Search

The surge of digital documents in various formats, including less standardized documents such as business reports and environmental assessments, underscores the growing importance of Document Understanding. While Large Language Models (LLMs) have showcased prowess across diverse natural language processing tasks, their direct application to Document Understanding remains a challenge. Previous research has demonstrated the utility of LLMs in this domain, yet their significant computational demands make them challenging to deploy effectively. Additionally, proprietary Blackbox LLMs often outperform their open-source counterparts, posing a barrier to widespread accessibility. In this paper, we delve into the realm of document understanding, leveraging distillation methods to harness the power of large LLMs while accommodating computational limitations. Specifically, we present a novel approach wherein we distill document understanding knowledge from the proprietary LLM ChatGPT into FLAN-T5. Our methodology integrates labeling and curriculum-learning mechanisms to facilitate efficient knowledge transfer. This work contributes to the advancement of document understanding methodologies by offering a scalable solution that bridges the gap between resource-intensive LLMs and practical applications. Our findings underscore the potential of distillation techniques in facilitating the deployment of sophisticated language models in real-world scenarios, thereby fostering advancements in natural language processing and document comprehension domains., Comment: Presented at AI@WORK-Workshop / Informatik-Festival (GI-Jahrestagung) (Wiesbaden, Germany, 2024)

This paper investigates how generative AI can potentially undermine the integrity of collective knowledge and the processes we rely on to acquire, assess, and trust information, posing a significant threat to our knowledge ecosystem and democratic discourse. Grounded in social and political philosophy, we introduce the concept of \emph{generative algorithmic epistemic injustice}. We identify four key dimensions of this phenomenon: amplified and manipulative testimonial injustice, along with hermeneutical ignorance and access injustice. We illustrate each dimension with real-world examples that reveal how generative AI can produce or amplify misinformation, perpetuate representational harm, and create epistemic inequities, particularly in multilingual contexts. By highlighting these injustices, we aim to inform the development of epistemically just generative AI systems, proposing strategies for resistance, system design principles, and two approaches that leverage generative AI to foster a more equitable information ecosystem, thereby safeguarding democratic values and the integrity of knowledge production.

BM25, a widely-used lexical search algorithm, remains crucial in information retrieval despite the rise of pre-trained and large language models (PLMs/LLMs). However, it neglects query-document similarity and lacks semantic understanding, limiting its performance. We revisit BM25 and introduce BMX, a novel extension of BM25 incorporating entropy-weighted similarity and semantic enhancement techniques. Extensive experiments demonstrate that BMX consistently outperforms traditional BM25 and surpasses PLM/LLM-based dense retrieval in long-context and real-world retrieval benchmarks. This study bridges the gap between classical lexical search and modern semantic approaches, offering a promising direction for future information retrieval research. The reference implementation of BMX can be found in Baguetter, which was created in the context of this work. The code can be found here: https://github.com/mixedbread-ai/baguetter., Comment: correct the affiliation order

Atmospheric states derived from reanalysis comprise a substantial portion of weather and climate simulation outputs. Many stakeholders -- such as researchers, policy makers, and insurers -- use this data to better understand the earth system and guide policy decisions. Atmospheric states have also received increased interest as machine learning approaches to weather prediction have shown promising results. A key issue for all audiences is that dense time series of these high-dimensional states comprise an enormous amount of data, precluding all but the most well resourced groups from accessing and using historical data and future projections. To address this problem, we propose a method for compressing atmospheric states using methods from the neural network literature, adapting spherical data to processing by conventional neural architectures through the use of the area-preserving HEALPix projection. We investigate two model classes for building neural compressors: the hyperprior model from the neural image compression literature and recent vector-quantised models. We show that both families of models satisfy the desiderata of small average error, a small number of high-error reconstructed pixels, faithful reproduction of extreme events such as hurricanes and heatwaves, preservation of the spectral power distribution across spatial scales. We demonstrate compression ratios in excess of 1000x, with compression and decompression at a rate of approximately one second per global atmospheric state., Comment: 44 pages, 25 figures

Lagrangian modelling can be used to derive mathematical models for complex power electronic converters. This approach uses scalar quantities (kinetic and potential energy) to derive models, which is simpler than using (vector-based) force balance equations. It employs generalized coordinates, making it easier to deal with complex systems with constraints. This systematic approach results in equations that can be expressed in state-space form, which allows for the simplification of the simulation and design process and the use of many standard software packages for system analysis and simulation. In this work, contributions are made regarding the procedure to be followed for the Lagrangian modelling of power converters and the incorporation of constraints within the Lagrangian framework. Furthermore, for the first time, Lagrangian modelling is extended to non-ideal, high-fidelity descriptions of standard power electronic circuits.

