We present LEGAL-UQA, the first Urdu legal question-answering dataset derived from Pakistan's constitution. This parallel English-Urdu dataset includes 619 question-answer pairs, each with corresponding legal article contexts, addressing the need for domain-specific NLP resources in low-resource languages. We describe the dataset creation process, including OCR extraction, manual refinement, and GPT-4-assisted translation and generation of QA pairs. Our experiments evaluate the latest generalist language and embedding models on LEGAL-UQA, with Claude-3.5-Sonnet achieving 99.19% human-evaluated accuracy. We fine-tune mt5-large-UQA-1.0, highlighting the challenges of adapting multilingual models to specialized domains. Additionally, we assess retrieval performance, finding OpenAI's text-embedding-3-large outperforms Mistral's mistral-embed. LEGAL-UQA bridges the gap between global NLP advancements and localized applications, particularly in constitutional law, and lays the foundation for improved legal information access in Pakistan.

The rapid evolution of Natural Language Processing (NLP) has favored major languages such as English, leaving a significant gap for many others due to limited resources. This is especially evident in the context of data annotation, a task whose importance cannot be underestimated, but which is time-consuming and costly. Thus, any dataset for resource-poor languages is precious, in particular when it is task-specific. Here, we explore the feasibility of repurposing existing datasets for a new NLP task: we repurposed the Belebele dataset (Bandarkar et al., 2023), which was designed for multiple-choice question answering (MCQA), to enable extractive QA (EQA) in the style of machine reading comprehension. We present annotation guidelines and a parallel EQA dataset for English and Modern Standard Arabic (MSA). We also present QA evaluation results for several monolingual and cross-lingual QA pairs including English, MSA, and five Arabic dialects. Our aim is to enable others to adapt our approach for the 120+ other language variants in Belebele, many of which are deemed under-resourced. We also conduct a thorough analysis and share our insights from the process, which we hope will contribute to a deeper understanding of the challenges and the opportunities associated with task reformulation in NLP research.

When Pakistan's government censored the media, former Prime Minister Imran Khan's party posted campaign videos on TikTok. When the police barred his supporters from holding rallies, they hosted virtual gatherings online. And when Khan ended up behind bars, his supporters produced speeches using artificial intelligence to simulate his voice. Khan's message resonated with millions across the country who were frustrated by the country's economic crisis and old political dynasties: Pakistan has been on a steep decline for decades, he explained, and only he could restore its former greatness.

Imran Khan, the ousted prime minister of Pakistan, discusses the situation that could see him being jailed by the country he recently led. Pakistan heads to the polls Thursday with one of the leading candidates languishing in prison as some reports claim that voter enthusiasm is lacking among the country's nearly 128 million voters. Former Pakistani Prime Minister Imran Khan, founder of Pakistan Tehreek-e-Insaf (PTI), has been barred from contesting the national parliamentary elections. Khan was ousted as the country's 22nd prime minister after a no-confidence vote in April 2022. The former cricketer-turned-politician is currently serving more than 30 years in jail.

Iran and Pakistan have carried out air attacks on each other's territories, targeting armed groups near their 900km-long (559-mile) volatile border, which they say were meant to ensure their respective national security. Iran's powerful Islamic Revolutionary Guard Corps (IRGC) targeted an armed group in Panjgur town of Pakistan's Balochistan province late on Tuesday, prompting Pakistan to bomb hideouts of armed groups in the Sistan-Baluchestan province of Iran early on Thursday. Let's take a look at why the neighbours have resorted to direct military strikes, who the targets were, and what the attacks tell us. The IRGC, an elite force which is a vital part of the Iranian establishment but separate from Iran's army, hit the Jaish al-Adl armed group with missile and drone strikes in a mountainous region in Pakistan close to the Iranian border. Iran said it targeted the Iranian "terrorist" group it blamed for recent attacks in the Iranian city of Rask in the southeastern province of Sistan-Baluchestan.

