{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/bingoguard-llm-content-moderation-tools-with","title":"BingoGuard: LLM Content Moderation Tools with Risk Levels","arxiv_id":"2503.06550","date":"2025-03-09","proceeding":null,"authors":["Fan Yin","Philippe Laban","Xiangyu Peng","Yilun Zhou","Yixin Mao","Vaibhav Vats","Linnea Ross","Divyansh Agarwal","Caiming Xiong","Chien-Sheng Wu"],"abstract":"Malicious content generated by large language models (LLMs) can pose varying degrees of harm. Although existing LLM-based moderators can detect harmful content, they struggle to assess risk levels and may miss lower-risk outputs. Accurate risk assessment allows platforms with different safety thresholds to tailor content filtering and rejection. In this paper, we introduce per-topic severity rubrics for 11 harmful topics and build BingoGuard, an LLM-based moderation system designed to predict both binary safety labels and severity levels. To address the lack of annotations on levels of severity, we propose a scalable generate-then-filter framework that first generates responses across different severity levels and then filters out low-quality responses. Using this framework, we create BingoGuardTrain, a training dataset with 54,897 examples covering a variety of topics, response severity, styles, and BingoGuardTest, a test set with 988 examples explicitly labeled based on our severity rubrics that enables fine-grained analysis on model behaviors on different severity levels. Our BingoGuard-8B, trained on BingoGuardTrain, achieves the state-of-the-art performance on several moderation benchmarks, including WildGuardTest and HarmBench, as well as BingoGuardTest, outperforming best public models, WildGuard, by 4.3\\%. Our analysis demonstrates that incorporating severity levels into training significantly enhances detection performance and enables the model to effectively gauge the severity of harmful responses.","url_abs":"https://arxiv.org/abs/2503.06550v1","url_pdf":"https://arxiv.org/pdf/2503.06550v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[],"methods":[{"method_slug":"set","method_name":"SET"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2503.06550","atlas_url":"https://app.syntology.ai/?focus=2503.06550","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2503.06550"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"deterministic:regex_extraction","url":"https://github.com/argilla-io/argilla","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":3},"by_repo_kind":{"found_in_text":{"samples":3,"ran":3,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"92ee2cd14509b1ec","entry":"full_qualified_class_name","repo":"argilla-io/argilla","repo_kind":"found_in_text","path":"argilla-server/src/argilla_server/logging.py","file_url":"https://github.com/argilla-io/argilla/blob/HEAD/argilla-server/src/argilla_server/logging.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"92ee2cd14509b1ec"}},{"code_sha256_prefix":"991c771d8d16fb58","entry":"get_logger_for_class","repo":"argilla-io/argilla","repo_kind":"found_in_text","path":"argilla-server/src/argilla_server/logging.py","file_url":"https://github.com/argilla-io/argilla/blob/HEAD/argilla-server/src/argilla_server/logging.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"991c771d8d16fb58"}},{"code_sha256_prefix":"506be8e0370a57fb","entry":"remove_suffix","repo":"argilla-io/argilla","repo_kind":"found_in_text","path":"argilla-server/src/argilla_server/helpers.py","file_url":"https://github.com/argilla-io/argilla/blob/HEAD/argilla-server/src/argilla_server/helpers.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"506be8e0370a57fb"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}