{"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/arxiv-2605-11601","title":"DiffScore: Text Evaluation Beyond Autoregressive Likelihood","arxiv_id":"2605.11601","date":"2026-05-12","proceeding":null,"authors":["Wen Lai","Yingli Shen","Dingnan Jin","Qing Cui","Jun Zhou","Maosong Sun","Alexander Fraser"],"abstract":"Autoregressive language models are widely used for text evaluation, however, their left-to-right factorization introduces positional bias, i.e., early tokens are scored with only leftward context, conflating architectural asymmetry with true text quality. We propose masked reconstruction as an alternative paradigm, where every token is scored using full bidirectional context. We introduce DiffScore, an evaluation framework built on Masked Large Diffusion Language Models. By measuring text recoverability across continuous masking rates, DiffScore eliminates positional bias and naturally establishes an evaluation hierarchy from local fluency to global coherence. We further provide diagnostic tools unavailable to autoregressive frameworks: multi-timestep quality profiles that decompose scores across masking rates, and bidirectional PMI decomposition that disentangles fluency from faithfulness. Experiments across ten benchmarks show that DiffScore consistently outperforms autoregressive baselines in both zero-shot and fine-tuned settings. The code is released at: https://github.com/wenlai-lavine/DiffScore.","url_abs":"https://arxiv.org/abs/2605.11601","url_pdf":"https://arxiv.org/pdf/2605.11601","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2605.11601","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2605.11601"}},"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":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/wenlai-lavine/DiffScore","reach":null}],"summary":{"ran":1,"unverified":4},"by_repo_kind":{"found_in_text":{"samples":5,"ran":1,"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":5,"samples":[{"code_sha256_prefix":"f2d6bd011a85ed8f","entry":"RandomMasker","repo":"wenlai-lavine/DiffScore","repo_kind":"found_in_text","path":"src/scoring/diffscore.py","file_url":"https://github.com/wenlai-lavine/DiffScore/blob/HEAD/src/scoring/diffscore.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f2d6bd011a85ed8f"}},{"code_sha256_prefix":"9a395a0b9fdc99c6","entry":"BaseMasker","repo":"wenlai-lavine/DiffScore","repo_kind":"found_in_text","path":"src/scoring/diffscore.py","file_url":"https://github.com/wenlai-lavine/DiffScore/blob/HEAD/src/scoring/diffscore.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9a395a0b9fdc99c6"}},{"code_sha256_prefix":"f2ec5949747ed070","entry":"DiffScoreResult","repo":"wenlai-lavine/DiffScore","repo_kind":"found_in_text","path":"src/scoring/diffscore.py","file_url":"https://github.com/wenlai-lavine/DiffScore/blob/HEAD/src/scoring/diffscore.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f2ec5949747ed070"}},{"code_sha256_prefix":"80e3abf70c626f52","entry":"DiffScorer","repo":"wenlai-lavine/DiffScore","repo_kind":"found_in_text","path":"src/scoring/diffscore.py","file_url":"https://github.com/wenlai-lavine/DiffScore/blob/HEAD/src/scoring/diffscore.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"80e3abf70c626f52"}},{"code_sha256_prefix":"4e44af40337c9eae","entry":"MDLMWrapper","repo":"wenlai-lavine/DiffScore","repo_kind":"found_in_text","path":"src/scoring/diffscore.py","file_url":"https://github.com/wenlai-lavine/DiffScore/blob/HEAD/src/scoring/diffscore.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4e44af40337c9eae"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.CL","source":"arxiv_2026.jsonl"},"syntology_extracted_results":null}