{"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/self-contradictory-hallucinations-of-large","title":"Self-contradictory Hallucinations of Large Language Models: Evaluation, Detection and Mitigation","arxiv_id":"2305.15852","date":"2023-05-25","proceeding":null,"authors":["Niels Mündler","Jingxuan He","Slobodan Jenko","Martin Vechev"],"abstract":"Large language models (large LMs) are susceptible to producing text that contains hallucinated content. An important instance of this problem is self-contradiction, where the LM generates two contradictory sentences within the same context. In this work, we present a comprehensive investigation into self-contradiction for various instruction-tuned LMs, covering evaluation, detection, and mitigation. Our primary evaluation task is open-domain text generation, but we also demonstrate the applicability of our approach to shorter question answering. Our analysis reveals the prevalence of self-contradictions, e.g., in 17.7% of all sentences produced by ChatGPT. We then propose a novel prompting-based framework designed to effectively detect and mitigate self-contradictions. Our detector achieves high accuracy, e.g., around 80% F1 score when prompting ChatGPT. The mitigation algorithm iteratively refines the generated text to remove contradictory information while preserving text fluency and informativeness. Importantly, our entire framework is applicable to black-box LMs and does not require retrieval of external knowledge. Rather, our method complements retrieval-based methods, as a large portion of self-contradictions (e.g., 35.2% for ChatGPT) cannot be verified using online text. Our approach is practically effective and has been released as a push-button tool to benefit the public at https://chatprotect.ai/.","url_abs":"https://arxiv.org/abs/2305.15852v3","url_pdf":"https://arxiv.org/pdf/2305.15852v3.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":[{"paper_slug":"self-contradictory-hallucinations-of-large","repo_url":"https://github.com/eth-sri/chatprotect","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":null,"task_name":"Hallucination Pair-wise Detection (1-ref)"},{"task_slug":"informativeness","task_name":"Informativeness"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"text-generation","task_name":"Text Generation"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt-4","method_name":"GPT-4"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2305.15852","atlas_url":"https://app.syntology.ai/?focus=2305.15852","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2305.15852"}},"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":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/eth-sri/chatprotect","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":9},"by_repo_kind":{"official":{"samples":9,"ran":0,"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":"07b62c885f7ea17e","entry":"llama_v2_prompt","repo":"eth-sri/chatprotect","repo_kind":"official","path":"chatprotect/bots/together_ai_api.py","file_url":"https://github.com/eth-sri/chatprotect/blob/HEAD/chatprotect/bots/together_ai_api.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"07b62c885f7ea17e"}},{"code_sha256_prefix":"ec7bbc872d252c0f","entry":"question_generation_beamsearch","repo":"eth-sri/chatprotect","repo_kind":"official","path":"chatprotect/selfcheckgpt/modeling_mqag.py","file_url":"https://github.com/eth-sri/chatprotect/blob/HEAD/chatprotect/selfcheckgpt/modeling_mqag.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ec7bbc872d252c0f"}},{"code_sha256_prefix":"2ff8cbbaf8e530db","entry":"question_generation_sampling","repo":"eth-sri/chatprotect","repo_kind":"official","path":"chatprotect/selfcheckgpt/modeling_mqag.py","file_url":"https://github.com/eth-sri/chatprotect/blob/HEAD/chatprotect/selfcheckgpt/modeling_mqag.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"2ff8cbbaf8e530db"}},{"code_sha256_prefix":"d0dc64dd7b3c3112","entry":"question_generation_sentence_level","repo":"eth-sri/chatprotect","repo_kind":"official","path":"chatprotect/selfcheckgpt/modeling_mqag.py","file_url":"https://github.com/eth-sri/chatprotect/blob/HEAD/chatprotect/selfcheckgpt/modeling_mqag.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d0dc64dd7b3c3112"}},{"code_sha256_prefix":"10de4d9e39ba3715","entry":"read_entities","repo":"eth-sri/chatprotect","repo_kind":"official","path":"chatprotect/util.py","file_url":"https://github.com/eth-sri/chatprotect/blob/HEAD/chatprotect/util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"10de4d9e39ba3715"}},{"code_sha256_prefix":"c260837e886b1746","entry":"read_sent_file","repo":"eth-sri/chatprotect","repo_kind":"official","path":"chatprotect/util.py","file_url":"https://github.com/eth-sri/chatprotect/blob/HEAD/chatprotect/util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c260837e886b1746"}},{"code_sha256_prefix":"20bd07858a5fc478","entry":"rephrase_sent_remove_conflict","repo":"eth-sri/chatprotect","repo_kind":"official","path":"chatprotect/rephrase.py","file_url":"https://github.com/eth-sri/chatprotect/blob/HEAD/chatprotect/rephrase.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"20bd07858a5fc478"}},{"code_sha256_prefix":"00998fc8648726d1","entry":"split_sentences","repo":"eth-sri/chatprotect","repo_kind":"official","path":"chatprotect/util.py","file_url":"https://github.com/eth-sri/chatprotect/blob/HEAD/chatprotect/util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"00998fc8648726d1"}},{"code_sha256_prefix":"20c4db5099969535","entry":"to_hour","repo":"eth-sri/chatprotect","repo_kind":"official","path":"chatprotect/bots/openai_chat_stream_api.py","file_url":"https://github.com/eth-sri/chatprotect/blob/HEAD/chatprotect/bots/openai_chat_stream_api.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"20c4db5099969535"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}