People with disabilities experience high levels of social discrimination worldwide. But, these harms are more pronounced in the Global South due to the intense stigma around disability and its intersections with structural embeddings of patriarchy. The massive growth of social media in the Global South provides people with disabilities a unique opportunity to advocate for disability rights and challenge regressive ableist norms. Yet, little is known about the challenges they face in doing their advocacy work on social media. Through interviews with 20 disability advocates in India with diverse gender identities and abilities, we found that disability advocates routinely face ableist hate and harassment, patronizing and invalidating comments, and lack of visibility and support, which forces them to self-censor as a form of self-protection, leading to low advocacy outcomes. We draw on these findings to illuminate the role of social media in the invisibilization of people with disabilities in the online sphere.

Evaluation of multilingual Large Language Models (LLMs) is challenging due to a variety of factors -the lack of benchmarks with sufficient linguistic diversity, contamination of popular benchmarks into LLM pre-training data and the lack of local, cultural nuances in translated benchmarks.In this work, we study human and LLM-based evaluation in a multilingual, multi-cultural setting.We evaluate 30 models across 10 Indic languages by conducting 90K human evaluations and 30K LLMbased evaluations and find that models such as GPT-4o and Llama-3 70B consistently perform best for most Indic languages.We build leaderboards for two evaluation settings -pairwise comparison and direct assessment and analyse the agreement between humans and LLMs.We find that humans and LLMs agree fairly well in the pairwise setting but the agreement drops for direct assessment evaluation especially for languages such as Bengali and Odia.We also check for various biases in human and LLMbased evaluation and find evidence of self-bias in the GPT-based evaluator.Our work presents a significant step towards scaling up multilingual evaluation of LLMs. 1 few-shot cross-lingual transfer.In

The Deaf or Hard-of-Hearing (DHH) community constitutes over 430 million people globally, with about 70 million of them using sign language as their primary means of communication. Learning sign language poses significant challenges, and researchers have explored the potential of digital games as educational tools for teaching sign language. However, existing educational games are limited to a narrow range of words and feature only single-player modes. Consequently, our objective was to design a sign language-based game that not only facilitates learning but also enables social gameplay and encourages user-generated content. In this work, we present SignIt!, an accessible quiz platform co-designed with the DHH community. SignIt! empowers DHH players to play sign language-based quizzes either individually or in competitive settings, as well as to create their own quiz content. Through a user study involving five DHH participants, we found SignIt! was deemed easy-to-learn, intuitive, and accessible by our participants. They expressed various motivations for using SignIt!, including passing time, learning quiz content, and connecting with friends.

“CCGRID will act with responsibility as its primary consideration; with equity, diversity, and inclusion as its central goals.” from the CCGRID 2023 web site [1]

Current infrastructure design, discouragement by parents, and lack of internal motivation act as barriers for people with visual impairments (PVIs) to perform physical activities at par with sighted individuals. This has triggered accessible exercise technologies to be an emerging area of research. However, most current solutions have either safety concerns and/or are expensive, hence limiting their mass adoption. In our work, we propose VStroll, a smartphone app to promote walking among PVIs, by enabling them to virtually explore real-world locations, while physically walking in the safety and comfort of their homes. Walking is a cheap, accessible, and a common physical activity for people with blindness. VStroll has several added features, such as places-of-interest (POI) announcement using spatial audio and voice input for route selection at every intersection, which helps the user to gain spatial awareness while walking. To understand the usability of VStroll, 16 participants used our app for five days, followed by a semi-structured interview. Overall, our participants took 253 trips, walked for 50.8 hours covering 121.6 kms. We uncovered novel insights, such as discovering new POIs and fitness-related updates acted as key motivators, route selection boosted their confidence in navigation, and spatial audio resulted in an immersive experience. We conclude the paper with key lessons learned to promote accessible exercise technologies.

Most children who are blind live in low-resource settings and attend schools that have poor technical infrastructure, overburdened teachers, and outdated curriculum. Our work explores the role digital games can play to develop digital skills of such children in the early grades. Recognizing the critical role of teachers in introducing children to technology, we conducted a mixed-methods study to examine which attributes of digital games teachers find useful for children and what challenges they perceive in integrating digital games in schools for the blind. Our findings indicate that teachers prefer games that align well with curriculum objectives, promote learning, improve soft skills, and increase engagement with computers. Despite being overburdened and lacking technological support, teachers expressed strong enthusiasm to integrate these games in school curriculum and schedule. We conclude by discussing design implications for designers of accessible games in low-resource settings.

