Automated Analysis of Global AI Safety Initiatives: A Taxonomy-Driven LLM Approach

• URL: http://arxiv.org/abs/2604.03533v1
• Date: Sat, 04 Apr 2026 01:02:15 GMT
• Title: Automated Analysis of Global AI Safety Initiatives: A Taxonomy-Driven LLM Approach
• Abstract summary: We present an automated crosswalk framework that compares an AI safety policy document pair under a shared taxonomy of activities.<n>Using the activity categories defined in Activity Map on AI Safety as fixed aspects, the system extracts and maps relevant activities, then produces for each aspect a short summary.<n>We assess the stability and validity of LLM-based crosswalk analysis across public policy documents.
• Score: 0.012202203114453207
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We present an automated crosswalk framework that compares an AI safety policy document pair under a shared taxonomy of activities. Using the activity categories defined in Activity Map on AI Safety as fixed aspects, the system extracts and maps relevant activities, then produces for each aspect a short summary for each document, a brief comparison, and a similarity score. We assess the stability and validity of LLM-based crosswalk analysis across public policy documents. Using five large language models, we perform crosswalks on ten publicly available documents and visualize mean similarity scores with a heatmap. The results show that model choice substantially affects the crosswalk outcomes, and that some document pairs yield high disagreements across models. A human evaluation by three experts on two document pairs shows high inter-annotator agreement, while model scores still differ from human judgments. These findings support comparative inspection of policy documents.

Related papers

• The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality [70.45240108873001]
  The FACTS Leaderboard is an online leaderboard suite that comprehensively evaluates the ability of language models to generate factually accurate text.<n>The suite provides a holistic measure of factuality by aggregating the performance of models on four distinct sub-leaderboards.
  arXiv  Detail & Related papers  (2025-12-11T16:35:14Z)
• ABCD-LINK: Annotation Bootstrapping for Cross-Document Fine-Grained Links [57.514511353084565]
  We introduce a new domain-agnostic framework for selecting a best-performing approach and annotating cross-document links.<n>We apply our framework in two distinct domains -- peer review and news.<n>The resulting novel datasets lay foundation for numerous cross-document tasks like media framing and peer review.
  arXiv  Detail & Related papers  (2025-09-01T11:32:24Z)
• ProxAnn: Use-Oriented Evaluations of Topic Models and Document Clustering [52.19512723549318]
  We design a scalable human evaluation protocol that reflects practitioners' real-world usage of models.<n>We use this protocol to collect extensive crowdworker annotations of outputs from a diverse set of topic models.<n>We then use these annotations to validate automated proxies, finding that the best LLM proxies are statistically indistinguishable from a human annotator.
  arXiv  Detail & Related papers  (2025-07-01T15:00:55Z)
• DocXplain: A Novel Model-Agnostic Explainability Method for Document Image Classification [5.247930659596986]
  This paper introduces DocXplain, a novel model-agnostic explainability method specifically designed for generating high interpretability feature attribution maps. We extensively evaluate our proposed approach in the context of document image classification, utilizing 4 different evaluation metrics. To the best of the authors' knowledge, this work presents the first model-agnostic attribution-based explainability method specifically tailored for document images.
  arXiv  Detail & Related papers  (2024-07-04T10:59:15Z)
• Evaluating the Fairness of Discriminative Foundation Models in Computer Vision [51.176061115977774]
  We propose a novel taxonomy for bias evaluation of discriminative foundation models, such as Contrastive Language-Pretraining (CLIP) We then systematically evaluate existing methods for mitigating bias in these models with respect to our taxonomy. Specifically, we evaluate OpenAI's CLIP and OpenCLIP models for key applications, such as zero-shot classification, image retrieval and image captioning.
  arXiv  Detail & Related papers  (2023-10-18T10:32:39Z)
• Fairness meets Cross-Domain Learning: a new perspective on Models and Metrics [80.07271410743806]
  We study the relationship between cross-domain learning (CD) and model fairness. We introduce a benchmark on face and medical images spanning several demographic groups as well as classification and localization tasks. Our study covers 14 CD approaches alongside three state-of-the-art fairness algorithms and shows how the former can outperform the latter.
  arXiv  Detail & Related papers  (2023-03-25T09:34:05Z)
• Document-Level Relation Extraction with Sentences Importance Estimation and Focusing [52.069206266557266]
  Document-level relation extraction (DocRE) aims to determine the relation between two entities from a document of multiple sentences. We propose a Sentence Estimation and Focusing (SIEF) framework for DocRE, where we design a sentence importance score and a sentence focusing loss. Experimental results on two domains show that our SIEF not only improves overall performance, but also makes DocRE models more robust.
  arXiv  Detail & Related papers  (2022-04-27T03:20:07Z)
• Automating Document Classification with Distant Supervision to Increase the Efficiency of Systematic Reviews [18.33687903724145]
  Well-done systematic reviews are expensive, time-demanding, and labor-intensive. We propose an automatic document classification approach to significantly reduce the effort in reviewing documents.
  arXiv  Detail & Related papers  (2020-12-09T22:45:40Z)
• Aspect-based Document Similarity for Research Papers [4.661692753666685]
  We extend similarity with aspect information by performing a pairwise document classification task. We evaluate our aspect-based document similarity for research papers. Our results show SciBERT as the best performing system.
  arXiv  Detail & Related papers  (2020-10-13T13:51:21Z)
• Hierarchical Interaction Networks with Rethinking Mechanism for Document-level Sentiment Analysis [37.20068256769269]
  Document-level Sentiment Analysis (DSA) is more challenging due to vague semantic links and complicate sentiment information. We study how to effectively generate a discriminative representation with explicit subject patterns and sentiment contexts for DSA. We design a Sentiment-based Rethinking mechanism (SR) by refining the HIN with sentiment label information to learn a more sentiment-aware document representation.
  arXiv  Detail & Related papers  (2020-07-16T16:27:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.