importance. In this work, we propose a two stage semi-supervised approach to bootstrap large-scale data for automatic sensitive language detection from publicly available web resources. We explore various data selection methods including 1) using a blacklist to rank online discussion forums by the level of their sensitiveness followed by randomly sampling utterances and 2) training a weakly supervised model in conjunction with the blacklist for scoring sentences from online discussion forums to curate a dataset. Our data collection strategy is flexible and allows the models to detect implicit sensitive content for which manual annotations may be difficult. We train models using publicly available annotated datasets as well as using the proposed large-scale semi-supervised datasets. We evaluate…

Alexa Conversations: An Extensible Data-driven Approach for Building Task-oriented Dialogue Systems

Anish Acharya, Suranjit Adhikari, Sanchit Agarwal +28 more

cs.CL · · cited 27× · arXiv ↗ · more like this ·

Traditional goal-oriented dialogue systems rely on various components such as natural language understanding, dialogue state tracking, policy learning and response generation. Training each component requires annotations which are hard to obtain for every new domain, limiting scalability of such systems. Similarly, rule-based dialogue systems require extensive writing and maintenance of rules and do not scale either. End-to-End dialogue systems, on the other hand, do not require module-specific annotations but need a large amount of data for training. To overcome these problems, in this demo, we present Alexa Conversations, a new approach for building goal-oriented dialogue systems that is scalable, extensible as well as data efficient. The components of this system are trained in a data-driven manner, but instead of collecting annotated conversations for training, we generate them…

Contextual Topic Modeling For Dialog Systems

Chandra Khatri, Rahul Goel, Behnam Hedayatnia +4 more

cs.CL · · cited 26× · arXiv ↗ · more like this ·

Accurate prediction of conversation topics can be a valuable signal for creating coherent and engaging dialog systems. In this work, we focus on context-aware topic classification methods for identifying topics in free-form human-chatbot dialogs. We extend previous work on neural topic classification and unsupervised topic keyword detection by incorporating conversational context and dialog act features. On annotated data, we show that incorporating context and dialog acts leads to relative gains in topic classification accuracy by 35% and on unsupervised keyword detection recall by 11% for conversational interactions where topics frequently span multiple utterances. We show that topical metrics such as topical depth is highly correlated with dialog evaluation metrics such as coherence and engagement implying that conversational topic models can predict user satisfaction. Our work for…

Misinformation Detection on YouTube Using Video Captions

Raj Jagtap, Abhinav Kumar, Rahul Goel +3 more

cs.LG · · cited 25× · arXiv ↗ · more like this ·

Millions of people use platforms such as YouTube, Facebook, Twitter, and other mass media. Due to the accessibility of these platforms, they are often used to establish a narrative, conduct propaganda, and disseminate misinformation. This work proposes an approach that uses state-of-the-art NLP techniques to extract features from video captions (subtitles). To evaluate our approach, we utilize a publicly accessible and labeled dataset for classifying videos as misinformation or not. The motivation behind exploring video captions stems from our analysis of videos metadata. Attributes such as the number of views, likes, dislikes, and comments are ineffective as videos are hard to differentiate using this information. Using caption dataset, the proposed models can classify videos among three classes (Misinformation, Debunking Misinformation, and Neutral) with 0.85 to 0.90 F1-score. To…

Flexible and Scalable State Tracking Framework for Goal-Oriented Dialogue Systems

Rahul Goel, Shachi Paul, Tagyoung Chung +3 more

cs.AI · · cited 15× · arXiv ↗ · more like this ·

Goal-oriented dialogue systems typically rely on components specifically developed for a single task or domain. This limits such systems in two different ways: If there is an update in the task domain, the dialogue system usually needs to be updated or completely re-trained. It is also harder to extend such dialogue systems to different and multiple domains. The dialogue state tracker in conventional dialogue systems is one such component - it is usually designed to fit a well-defined application domain. For example, it is common for a state variable to be a categorical distribution over a manually-predefined set of entities (Henderson et al., 2013), resulting in an inflexible and hard-to-extend dialogue system. In this paper, we propose a new approach for dialogue state tracking that can generalize well over multiple domains without incorporating any domain-specific knowledge. Under…

MultiWOZ 2.1: A Consolidated Multi-Domain Dialogue Dataset with State Corrections and State Tracking Baselines

Mihail Eric, Rahul Goel, Shachi Paul +7 more

cs.CL · · cited 13× · arXiv ↗ · more like this ·

MultiWOZ 2.0 (Budzianowski et al., 2018) is a recently released multi-domain dialogue dataset spanning 7 distinct domains and containing over 10,000 dialogues. Though immensely useful and one of the largest resources of its kind to-date, MultiWOZ 2.0 has a few shortcomings. Firstly, there is substantial noise in the dialogue state annotations and dialogue utterances which negatively impact the performance of state-tracking models. Secondly, follow-up work (Lee et al., 2019) has augmented the original dataset with user dialogue acts. This leads to multiple co-existent versions of the same dataset with minor modifications. In this work we tackle the aforementioned issues by introducing MultiWOZ 2.1. To fix the noisy state annotations, we use crowdsourced workers to re-annotate state and utterances based on the original utterances in the dataset. This correction process results in changes…