People with vision impairments access smartphones with the help of screen reader apps such as TalkBack for Android and VoiceOver for iPhone. Prior research has mostly focused on understanding touchscreen phone adoption and typing performance of novice blind users by logging their real-world smartphone usage. Understanding smartphone usage pattern and practices of expert users can help in developing tools and tutorials for transitioning novice and intermediate users to expert users. In this work, we logged smartphone usage data of eight expert Android smartphone users with visual impairments for four weeks, and then interviewed them. This paper presents a detailed analysis that uncovered novel usage patterns, such as extensive usage of directional gestures, reliance on voice and external keyboard for text input, and repurposed explore by touch for single-tap. We conclude with design recommendations to inform the future of mobile accessibility, including hardware guidelines and rethinking accessible software design.

Camera based assistive technologies such as smart glasses can provide people with visual impairments (PVIs) information about people in their vicinity. Although such ‘visually available’ information can enhance one’s social interactions, the privacy implications for bystanders from the perspective of PVIs remains underexplored. Motivated by prior findings of bystanders’ perspectives, we conducted two online surveys with visually impaired (N=128) and sighted (N=136) participants with two ‘field-of-view’ (FoV) experimental conditions related to whether information about bystanders was gathered from the front of the glasses or all directions. We found that PVIs considered it as ‘fair’ and equally useful to receive information from all directions. However, they reported being uncomfortable in receiving some visually apparent information (such as weight and gender) about bystanders as they felt it was ‘impolite’ or ‘improper’. Both PVIs and bystanders shared concerns about the fallibility of AI, where bystanders can be misrepresented by the devices. Our finding suggests that beyond issues of social stigma, both PVIs and bystanders have shared concerns that need to be considered to improve the social acceptability of camera based assistive technologies.

Ambient air pollution in urban areas is a significant health hazard, with over 4.2 million deaths annually attributed to it. A crucial step in tackling these challenge is to measure air quality at a fine spatiotemporal granularity. A promising approach for several smart city projects, called drive-by sensing, is to leverage vehicles retrofitted with different sensors (pollution monitors, etc.) that can provide the desired spatiotemporal coverage at a fraction of the cost. However, deploying a drive-by sensing network at a city-scale to optimally select vehicles from a large fleet is still unexplored. In this paper, we propose Modulo -- a system to bootstrap drive-by sensing deployment by taking into consideration a variety of aspects such as spatiotemporal coverage, budget constraints. Modulo is well-suited to satisfy unique deployment constraints such as colocations with other sensors (needed for gas and PM sensor calibration), etc. We compare Modulo with two baseline algorithms on real-world taxi and bus datasets. Modulo significantly outperforms the baselines when a fleet comprises of both taxis and fixed-route vehicles such as public transport buses. Finally, we present a real-world case study that uses Modulo to select vehicles for an air pollution sensing application.

Project Torino is a physical programming environment designed for teaching computational thinking (CT) to school children in the UK, regardless of the level of vision. We introduced Project Torino to children in three schools for the blind in Bangalore, India as a toy for playing with songs, rhymes, and stories. We present the results of 103 semi-structured play sessions spread over three months with 12 children (2 girls, 10 boys) with diverse backgrounds. We found that children progressed from playing with pre-connected examples, to making changes, to actively participating in what items are played. Engaging the children in conversation while they played, we established that the teams had grasped three basic concepts of computational thinking--flow of control, variables, and loops, without any explicit instructions towards learning them. We propose that play-based approaches can be successfully used with low resource overhead to introduce fundamental concepts of CT.

Fine-grained air pollution monitoring is a fundamental step towards curbing pollution levels. This is sought to be achieved by the large-scale deployment of low-cost sensors at high spatio-temporal resolution. Due to the nature of these deployments, in-the-wild and in harsh environments, sensors are prone to failures and hence ensuring data reliability is challenging. Furthermore, detecting a fault by analyzing the sensor data using existing data-centric approaches is non-trivial. This demonstration presents a sensor fault diagnosis system that employs the current signature of the sensor to address data reliability issues. The current signature captures the electrical characteristics of the hardware components enabling accurate detection and isolation of faults in low-cost pollution sensors.

