{"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/pseudolikelihood-reranking-with-masked","title":"Masked Language Model Scoring","arxiv_id":"1910.14659","date":"2019-10-31","proceeding":"ACL 2020 6","authors":["Julian Salazar","Davis Liang","Toan Q. Nguyen","Katrin Kirchhoff"],"abstract":"Pretrained masked language models (MLMs) require finetuning for most NLP tasks. Instead, we evaluate MLMs out of the box via their pseudo-log-likelihood scores (PLLs), which are computed by masking tokens one by one. We show that PLLs outperform scores from autoregressive language models like GPT-2 in a variety of tasks. By rescoring ASR and NMT hypotheses, RoBERTa reduces an end-to-end LibriSpeech model's WER by 30% relative and adds up to +1.7 BLEU on state-of-the-art baselines for low-resource translation pairs, with further gains from domain adaptation. We attribute this success to PLL's unsupervised expression of linguistic acceptability without a left-to-right bias, greatly improving on scores from GPT-2 (+10 points on island effects, NPI licensing in BLiMP). One can finetune MLMs to give scores without masking, enabling computation in a single inference pass. In all, PLLs and their associated pseudo-perplexities (PPPLs) enable plug-and-play use of the growing number of pretrained MLMs; e.g., we use a single cross-lingual model to rescore translations in multiple languages. We release our library for language model scoring at https://github.com/awslabs/mlm-scoring.","url_abs":"https://arxiv.org/abs/1910.14659v3","url_pdf":"https://arxiv.org/pdf/1910.14659v3.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":"pseudolikelihood-reranking-with-masked","repo_url":"https://github.com/awslabs/mlm-scoring","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"mxnet","reach":null},{"paper_slug":"pseudolikelihood-reranking-with-masked","repo_url":"https://github.com/google-research/electra","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"pseudolikelihood-reranking-with-masked","repo_url":"https://github.com/ipieter/universal-distillation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"pseudolikelihood-reranking-with-masked","repo_url":"https://github.com/phueb/unmasked","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"pseudolikelihood-reranking-with-masked","repo_url":"https://github.com/soyoung97/fairseq-mlm-scoring","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"pseudolikelihood-reranking-with-masked","repo_url":"https://github.com/soyoung97/mlm-scoring","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"linguistic-acceptability","task_name":"Linguistic Acceptability"},{"task_slug":"nmt","task_name":"NMT"},{"task_slug":"translation","task_name":"Translation"},{"task_slug":"model","task_name":"model"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bert","method_name":"BERT"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"cosine-annealing","method_name":"Cosine Annealing"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"discriminative-fine-tuning","method_name":"Discriminative Fine-Tuning"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt-2","method_name":"GPT-2"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-cosine-annealing","method_name":"Linear Warmup With Cosine Annealing"},{"method_slug":"linear-warmup-with-linear-decay","method_name":"Linear Warmup With Linear Decay"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"roberta","method_name":"RoBERTa"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"weight-decay","method_name":"Weight Decay"},{"method_slug":"wordpiece","method_name":"WordPiece"},{"method_slug":"xlm","method_name":"XLM"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1910.14659","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1910.14659"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+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/soyoung97/fairseq-mlm-scoring","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/awslabs/mlm-scoring","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/phueb/unmasked","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/google-research/electra","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/soyoung97/mlm-scoring","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ipieter/universal-distillation","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran_honours":2,"ran_draft_wrong":3,"unverified":5},"by_repo_kind":{"listed":{"samples":7,"ran":2,"repositories":2}},"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":3,"samples":[{"code_sha256_prefix":"eb725d5794b15f6b","entry":"accuracy","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"eb725d5794b15f6b"}},{"code_sha256_prefix":"52a1f0891fc0ad6d","entry":"convert_example_to_features","repo":"soyoung97/mlm-scoring","repo_kind":"listed","path":"examples/run_lm_finetuning.py","file_url":"https://github.com/soyoung97/mlm-scoring/blob/HEAD/examples/run_lm_finetuning.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"52a1f0891fc0ad6d"}},{"code_sha256_prefix":"b3feaffc79a790e6","entry":"convert_examples_to_features","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"b3feaffc79a790e6"}},{"code_sha256_prefix":"fab8efcc34ebd909","entry":"random_word","repo":"soyoung97/mlm-scoring","repo_kind":"listed","path":"examples/run_lm_finetuning.py","file_url":"https://github.com/soyoung97/mlm-scoring/blob/HEAD/examples/run_lm_finetuning.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"fab8efcc34ebd909"}},{"code_sha256_prefix":"c58d57224530d17e","entry":"warmup_linear","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"c58d57224530d17e"}},{"code_sha256_prefix":"ce3259a298ec3e93","entry":"calc_accuracy_from_scores","repo":"phueb/unmasked","repo_kind":"listed","path":"unmasked/utils.py","file_url":"https://github.com/phueb/unmasked/blob/HEAD/unmasked/utils.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":"ce3259a298ec3e93"}},{"code_sha256_prefix":"e61b5f8d71122e38","entry":"get_group_names","repo":"phueb/unmasked","repo_kind":"listed","path":"unmasked/utils.py","file_url":"https://github.com/phueb/unmasked/blob/HEAD/unmasked/utils.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":"e61b5f8d71122e38"}},{"code_sha256_prefix":"e2a116c375db5070","entry":"get_legend_label","repo":"phueb/unmasked","repo_kind":"listed","path":"unmasked/utils.py","file_url":"https://github.com/phueb/unmasked/blob/HEAD/unmasked/utils.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":"e2a116c375db5070"}},{"code_sha256_prefix":"b5af14201018f768","entry":"make_ax_title","repo":"phueb/unmasked","repo_kind":"listed","path":"unmasked/visualizer.py","file_url":"https://github.com/phueb/unmasked/blob/HEAD/unmasked/visualizer.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":"b5af14201018f768"}},{"code_sha256_prefix":"d6af7b5cef80b3a2","entry":"shorten_phenomenon","repo":"phueb/unmasked","repo_kind":"listed","path":"unmasked/helpers.py","file_url":"https://github.com/phueb/unmasked/blob/HEAD/unmasked/helpers.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":"d6af7b5cef80b3a2"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}