The Taliban's acting defence minister has said Pakistan allowed American drones to use its airspace to access Afghanistan, a charge Pakistan has recently denied following a US air strike in Kabul. Acting Minister of Defence Mullah Mohammad Yaqoob told reporters at a news conference in Kabul on Sunday that American drones have been entering Afghanistan via Pakistan. "According to our information the drones are entering through Pakistan to Afghanistan, they use Pakistan's airspace, we ask Pakistan, don't use your airspace against us," he said. Pakistan's foreign ministry did not immediately respond to a request for comment. Pakistani authorities have denied involvement in or advanced knowledge of a drone strike the United States said it carried out in Kabul in July that killed al-Qaeda leader Ayman al-Zawahiri.

When Imran Khan became prime minister of Pakistan in August 2018, the country's economy faced twin deficits – foreign exchange reserves were …

Among the many adverse impacts of climate change in the most vulnerable countries, climate change–induced displacement increasingly caused by extreme weather events is a serious concern, particularly in densely populated Asian countries. Reports by the Intergovernmental Panel on Climate Change (IPCC) project a grim picture for South Asia, the most populous region on Earth, home to about one-quarter of global population, with the highest poverty incidence. A combination of poor socioeconomic indicators and increased frequency and intensity of cyclones and floods renders the region extremely vulnerable. Meanwhile, slow-onset climate hazards, such as sea level rise, salinity intrusion, water stress, and crop failures gradually turn into larger disasters. Within South Asia, Bangladesh stands as the most vulnerable: 4.1 million people were displaced as a result of climate disasters in 2019 (2.5% of the population), 13.3 million people could be displaced by climate change by 2050, and 18% of its coastland will remain inundated by 2080 ([ 1 ][1]). We describe how, faced with such natural and human-made adversities, Bangladesh can stand as a model of disaster management, adaptation, and resilience. The Paris Agreement goal of keeping the temperature rise at 1.5°C or well below 2°C compared to pre-industrial times may not be achieved, given the lack of ambitious mitigation. As a result, the number of people estimated to be displaced by slow-onset events will stand at ∼22.5 million by 2030 and ∼34.4 million by 2050 ([ 2 ][2]). A combination of sudden and slow-onset climate events, which affect all elements of the environment, becomes the main driver of environmental displacement. Migration is an adaptation strategy. An estimated half a million people move to Dhaka, the capital city of Bangladesh, each year. Migration of this magnitude presents a challenge for Bangladesh given its small land area (147,570 km2) and high population density (∼1100/km2). There is simply little space for retreat: Bangladesh's population is half that of the United States, living on ∼1.5% of the land area of the United States. Usually, three pathways can be discerned with respect to how displaced people are settled: autonomous relocation by displaced individuals (without much government support), government-supported temporary settlement, and planned relocation. In Bangladesh, the first option overwhelms, followed by efforts for temporary settlement, until the government rehabilitates their former residences. Planned relocation or managed retreat in response to climate change ([ 3 ][3]) is not yet happening widely because of space and resource constraints. ![Figure][4] Building migrant-friendly, climate-resilient cities in Bangladesh The map shows some activities being undertaken to build migrant-friendly and climate-resilient cities in Bangladesh. Descriptions of activities are based on publicly available information about the programs, and on discussions with representatives of the NGO BRAC. GRAPHIC: N. DESAI/ SCIENCE Since the founding of Bangladesh in 1971, and even earlier in Pakistan, government-planned relocation of people displaced by riverbank erosion has fueled ethnic conflicts in the Chittagong Hill Tracts in the southeast part of the country, because the move was not backed by consultations with tribal communities. About 100,000 of more than a million Rohingya refugees in Bangladesh, fleeing persecution in Myanmar, are being relocated to Bhasan Char, an island in the Bay of Bengal. In land-hungry Bangladesh, most of the 30+ such Chars/mudflats in the bay are already inhabited at different degrees by people displaced by riverbank erosion and climate change. Despite these odds, Bangladesh is a leader in economic growth among developing countries and in mainstreaming climate change into its development strategy. Partially in response to scientific findings, the National Strategy on the Management of Disaster