Torino is a tangible programming environment designed for teaching the computational thinking curriculum in the UK to children who are blind or low vision (henceforth, just children) in an inclusive setting. In this paper we describe the experience of children in Bangalore, India, when Torino was introduced to them as a toy for creating and sharing stories, songs and music. We conducted 12 play sessions with 12 children (4 girls and 8 boys) with diverse backgrounds belonging to three different schools for the blind. We briefly present the reasons for play being central to our effort of bringing computational thinking to children who are blind and low vision in India, and share some experiences of the children and some insights that we have gathered so far: Children not only enjoyed every session, they rapidly moved from playing with pre-created examples, to making changes, to demanding that their favorite stories be told. In observing such play, we could infer that they have grasped the basic concepts of computational thinking? flow of control, variables, loops? though not articulated in that vocabulary.

Recent work in networked systems has shown that using aerial imagery for farm monitoring can enable precision agriculture by lowering the cost and reducing the overhead of large scale sensor deployment. However, acquiring aerial imagery requires a drone, which has high capital and operational costs, often beyond the reach of farmers in the developing world. In this paper, we present TYE (Tethered eYE), an inexpensive platform for aerial imagery. It consists of a tethered helium balloon with a custom mount that can hold a smartphone (or a camera) with a battery pack. The balloon can be carried using a tether by a person or a vehicle. We incorporate various techniques to increase the operational time of the system, and to provide actionable insights even with unstable imagery. We develop path-planning algorithms and use that to develop an interactive mobile phone application that provides the user instant feedback to guide users to efficiently traverse large areas of land. We use computer vision algorithms to stitch orthomosaics by effectively countering wind-induced motion of the camera. We have used TYE for aerial imaging of agricultural land for over a year, and envision it as a low-cost aerial imaging platform for similar applications.

Drive-by sensing has emerged as a popular way to achieve fine-grained sensing of physical phenomena. However, for it to be effective at a city-scale, there is a need to optimally select a subset of vehicles from a larger available fleet. These chosen vehicles must maximize coverage of the entire city. Simultaneously, they must fulfill other deployment requirements specific to the sensing application such as reference-monitor colocation instances for gas sensors. In this paper, we describe a system to evaluate the coverage offered by different subsets of vehicles for sensor deployment based on historical vehicle mobility data. Our system allows evaluation of different vehicle selection algorithms, and also provides two in-built baselines --- i) Random-MP, and ii) MaxPoints --- for comparison. Finally, we provide visualizations showing coverage to gauge the efficacy of different vehicle selections.

Mainstream video games are predominantly inaccessible to people with visual impairments (VIPs). We present ongoing research that aims to make such games go beyond accessibility, by making them engaging and enjoyable for visually impaired players. We have built a new interaction toolkit called the Responsive Spatial Audio Cloud (ReSAC), developed around spatial audio technology, to enable visually impaired players to play video games. VIPs successfully finished a simple video game integrated with ReSAC and reported enjoying the experience.

In recent times, programming environments like Visual Studio are widely used to enhance programmer productivity. However, inadequate accessibility prevents Visually Impaired (VI) developers from taking full advantage of these environments. In this paper, we focus on the accessibility challenges faced by the VI developers in using Graphical User Interface (GUI) based programming environments. Based on a survey of VI developers and based on two of the authors' personal experiences, we categorize the accessibility difficulties into Discoverability, Glanceability, Navigability, and Alertability. We propose solutions to some of these challenges and implement these in CodeTalk, a plugin for Visual Studio. We show how CodeTalk improves developer experience and share promising early feedback from VI developers who used our plugin.

The proliferation of Internet of Things (IoT) devices has led to the deployment of various types of sensors in the homes, offices, buildings, lawns, cities, and even in agricultural farms. Since IoT applications rely on the fidelity of data reported by the sensors, it is important to detect a faulty sensor and isolate the cause of the fault. Existing fault detection techniques demand sensor domain knowledge along with the contextual information and historical data from similar near-by sensors. However, detecting a sensor fault by analyzing just the sensor data is non-trivial since a faulty sensor reading could mimic non-faulty sensor data. This paper presents a novel primitive, which we call the Fall-curve - a sensor's voltage response when the power is turned off - that can be used to characterize sensor faults. The Fall-curve constitutes a unique signature independent of the phenomenon being monitored which can be used to identify the sensor and determine whether the sensor is correctly operating.

The proliferation of Internet of Things (IoT) devices has led to the deployment of various types of sensors in the homes, offices, buildings, lawns, cities, and even in agricultural farms. Due to the diverse nature of IoT deployments and the likelihood of sensor failures in-the-wild, a key challenge in the design of IoT systems is ensuring the integrity, accuracy, and fidelity of sensor data.