{"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/semantic-entropy-probes-robust-and-cheap","title":"Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMs","arxiv_id":"2406.15927","date":"2024-06-22","proceeding":null,"authors":["Jannik Kossen","Jiatong Han","Muhammed Razzak","Lisa Schut","Shreshth Malik","Yarin Gal"],"abstract":"We propose semantic entropy probes (SEPs), a cheap and reliable method for uncertainty quantification in Large Language Models (LLMs). Hallucinations, which are plausible-sounding but factually incorrect and arbitrary model generations, present a major challenge to the practical adoption of LLMs. Recent work by Farquhar et al. (2024) proposes semantic entropy (SE), which can detect hallucinations by estimating uncertainty in the space semantic meaning for a set of model generations. However, the 5-to-10-fold increase in computation cost associated with SE computation hinders practical adoption. To address this, we propose SEPs, which directly approximate SE from the hidden states of a single generation. SEPs are simple to train and do not require sampling multiple model generations at test time, reducing the overhead of semantic uncertainty quantification to almost zero. We show that SEPs retain high performance for hallucination detection and generalize better to out-of-distribution data than previous probing methods that directly predict model accuracy. Our results across models and tasks suggest that model hidden states capture SE, and our ablation studies give further insights into the token positions and model layers for which this is the case.","url_abs":"https://arxiv.org/abs/2406.15927v1","url_pdf":"https://arxiv.org/pdf/2406.15927v1.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":"semantic-entropy-probes-robust-and-cheap","repo_url":"https://github.com/oatml/semantic-entropy-probes","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"hallucination","task_name":"Hallucination"},{"task_slug":"uncertainty-quantification","task_name":"Uncertainty Quantification"}],"methods":[{"method_slug":"set","method_name":"SET"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2406.15927","atlas_url":"https://app.syntology.ai/?focus=2406.15927","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2406.15927"}},"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/oatml/semantic-entropy-probes","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":12,"unverified":1},"by_repo_kind":{"listed":{"samples":13,"ran":12,"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":"c425058e87c88adb","entry":"accuracy_at_quantile","repo":"oatml/semantic-entropy-probes","repo_kind":"listed","path":"semantic_uncertainty/uncertainty/utils/eval_utils.py","file_url":"https://github.com/oatml/semantic-entropy-probes/blob/HEAD/semantic_uncertainty/uncertainty/utils/eval_utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c425058e87c88adb"}},{"code_sha256_prefix":"ccde4062eb35a7ad","entry":"auroc","repo":"oatml/semantic-entropy-probes","repo_kind":"listed","path":"semantic_uncertainty/uncertainty/utils/eval_utils.py","file_url":"https://github.com/oatml/semantic-entropy-probes/blob/HEAD/semantic_uncertainty/uncertainty/utils/eval_utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ccde4062eb35a7ad"}},{"code_sha256_prefix":"4b00bf597ecfdaac","entry":"bootstrap","repo":"oatml/semantic-entropy-probes","repo_kind":"listed","path":"semantic_uncertainty/uncertainty/utils/eval_utils.py","file_url":"https://github.com/oatml/semantic-entropy-probes/blob/HEAD/semantic_uncertainty/uncertainty/utils/eval_utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4b00bf597ecfdaac"}},{"code_sha256_prefix":"ab0a03d09dc95c36","entry":"calculate_p_true","repo":"oatml/semantic-entropy-probes","repo_kind":"listed","path":"semantic_uncertainty/uncertainty/uncertainty_measures/p_true.py","file_url":"https://github.com/oatml/semantic-entropy-probes/blob/HEAD/semantic_uncertainty/uncertainty/uncertainty_measures/p_true.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ab0a03d09dc95c36"}},{"code_sha256_prefix":"0b79bb92328a4653","entry":"construct_fewshot_prompt_from_indices","repo":"oatml/semantic-entropy-probes","repo_kind":"listed","path":"semantic_uncertainty/uncertainty/utils/utils.py","file_url":"https://github.com/oatml/semantic-entropy-probes/blob/HEAD/semantic_uncertainty/uncertainty/utils/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0b79bb92328a4653"}},{"code_sha256_prefix":"0814f57f66c87087","entry":"context_entails_response","repo":"oatml/semantic-entropy-probes","repo_kind":"listed","path":"semantic_uncertainty/uncertainty/uncertainty_measures/semantic_entropy.py","file_url":"https://github.com/oatml/semantic-entropy-probes/blob/HEAD/semantic_uncertainty/uncertainty/uncertainty_measures/semantic_entropy.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0814f57f66c87087"}},{"code_sha256_prefix":"e5a9b05c0453abbc","entry":"get_parser","repo":"oatml/semantic-entropy-probes","repo_kind":"listed","path":"semantic_uncertainty/uncertainty/utils/utils.py","file_url":"https://github.com/oatml/semantic-entropy-probes/blob/HEAD/semantic_uncertainty/uncertainty/utils/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e5a9b05c0453abbc"}},{"code_sha256_prefix":"80a6075a64b753ae","entry":"get_semantic_ids","repo":"oatml/semantic-entropy-probes","repo_kind":"listed","path":"semantic_uncertainty/uncertainty/uncertainty_measures/semantic_entropy.py","file_url":"https://github.com/oatml/semantic-entropy-probes/blob/HEAD/semantic_uncertainty/uncertainty/uncertainty_measures/semantic_entropy.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"80a6075a64b753ae"}},{"code_sha256_prefix":"b95c0c18affed908","entry":"logsumexp_by_id","repo":"oatml/semantic-entropy-probes","repo_kind":"listed","path":"semantic_uncertainty/uncertainty/uncertainty_measures/semantic_entropy.py","file_url":"https://github.com/oatml/semantic-entropy-probes/blob/HEAD/semantic_uncertainty/uncertainty/uncertainty_measures/semantic_entropy.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b95c0c18affed908"}},{"code_sha256_prefix":"a7132c6a5c4b94e4","entry":"md5hash","repo":"oatml/semantic-entropy-probes","repo_kind":"listed","path":"semantic_uncertainty/uncertainty/utils/openai.py","file_url":"https://github.com/oatml/semantic-entropy-probes/blob/HEAD/semantic_uncertainty/uncertainty/utils/openai.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a7132c6a5c4b94e4"}},{"code_sha256_prefix":"50909f87202233f8","entry":"remove_split_layer","repo":"oatml/semantic-entropy-probes","repo_kind":"listed","path":"semantic_uncertainty/uncertainty/models/huggingface_models.py","file_url":"https://github.com/oatml/semantic-entropy-probes/blob/HEAD/semantic_uncertainty/uncertainty/models/huggingface_models.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"50909f87202233f8"}},{"code_sha256_prefix":"eb7cd0b37505f35f","entry":"split_dataset","repo":"oatml/semantic-entropy-probes","repo_kind":"listed","path":"semantic_uncertainty/uncertainty/utils/utils.py","file_url":"https://github.com/oatml/semantic-entropy-probes/blob/HEAD/semantic_uncertainty/uncertainty/utils/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"eb7cd0b37505f35f"}},{"code_sha256_prefix":"f85264d7c7639ee8","entry":"get_p_ik","repo":"oatml/semantic-entropy-probes","repo_kind":"listed","path":"semantic_uncertainty/uncertainty/uncertainty_measures/p_ik.py","file_url":"https://github.com/oatml/semantic-entropy-probes/blob/HEAD/semantic_uncertainty/uncertainty/uncertainty_measures/p_ik.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f85264d7c7639ee8"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}