We interviewed twenty professional comedians who perform live shows in front of audiences and who use artificial intelligence in their artistic process as part of 3-hour workshops on ``AI x Comedy'' conducted at the Edinburgh Festival Fringe in August 2023 and online. The workshop consisted of a comedy writing session with large language models (LLMs), a human-computer interaction questionnaire to assess the Creativity Support Index of AI as a writing tool, and a focus group interrogating the comedians' motivations for and processes of using AI, as well as their ethical concerns about bias, censorship and copyright. Participants noted that existing moderation strategies used in safety filtering and instruction-tuned LLMs reinforced hegemonic viewpoints by erasing minority groups and their perspectives, and qualified this as a form of censorship. At the same time, most participants felt the LLMs did not succeed as a creativity support tool, by producing bland and biased comedy tropes, akin to ``cruise ship comedy material from the 1950s, but a bit less racist''. Our work extends scholarship about the subtle difference between, one the one hand, harmful speech, and on the other hand, ``offensive'' language as a practice of resistance, satire and ``punching up''. We also interrogate the global value alignment behind such language models, and discuss the importance of community-based value alignment and data ownership to build AI tools that better suit artists' needs., Comment: 15 pages, 1 figure, published at ACM FAccT 2024

Excitons in monolayer transition metal dichalcogenides (TMDCs) offer intriguing new possibilities for optoelectronics with no analogues in bulk semiconductors. Yet, intrinsic defects in TMDCs limit the radiative exciton recombination pathways. As a result, the photoluminescence (PL) quantum yield (QY) is limited. Methods like superacid treatment, electrical doping, and plasmonic engineering can inhibit nonradiative decay channels and enhance PL. Here, we show a more straightforward approach that allows PL enhancement. An engineered vertical van der Waals (vdW) metal-monolayer semiconductor junction (MSJ) results in PL enhancement of more than an order of magnitude at technologically relevant excitation powers. Such MSJ can be constructed by vertically stacking metals with suitable work function either above or below a monolayer semiconducting TMDC. Our experiments reveal that the underlying PL enhancement mechanism is to be the suppressed exciton quenching due to the absence of metal-induced gap states and weak Fermi level pinning, thanks to the vdW gapped interface between the metal and the TMDC. Our time-resolved PL measurements further indicate that reduced exciton-exciton annihilation, even at high generation rates, contributes to the observed PL enhancement. The PL intensity is further increased by the proximity of surface plasmons in the metal with the TMDC layer. Our findings shed light on the interaction at vdW metal-semiconductor interfaces and offer a path to improving the optoelectronic performance of semiconducting TMDCs.

A major challenge facing the world is the provision of equitable and universal access to quality education. Recent advances in generative AI (gen AI) have created excitement about the potential of new technologies to offer a personal tutor for every learner and a teaching assistant for every teacher. The full extent of this dream, however, has not yet materialised. We argue that this is primarily due to the difficulties with verbalising pedagogical intuitions into gen AI prompts and the lack of good evaluation practices, reinforced by the challenges in defining excellent pedagogy. Here we present our work collaborating with learners and educators to translate high level principles from learning science into a pragmatic set of seven diverse educational benchmarks, spanning quantitative, qualitative, automatic and human evaluations; and to develop a new set of fine-tuning datasets to improve the pedagogical capabilities of Gemini, introducing LearnLM-Tutor. Our evaluations show that LearnLM-Tutor is consistently preferred over a prompt tuned Gemini by educators and learners on a number of pedagogical dimensions. We hope that this work can serve as a first step towards developing a comprehensive educational evaluation framework, and that this can enable rapid progress within the AI and EdTech communities towards maximising the positive impact of gen AI in education.

Human participants play a central role in the development of modern artificial intelligence (AI) technology, in psychological science, and in user research. Recent advances in generative AI have attracted growing interest to the possibility of replacing human participants in these domains with AI surrogates. We survey several such "substitution proposals" to better understand the arguments for and against substituting human participants with modern generative AI. Our scoping review indicates that the recent wave of these proposals is motivated by goals such as reducing the costs of research and development work and increasing the diversity of collected data. However, these proposals ignore and ultimately conflict with foundational values of work with human participants: representation, inclusion, and understanding. This paper critically examines the principles and goals underlying human participation to help chart out paths for future work that truly centers and empowers participants., Comment: Proceedings of the CHI Conference on Human Factors in Computing Systems (CHI 2024)

Weather forecasts are fundamentally uncertain, so predicting the range of probable weather scenarios is crucial for important decisions, from warning the public about hazardous weather, to planning renewable energy use. Here, we introduce GenCast, a probabilistic weather model with greater skill and speed than the top operational medium-range weather forecast in the world, the European Centre for Medium-Range Forecasts (ECMWF)'s ensemble forecast, ENS. Unlike traditional approaches, which are based on numerical weather prediction (NWP), GenCast is a machine learning weather prediction (MLWP) method, trained on decades of reanalysis data. GenCast generates an ensemble of stochastic 15-day global forecasts, at 12-hour steps and 0.25 degree latitude-longitude resolution, for over 80 surface and atmospheric variables, in 8 minutes. It has greater skill than ENS on 97.4% of 1320 targets we evaluated, and better predicts extreme weather, tropical cyclones, and wind power production. This work helps open the next chapter in operational weather forecasting, where critical weather-dependent decisions are made with greater accuracy and efficiency., Comment: Main text 11 pages, Appendices 76 pages

In recent years, the amount of data available on the internet and the number of users who utilize the Internet have increased at an unparalleled pace. The exponential development in the quantity of digital information accessible and the number of Internet users has created the possibility for information overload, impeding fast access to items of interest on the Internet. Information retrieval systems like as Google, DevilFinder, and Altavista have partly overcome this challenge, but prioritizing and customization of information (where a system maps accessible material to a user's interests and preferences) were lacking. This has resulted in a higher-than-ever need for recommender systems. Recommender systems are information filtering systems that address the issue of information overload by filtering important information fragments from a huge volume of dynamically produced data based on the user's interests, favorite things, preferences and ratings on the desired item. Recommender systems can figure out if a person would like an item or not based on their profile., Comment: want to do experiment on proposed methods