and Climate Induced Internal Displacement (N SMDCIID) adopted in 2015 incorporated disaster risk reduction and rights-based approaches, so that vulnerable communities can enjoy their basic rights to livelihood, food, health, and housing. The Strategy is built on an integrated Displacement Management Framework, in line with the migration management cycle of the International Organization of Migration (IOM). This Framework elaborates responses during the three phases of mobility management: pre-displacement [disaster risk reduction (DRR)], displacement (emergency), and post-displacement (rehabilitation/relocation). Under the Strategy, the government has initiated support for livelihood opportunities, housing, and human development of displaced people in vulnerable hotspots. It is likely that the government-supported community mobilization and disaster management and DRR policies, both before and after adoption of this Strategy, were helpful in lessening the number of casualties from the supercyclone Amphan in May 2020. One way to address displacements under increasing urbanization across the world could be the establishment of peri-urban growth centers and transformation of cities and towns to be migrant-friendly. This option appears practicable for populous countries such as Bangladesh, having little space for retreat from vulnerable hotspots. To achieve this, institutional changes in a city need to be fostered by research, planning, design, and capacity building. Examples from cities such as Durban, Quito, Semarang, and Malé indicate that cities may need to develop general as well as sector-based strategies to manage effective climate change adaptation ([ 4 ][5]). This warrants the linking of adaptation planning and implementation to city priorities. Cities must have access to reliable information and opportunities to share experiences through local, regional, national, and international networks ([ 4 ][5]). National and local governments should develop migrant-friendly plans along three lines: building of resilient hardware, such as low-cost housing, industries for employment generation, and other infrastructure; software, such as legal, policy, and institutional frameworks; and “heart-ware”—the promotion of awareness, reflecting values and ethics. The basic parameters for safe and orderly movement for migrants are to ensure employment, social protection, access to education, housing, health services, utilities, etc. Although government support is important, engagement of the private sector, nongovernmental organizations (NGOs), civil society, and university-led research can strengthen municipal adaptation efforts. This is what the International Centre for Climate Change and Development (ICCCAD) in Bangladesh has been doing—to facilitate the transformation of smaller peripheral towns to be migrant-friendly as a climate adaptation strategy (see the figure). Our work has multiple purposes: to shift the tide of migration away from Dhaka and other large cities toward smaller towns, and to decentralize climate-resilient development and facilitate planning for basic services and amenities. In Bangladesh, a majority of those displaced by climate change prefer non-migration from their ancestral roots ([ 5 ][6]) if they are provided support for improving their livelihood, housing, etc. Settlement of displaced people in a town nearer to their ancestral home allows them to maintain psychological kinship and cultural comforts. On the basis of such local context and needs, each migrant-friendly town needs its own development and adaptation plans to address climate risks and economic opportunities. The NGO BRAC has initially identified about 20 towns and municipalities, considering their economic potential and climate stress, to determine whether they can absorb a sizeable number of displaced people. A number of satellite towns adjacent to economic hubs, such as relatively elevated sea and river ports and export processing zones (EPZs), can potentially employ millions of migrants. Investment in manufacturing and/or services is generating jobs through public, private, and community partnerships, such as private investments, government support, and microfinancing from BRAC and Grameen Bank. ICCCAD has formal agreements with many ministries and agencies including the Local Government Engineering Department (LGED), the agency for building and maintenance of rural infrastructure. ICCCAD has been working as an advisor and co-implementer of programs with all stakeholders, including mayors in two small towns in coastal Bangladesh, Mongla and Noapara (see the figure). It is helping town authorities in planning and implementing initiatives that are intended to be hospitable to incoming settlers, so that they can gradually be mainstreamed into citizenship ([ 6 ][7]). The process is based on a participatory, consultative process involving the municipal