mlsanityData to 7 Oct 2026

Retrieval-Augmented Detection of Potentially Abusive Clauses in Chilean Terms of Service

arXiv 2605.26019 ↗First posted 25 May 2026cs.LG cs.AI cs.CLcited by 0

On the map: Retrieval-augmented generation → Legal and financial RAG

Abstract (arXiv)

Online Terms of Service often function as contracts of adhesion, creating asymmetries that may expose consumers to potentially abusive clauses. In Chile, assessing such clauses is legally challenging because some provisions clearly violate mandatory consumer law, whereas others depend on broader standards such as good faith and contractual imbalance. We present a retrieval-augmented generation framework for the automated detection and classification of potentially abusive clauses in Chilean Terms of Service. Designed for local execution, it combines efficient clause detection, hybrid dense--sparse retrieval, reranking, and prompt augmentation to support medium-sized open-weight language models. We also introduce the Chilean Abusive Terms of Service Extended corpus, comprising 100 contracts and 10,029 annotated clauses in 24 legally grounded categories spanning illegal, dark, and gray clauses. Experiments comparing commercial and open-weight language models, fine-tuned encoders, and traditional baselines show that retrieval-augmented prompting substantially improves performance and enables local models to approach larger cloud-based systems at lower computational and token cost. The study also contributes a refined legal annotation scheme and a practical design for AI-assisted consumer contract review.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Legal and financial RAG, one of 15 directions in Retrieval-augmented generation. Builds on 37 of its 61 references that are in the map. First posted on arXiv 25 May 2026.

Builds on

37 of 61 references.

References · 37

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2024The Llama 3 Herd of Modelscited by 14,192
  4. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  5. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  6. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,049
  7. 2025Qwen3 Technical Reportcited by 8,195
  8. 2019Unsupervised Cross-lingual Representation Learning at Scalecited by 3,907
  9. 2020Longformer: The Long-Document Transformercited by 3,362
  10. 2023Retrieval-Augmented Generation for Large Language Models: A Surveycited by 2,185
  11. 2022GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformerscited by 1,838
  12. 2016MS MARCO: A Human Generated MAchine Reading COmprehension Datasetcited by 1,591
25 more
  1. 2024From Local to Global: A Graph RAG Approach to Query-Focused Summarizationcited by 1,411
  2. 2025gpt-oss-120b & gpt-oss-20b Model Cardcited by 1,258
  3. 2023AWQ: Activation-aware Weight Quantization for LLM Compression and Accelerationcited by 1,234
  4. 2022Text Embeddings by Weakly-Supervised Contrastive Pre-trainingcited by 1,216
  5. 2023Large Language Models Can Be Easily Distracted by Irrelevant Contextcited by 820
  6. 2025DeepSeek-V3.2: Pushing the Frontier of Open Large Language Modelscited by 755
  7. 2023In-Context Retrieval-Augmented Language Modelscited by 694
  8. 2019How multilingual is Multilingual BERT?cited by 642
  9. 20218-bit Optimizers via Block-wise Quantizationcited by 395
  10. 2024The Faiss librarycited by 370
  11. 2024Multilingual E5 Text Embeddings: A Technical Reportcited by 357
  12. 2022The case for 4-bit precision: k-bit Inference Scaling Lawscited by 247
  13. 2024ClashEval: Quantifying the tug-of-war between an LLM's internal prior and external evidencecited by 109
  14. 2023Spanish Pre-trained BERT Model and Evaluation Datacited by 81
  15. 2024Evaluating Quantized Large Language Modelscited by 75
  16. 2018CLAUDETTE: an Automated Detector of Potentially Unfair Clauses in Online Terms of Servicecited by 72
  17. 2025jina-reranker-v3: Last but Not Late Interaction for Listwise Document Rerankingcited by 23
  18. 2021Few-Shot Learning with Class Imbalancecited by 18
  19. 2023Bringing order into the realm of Transformer-based language models for artificial intelligence and lawcited by 13
  20. 2021Spanish Legalese Language Model and Corporacited by 11
  21. 2025Pathway to Relevance: How Cross-Encoders Implement a Semantic Variant of BM25cited by 6
  22. 2026The Reasoning Bottleneck in Graph-RAG: Structured Prompting and Context Compression for Multi-Hop QAcited by 4
  23. 2025OASIS: Outlier-Aware LUT-Based GEMM with Dual-Side Quantization for LLM Inference Accelerationcited by 2
  24. 2025Predicting potentially abusive clauses in Chilean terms of services with natural language processingcited by 1
  25. 2025ParetoRAG: Leveraging Sentence-Context Attention for Robust and Efficient Retrieval-Augmented Generationcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Enhancing Legal LLMs through Metadata-Enriched RAG Pipelines and Direct Preference Optimization
  2. 2025Better Call CLAUSE: A Discrepancy Benchmark for Auditing LLMs Legal Reasoning Capabilities
  3. 2026Building Legal Reward Models for Grounding and Abstention
  4. 2024LegalBench-RAG: A Benchmark for Retrieval-Augmented Generation in the Legal Domain
  5. 2025Metadata Extraction Leveraging Large Language Models
  6. 2025Towards Reliable Retrieval in RAG Systems for Large Legal Datasets
  7. 2026Fine-grained Claim-level RAG Benchmark for Law
  8. 2025Scaling Legal AI: Benchmarking Mamba and Transformers for Statutory Classification and Case Law Retrieval