Understanding gender segregation through Call Data Records: An Estonian case study

Rahul Goel, Rajesh Sharma, Anto Aasa

Transportation · PLoS ONE · · cited 12× · DOI ↗ · more like this ·

Understanding segregation plays a significant role in determining the development pathways of a country as it can help governmental and other concerned agencies to prepare better-targeted policies for the needed groups. However, inferring segregation through alternative data, apart from governmental surveys remains limited due to the non-availability of representative datasets. In this work, we utilize Call Data Records (CDR) provided by one of Estonia's major telecom operators to research the complexities of social interaction and human behavior in order to explain gender segregation. We analyze the CDR with two objectives. First, we study gender segregation by exploring the social network interactions of the CDR. We find that the males are tightly linked which allows information to spread faster among males compared to females. Second, we perform the micro-analysis using various…

Distance-decay functions of travel to work trips in India

Rahul Goel

Transportation · Data in Brief · · cited 12× · DOI ↗ · more like this ·

In 2011, for the first time, Census in India reported travel distance and mode of travel for the workers. The distance reported is in the form of aggregate counts for each mode of travel in 7 distance bins (0-1 km, 2-5 km, 6-10 km, 11-20 km, 21-30 km, 31-50 km, and >50 km). In this data article, methods are described to model categorical count data as distance-decay functions using continuous probability distributions. The distributions have been developed for 8 categories of modes-walk, cycle, motorised two-wheelers, car, tempo/auto rickshaw/taxi, bus, train, and all modes combined, for the 33 mainland states of India and all states combined. Distance for walk is modelled using exponential distribution, and for all the other modes using lognormal or Weibull distribution. For estimating parameters of the distributions, chi-square minimization has been used in a spreadsheet program. The…

An integrated approach for optimal rail transit corridor identification and scheduling using geographical information system

Ashish Verma, Devendra Upadhyay, Rahul Goel

Transportation · Journal of King Saud University - Science · · cited 12× · DOI ↗ · more like this ·

In this work an integrated approach has been proposed to identify and schedule a rail transit corridor in a city which has a potential demand for a new rail based mass transit system besides the street transit system and existing rail-based system (if any). The motivation for the present study comes from the present Indian context where, because of recent initiatives of Central Urban Development Ministry, many Urban Local Bodies (ULBs) and transit agencies need to master plan new rail based systems in an optimum way and at the same time attempting to integrate them with other existing public transport (PT) modes. While aligning the new rail transit system, the primary focus of the integrated approach is to optimize the origin–destination (O–D) travel time of the user and at the same time attempting to minimize the operators’ cost. The proposed rail corridor identification model consists…

PRESTO: A Multilingual Dataset for Parsing Realistic Task-Oriented Dialogs

Rahul Goel, Waleed Ammar, Aditya Gupta +13 more

cs.CL · · cited 9× · arXiv ↗ · more like this ·

Research interest in task-oriented dialogs has increased as systems such as Google Assistant, Alexa and Siri have become ubiquitous in everyday life. However, the impact of academic research in this area has been limited by the lack of datasets that realistically capture the wide array of user pain points. To enable research on some of the more challenging aspects of parsing realistic conversations, we introduce PRESTO, a public dataset of over 550K contextual multilingual conversations between humans and virtual assistants. PRESTO contains a diverse array of challenges that occur in real-world NLU tasks such as disfluencies, code-switching, and revisions. It is the only large scale human generated conversational parsing dataset that provides structured context such as a user's contacts and lists for each example. Our mT5 model based baselines demonstrate that the conversational…

Population-level estimate of bicycle use and fatality risk in a data-poor setting

Rahul Goel

Transportation · International Journal of Injury Control and Safety Promotion · · cited 9× · DOI ↗ · more like this ·

Lack of data on exposure for walking and cycling poses a significant barrier to understanding the injury risk of these road users. Though this data paucity is most prevalent across low-and-middle-income countries (LMICs), it remains a challenge in many high-income countries as well. A new and simple method has been proposed to estimate population-level cycling distance travelled, with New Delhi, India as a case study. I used two independent estimates to calculate this distance. First, a ratio of motorcycle volume counts to that of cycle volume counts across major roads, and second, the total annual distance travelled by motorcycles. I validate this method using data from London, where cycling distance estimates are available from city-wide traffic volume counts as well as household travel survey. Combining the distance estimates with annual fatalities of corresponding road users, I…