authorities, host community leaders, and settlers. The Strategy (NSMDCIID) includes options such as supporting livelihood for new settlers and skill development, both in displacement hotspots and in new settlements. Although these towns do not yet have adaptation plans as such, the programs consider risk-informed and socially conducive adaptation measures. BRAC with its Climate Bridge Fund is also currently implementing different programs in five cities: Khulna, Rajshahi, Satkhira, Barisal, and Sirajgonj. For programs under implementation in these cities, the target groups are incoming migrants, who crowd the slums. The activities undertaken in these cities are similar, with some specific activities in each town (see the figure). Most of the new settlers have moved from rural areas rendered inhospitable as a consequence of slow and sudden-onset climate impacts. ICCCAD started facilitating this program 3 years ago with a strategy of learning by doing. Among the lessons learned: (i) Vibrant economic activities in these rapidly growing towns are absorbing increasing numbers of migrants from vulnerable hotspots, and (ii) migrants with energy and agency are engaging themselves in different small businesses, with government support and microcredits from Grameen Bank and BRAC. The fact that an overwhelming share of those displaced by climate change around the world resettle internally indicates that adaptation in-country is the most viable option. The global community dealing with disaster displacement, including the United Nations Framework Convention on Climate Change (UNFCCC), primarily recommends this option. However, it requires adequate international support, which developed countries are obligated to deliver (with the language “shall provide”) under the UNFCCC and the Paris Agreement. Unfortunately, adaptation finance continues to remain the “poor cousin” of mitigation, the ratio remaining 20:80 despite repeated pledges by developed countries and agencies. For domestic resource mobilization, some countries (for example, Fiji) have introduced an adaptation levy on all goods and services produced and consumed in the country. There are limits to relocation in-country; sudden and slow-onset events sometimes trigger cross-border movement of individuals seeking jobs and protection. The UN Commission on Human Rights argues for looking at such mobility from a human-rights perspective (i.e., the space for realizing the basic human rights of livelihood, health, housing, etc.). Currently, those displaced by climate change suffer an international protection deficit, not qualifying as “refugees” under the 1951 Geneva Convention. Consideration of those displaced by climate change began in 2008 under the UNFCCC, with research and advocacy. The Cancun Adaptation Framework (Decision 1./CP16, paragraph 14f ) provides for different types of climate-induced human mobility (displacement, migration, and planned relocation), different scales of mobility (national, regional, and international), and different actions (research, cooperation, and coordination). This decision recognized migration as an adaptation strategy. The Nansen Initiative in 2011–2012 focused on promoting research and planned relocation. The Paris Agreement established a Task Force on Displacement under the Warsaw International Mechanism, with mandates to make recommendations for averting, minimizing, and addressing climate change–induced displacement. Finally, the Global Compact on Safe, Orderly, Regular and Responsive Migration was adopted in 2018 as the first multilateral framework to cooperate on migration, including in response to climate change. Many major countries and think tanks started looking at climate displacement through a lens of national security, with its characterization as a “threat multiplier,” and a number of nationally determined contributions under the Paris Agreement refer to those displaced by climate change as potentially fueling national and regional conflicts ([ 7 ][8]). However, climate security can be looked at either from a conflict perspective or from a lens of vulnerability-focused human and global security ([ 8 ][9]). The “conflict view” proponents call for closing the borders, but still the result of such a policy ends up being a humanitarian disaster, caused primarily by actions beyond the control of those being displaced or of their home countries. Should we see more of these displaced and disgruntled youth as victims in the hands of human traffickers? If not, we then argue—viewing this displacement in terms of vulnerability-focused human security—that planned relocation internationally can be an effective way forward under paragraph 14f of the Cancun agreement. As multilateral processes are typically very cumbersome and painstakingly slow, bilateral action can be more rapid and effective, and may then gradually feed into regional and global initiatives. For