ToxiREX: A Dataset on Toxic REasoning in ConteXt

• URL: http://arxiv.org/abs/2606.27981v1
• Date: Fri, 26 Jun 2026 11:30:42 GMT
• Title: ToxiREX: A Dataset on Toxic REasoning in ConteXt
• Abstract summary: We introduce ToxiREX: Toxic REasoning in ConteXt.<n>The dataset consists of threads of Reddit comments and structured characterizations of what the comments imply.<n>ToxiREX is the first dataset to simultaneously incorporate multiple languages, conversational context, and implicit toxicity.
• Score: 7.155412580325742
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We introduce a new, contextual, multilingual dataset called ToxiREX: Toxic REasoning in ConteXt. The dataset consists of threads of Reddit comments and structured characterizations of what the comments imply, following a systematic toxic reasoning schema developed in a previous paper. Using the schema allows us to capture and explain implicit and context-dependent toxicity, while supporting mappings to existing toxicity taxonomies. The dataset includes comments in six languages (English, Arabic, Turkish, Spanish, German, and Dutch), collected from posts connected to specific major events (e.g. the 2023 Turkey earthquakes; the Russian invasion of Ukraine). We describe the context-preserving preprocessing of the threads. We create a training set of 125 thousand comments which is annotated by a commercially available LLM, and a test set of just under three thousand comments that is annotated by native speakers. We show that apparent disagreements in the test set annotations often reflect defensible alternative interpretations rather than noise. Finally, we provide baseline results by prompting and fine-tuning language models. To produce these results, we develop evaluation strategies for our hierarchical, schema-based predictions. While models perform better than random, there remains a lot of room for improvement, showing the task to be challenging. ToxiREX is the first dataset to simultaneously incorporate multiple languages, conversational context, and implicit toxicity, while using the toxic reasoning schema for rich, structured annotations. Dataset available at: https://github.com/cltl/toxirex

Related papers

• SynthDetoxM: Modern LLMs are Few-Shot Parallel Detoxification Data Annotators [61.82799141938912]
  Existing approaches to multilingual text detoxification are hampered by the scarcity of parallel multilingual datasets.<n>We introduce SynthDetoxM, a manually collected and synthetically generated multilingual parallel text detoxification dataset.
  arXiv  Detail & Related papers  (2025-02-10T12:30:25Z)
• Triples-to-isiXhosa (T2X): Addressing the Challenges of Low-Resource Agglutinative Data-to-Text Generation [9.80836683456026]
  We tackle data-to-text for isiXhosa, which is low-resource and agglutinative. We introduce Triples-to-isiXhosa (T2X), a new dataset based on a subset of WebNLG. We develop an evaluation framework for T2X that measures how accurately generated text describes the data.
  arXiv  Detail & Related papers  (2024-03-12T11:53:27Z)
• Text Detoxification as Style Transfer in English and Hindi [1.183205689022649]
  This paper focuses on text detoxification, i.e., automatically converting toxic text into non-toxic text. We present three approaches: knowledge transfer from a similar task, multi-task learning approach, and delete and reconstruct approach. Our results demonstrate that our approach effectively balances text detoxication while preserving the actual content and maintaining fluency.
  arXiv  Detail & Related papers  (2024-02-12T16:30:41Z)
• Constructing Colloquial Dataset for Persian Sentiment Analysis of Social Microblogs [0.0]
  This paper first constructs a user opinion dataset called ITRC-Opinion in a collaborative environment and insource way. Our dataset contains 60,000 informal and colloquial Persian texts from social microblogs such as Twitter and Instagram. Second, this study proposes a new architecture based on the convolutional neural network (CNN) model for more effective sentiment analysis of colloquial text in social microblog posts.
  arXiv  Detail & Related papers  (2023-06-22T05:51:22Z)
• XRICL: Cross-lingual Retrieval-Augmented In-Context Learning for Cross-lingual Text-to-SQL Semantic Parsing [70.40401197026925]
  In-context learning using large language models has recently shown surprising results for semantic parsing tasks. This work introduces the XRICL framework, which learns to retrieve relevant English exemplars for a given query. We also include global translation exemplars for a target language to facilitate the translation process for large language models.
  arXiv  Detail & Related papers  (2022-10-25T01:33:49Z)
• DiS-ReX: A Multilingual Dataset for Distantly Supervised Relation Extraction [15.649929244635269]
  We propose a new dataset, DiS-ReX, which alleviates these issues. Our dataset has more than 1.5 million sentences, spanning across 4 languages with 36 relation classes + 1 no relation (NA) class. We also modify the widely used bag attention models by encoding sentences using mBERT and provide the first benchmark results on multilingual DS-RE.
  arXiv  Detail & Related papers  (2021-04-17T22:44:38Z)
• XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization [98.61159823343036]
  We present the Word-in-Context dataset (WiC) for assessing the ability to correctly model distinct meanings of a word. We put forward a large multilingual benchmark, XL-WiC, featuring gold standards in 12 new languages. Experimental results show that even when no tagged instances are available for a target language, models trained solely on the English data can attain competitive performance.
  arXiv  Detail & Related papers  (2020-10-13T15:32:00Z)
• Abstractive Summarization of Spoken and Written Instructions with BERT [66.14755043607776]
  We present the first application of the BERTSum model to conversational language. We generate abstractive summaries of narrated instructional videos across a wide variety of topics. We envision this integrated as a feature in intelligent virtual assistants, enabling them to summarize both written and spoken instructional content upon request.
  arXiv  Detail & Related papers  (2020-08-21T20:59:34Z)
• XCOPA: A Multilingual Dataset for Causal Commonsense Reasoning [68.57658225995966]
  Cross-lingual Choice of Plausible Alternatives (XCOPA) is a typologically diverse multilingual dataset for causal commonsense reasoning in 11 languages. We evaluate a range of state-of-the-art models on this novel dataset, revealing that the performance of current methods falls short compared to translation-based transfer.
  arXiv  Detail & Related papers  (2020-05-01T12:22:33Z)
• ToTTo: A Controlled Table-To-Text Generation Dataset [61.83159452483026]
  ToTTo is an open-domain English table-to-text dataset with over 120,000 training examples. We introduce a dataset construction process where annotators directly revise existing candidate sentences from Wikipedia. While usually fluent, existing methods often hallucinate phrases that are not supported by the table.
  arXiv  Detail & Related papers  (2020-04-29T17:53:45Z)

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.