An official website of the United States government
Climate change demands urgent action, yet understanding the environmental impact (EI) of everyday objects and activities remains challenging for the general public. While Life Cycle Assessment (LCA) offers a comprehensive framework for EI analysis, its traditional implementation requires extensive domain expertise, structured input data, and significant time investment, creating barriers for non-experts seeking real-time sustainability insights. Here we present the first autonomous sustainability assessment tool that bridges this gap by transforming unstructured natural language descriptions into in-context, interactive EI visualizations. Our approach combines language modeling and AI agents, and achieves >97% accuracy in transforming natural language into a data abstraction designed for simplified LCA modeling. The system employs a non-parametric datastore to integrate proprietary LCA databases while maintaining data source attribution and allowing personalized source management. We demonstrate through case studies that our system achieves results within 11% of traditional full LCA, while accelerating from hours of expert time to real-time. We conducted a formative elicitation study (N=6) to inform the design objectives of such EI communication augmentation tools. We implemented and deployed the tool as a Chromium browser extension and further evaluated it through a user study (N=12). This work represents a significant step toward democratizing access to environmental impact information for the general public with zero LCA expertise.
more »
« less
CA 30x30 Planning & Assessment Prototype
Proof of concept for a decision support tool developed in partnership with California Biodiversity Network participants through a co-design process. The tool can answer complex, real world natural language queries asked by conservation partner organizations, responding with reproducible, verifiable data summaries, charts, maps and text through careful integration of open weights language models and cloud optimized data.
more »
« less
- Award ID(s):
- 2153040
- PAR ID:
- 10591775
- Publisher / Repository:
- Zenodo
- Date Published:
- Format(s):
- Medium: X
- Right(s):
- Creative Commons Attribution 4.0 International
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)We report on the development of a new instrument for measuring teachers' knowledge of language as an epistemic tool in science classes. Language is essential for science learning, as all learning requires the use of language to constitute one's own ideas and to engage with others' ideas. Teachers with knowledge of language as an epistemic tool can recognize the ways that language allows students to generate and validate knowledge for themselves, rather than to replicate canonical knowledge transmitted by other sources.We used a construct‐driven development approach with iterations of domain analysis, item revision, teacher feedback, expert review, and item piloting to address the content, substance, and structure aspects of validity. Data from 158 preservice and in‐service teachers on 27 preliminary items were collected. Findings from Rasch measurement modeling indicate a single dimension fits the items well and can distinguish teachers of higher and lower knowledge. We revised and selected 15 items for an updated instrument. This contributes to ongoing measurement projects and provides a potential instrument for future, broader use by the field to gauge teachers' knowledge of language as an epistemic tool.more » « less
-
Data visualization provides a powerful way for analysts to explore and make data-driven discoveries. However, current visual analytic tools provide only limited support for hypothesis-driven inquiry, as their built-in interactions and workflows are primarily intended for exploratory analysis. Visualization tools notably lack capabilities that would allow users to visually and incrementally test the fit of their conceptual models and provisional hypotheses against the data. This imbalance could bias users to overly rely on exploratory analysis as the principal mode of inquiry, which can be detrimental to discovery. In this paper, we introduce Visual (dis) Confirmation, a tool for conducting confirmatory, hypothesis-driven analyses with visualizations. Users interact by framing hypotheses and data expectations in natural language. The system then selects conceptually relevant data features and automatically generates visualizations to validate the underlying expectations. Distinctively, the resulting visualizations also highlight places where one's mental model disagrees with the data, so as to stimulate reflection. The proposed tool represents a new class of interactive data systems capable of supporting confirmatory visual analysis, and responding more intelligently by spotlighting gaps between one's knowledge and the data. We describe the algorithmic techniques behind this workflow. We also demonstrate the utility of the tool through a case study.more » « less
-
The data and compute requirements of current language modeling technology pose challenges for the processing and analysis of low-resource languages. Declarative linguistic knowledge has the potential to partially bridge this data scarcity gap by providing models with useful inductive bias in the form of language-specific rules. In this paper, we propose a retrieval augmented generation (RAG) framework backed by a large language model (LLM) to correct the output of a smaller model for the linguistic task of morphological glossing. We leverage linguistic information to make up for the lack of data and trainable parameters, while allowing for inputs from written descriptive grammars interpreted and distilled through an LLM. The results demonstrate that significant leaps in performance and efficiency are possible with the right combination of: a) linguistic inputs in the form of grammars, b) the interpretive power of LLMs, and c) the trainability of smaller token classification networks. We show that a compact, RAG-supported model is highly effective in data-scarce settings, achieving a new state-of-the-art for this task and our target languages. Our work also offers documentary linguists a more reliable and more usable tool for morphological glossing by providing well-reasoned explanations and confidence scores for each output.more » « less
-
null (Ed.)Customizing software should be as easy as using it. Unfortunately, most customization methods require users to abruptly shift from using a graphical interface to writing scripts in a programming language. We introduce data-driven customization, a new way for end users to extend software by direct manipulation without doing traditional programming. We augment existing user interfaces with a table view showing the structured data inside the application. When users edit the table, their changes are reflected in the original UI. This simple model accommodates a spreadsheet formula language and custom data-editing widgets, providing enough power to implement a variety of useful extensions. We illustrate the approach with Wildcard, a browser extension that implements data-driven customization on the web using web scraping. Through concrete examples, we show that this paradigm can support useful extensions to many real websites, and we share reflections from our experiences using the tool. Finally, we share our broader vision for data-driven customization: a future where end users have more access to the data inside their applications, and can more flexibly repurpose that data as part of everyday software usage.more » « less