example, the Seasonal Migrant Worker Program in Australia and New Zealand, or New Zealand's Climate Visa Program ([ 9 ][10]), attract migrants from the Pacific Small Island States (although these initiatives are not solely meant for absorbing migrants displaced by climate change). Canada and the United States offered immigration opportunities to typhoon Haiyan victims, but these were based on kinship relations ([ 10 ][11]). Although the EU does not have a common policy, Finland and Sweden changed their earlier liberal policies on climate-induced displacement after the refugee influx from Syria ([ 11 ][12]). There are also provisions of circular migration, as between Spain and Colombia. The IOM continues recommending such migration between developed and developing countries as an adaptation response to climate-induced vulnerability. The Bangladesh Strategy recommends such options as well. Many developed countries already suffer from demographic deficits, with negative growth, and increasingly aging cohorts. The rhetoric in many of these counties, which often is anti-immigrant, cannot change the reality that these countries will need more and more young and skilled labor. Using projected needs of specific skills, developed countries could thus enter into bilateral agreements with climate-vulnerable countries, where those displaced by climate change may be trained in jointly supported educational and training institutions, either for permanent or for circular migration. For example, under the “Triple Win” program, Germany recruits nurses from Serbia, Bosnia-Herzegovina, and the Philippines to meet their nursing shortage, while reducing unemployment and contributing to economic development in the countries of origin ([ 12 ][13]). It is only just and fair for developed-country emitters of greenhouse gases to take some responsibility under Article 3.1 of the UNFCCC for their disproportionate contributions to generating this increasing number of people displaced by climate change. Lessons suggest that migration to rich countries can have strong positive impacts on labor market, GDP growth, and public revenue for host countries ([ 13 ][14], [ 14 ][15]). Mig ration is also typically positive for countries of origin, through remittance, transfer of technology, skills, domestic consumption and GDP growth, housing, children's education, and more. In 2017, low- and middle-income countries received more than $466 billion in remittances, three times the amount of official aid ([ 15 ][16]). This presents an important indicator of the effects that bilateral agreements on migration of climate-displaced people may have on promoting many different Sustainable Development Goals. Such migration should be framed as a win-win option, not as climate humanitarianism ([ 10 ][11]). The Bangladesh Strategy (NSMDCIID) argues for creating “opportunities for international labor migration by one or few members of families from the displacement hotspots” (p. 115). Older and underage family members and spouses can stay behind and rebuild their lives with remittance support. We believe this option of selective, not wholesale, relocation as a pragmatic policy can be scaled gradually, as warranted by projected demands of skills over time in developed countries. This relocation is based on bilateral planning and preparation, unlike the conventional, voluntary migration of skilled labor to industrial countries. This option is challenging, though mutually rewarding. However, acceptance of this proposal by Western democracies depends on whether they are ready to embrace and enjoy more of “smart/pooled” sovereignty, with enlightened self-interests under climate-induced vulnerability interdependence, rather than holding on to a centuries-old “Westphalian” model of a zero-sum game in global cooperation. Many have argued that with the increasing number of global commons problems, we now live in a positive-sum world. But such a paradigm shift warrants a vigorous campaign to raise awareness among citizens in industrial countries about the “new normal” of increasing extreme and ever-growing slow-onset events. Those citizens and politicians must face the lead and obligatory responsibility their countries have assumed under the international climate regime to support adaptation in vulnerable countries. Such awareness must confront and overcome the xenophobia and anti-immigration sentiments that often surface in many countries, inhibiting the enjoyment of mutual dividends, which can contribute to real and sustainable global peace and security. Successful implementation of the two options raised above (migrant-friendly towns and bilateral agreements for international migration) could help to germinate coordinated implementation, as stipulated in the Cancun agreement, of global policy frameworks on climate change (UNFCCC), disaster risk reduction (Sendai Framework), and human migration (Global Compact for Migration). As many ideas and actions on planned internal or international relocation of climate change–induced displacement are relatively new in the national and global policy domains, continued research and science-policy interface are essential in order to determine the feasibility, efficacy, and scalability of these options. 1. [↵][17]1. K. Rigaud et al ., “Groundswell: Preparing for Internal Climate Migration” (World Bank, 2018). 2. [↵][18]1. H. Singh, 2. J. Faleiro, 3. T. Anderson, 4. S. Vashist , “Costs of Climate Inaction Displacement and Distress Migration” (Actionaid, 2020). 3. [↵][19]1. J. Carmin, 2. D. Roberts, 3. I. Anguelovski , “Planning Climate Resilient Cities: Early Lessons from Early Adapters” (2011), pp. 5–8. 4. [↵][20]1. S. Weerasinghe et al ., “Planned Relocation, Disasters and Climate Change: Consolidating Good Practices and Preparing for the Future” (UNHCR, 2014). 5. [↵][21]1. B. Mallick, 2. K. G. Rogers, 3. Z. Sultana , Ambio 10.1007/s13280-021-01552-8 (2021). 6. [↵][22]1. S. S. Alam, 2. S. Huq, 3. F. Islam, 4. H. M. A. Hoque , “Building Climate-Resilient, Migrant-Friendly Cities and Towns” (International Centre for Climate Change and Development, 2018). 7. [↵][23]1. E. Wright, 2. D. Tänzler, 3. L. Rüttinger , “Migration, Environment and Climate Change: Responding via Climate Change Adaptation Policy” (German Environment Agency, 2020). 8. [↵][24]1. M. R. Khan , Toward a Binding Climate Change Adaptation Regime: A Proposed Framework (Routledge, 2014), chapter 6. 9. [↵][25]1. H. Dempster , “New Zealand's ‘Climate Refugee’ Visas: Lessons for the Rest of the World” (Centre for Global Development, Washington, DC, 2020). 10. [↵][26]1. D. M. S. Matias , Clim. Change 160, 143 (2020). [OpenUrl][27] 11. [↵][28]1. A. Kraler, 2. K. Caitlin, 3. M. Wagner , “Climate Change and Migration: Legal and Policy Challenges and Responses to Environmentally-Induced Migration” (European Union, 2020). 12. [↵][29]German Development Agency, “Sustainable Recruitment of Nurses (Triple Win)” (2019); [www.giz.de/en/worldwide/41533.html][30]. 13. [↵][31]1. E.-j. Quak , “The effects of economic integration of migrants on the economy of host countries” (Institute of Development Studies, London, 2016). 14. [↵][32]1. V. Grossmann , “How Immigration Affects Investment and Productivity in Host and Home Countries” (IZA, 2016); . 15. [↵][33]World Bank, “Record high remittances to low- and middle-income countries in 2017” (2018); [www.worldbank.org/en/news/press-release/2018/04/23/record-high-remittances-to-low-and-middle-income-countries-in-2017][34]. [1]: #ref-1 [2]: #ref-2 [3]: #ref-3 [4]: pending:yes [5]: #ref-4 [6]: #ref-5 [7]: #ref-6 [8]: #ref-7 [9]: #ref-8 [10]: #ref-9 [11]: #ref-10 [12]: #ref-11 [13]: #ref-12 [14]: #ref-13 [15]: #ref-14 [16]: #ref-15 [17]: #xref-ref-1-1 "View reference 1 in text" [18]: #xref-ref-2-1 "View reference 2 in text" [19]: #xref-ref-3-1 "View reference 3 in text" [20]: #xref-ref-4-1 "View reference 4 in text" [21]: #xref-ref-5-1 "View reference 5 in text" [22]: #xref-ref-6-1 "View reference 6 in text" [23]: #xref-ref-7-1 "View reference 7 in text" [24]: #xref-ref-8-1 "View reference 8 in text" [25]: #xref-ref-9-1 "View reference 9 in text" [26]: #xref-ref-10-1 "View reference 10 in text" [27]: {openurl}?query=rft.jtitle%253DClim.%2BChange%26rft.volume%253D160%26rft.spage%253D143%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [28]: #xref-ref-11-1 "View reference 11 in text" [29]: #xref-ref-12-1 "View reference 12 in text" [30]: http://www.giz.de/en/worldwide/41533.html [31]: #xref-ref-13-1 "View reference 13 in text" [32]: #xref-ref-14-1 "View reference 14 in text" [33]: #xref-ref-15-1 "View reference 15 in text" [34]: http://www.worldbank.org/en/news/press-release/2018/04/23/record-high-remittances-to-low-and-middle-income-countries-in-2017

Finance Minister Asad Umar recently spoke in the National Assembly about amendments to the Finance Bill and hinted at using data and AI to identify tax evaders. "Algorithms are now very advanced. Global tax collection agencies don't rely solely on traditional methods for collection, they use modern technology," he said. "We want to use anonymised data to run algorithms which will help us identify tax evaders." But what algorithms was Umar referring to and how do they work?

While at King Edward Medical University (KEMU), one of my responsibilities was giving a series of lectures on cardiac surgery to the final year MBBS class every year. Cardiac surgery is not an examination subject so attendance in my lectures was rather thin and the atmosphere a bit relaxed. In my first lecture to the new class, I often started off with asking students three questions not related to cardiac surgery. The first question was about Pakistan history. The question was, who was the second prime minister of Pakistan, and who was the second governor general of Pakistan.