{"id":"https://openalex.org/W7163711728","doi":"https://doi.org/10.48550/arxiv.2606.06350","title":"EDIT: Evidence-Diagnosed Intervention Training for Rule-Faithful LLM Grading","display_name":"EDIT: Evidence-Diagnosed Intervention Training for Rule-Faithful LLM Grading","publication_year":2026,"publication_date":"2026-06-04","ids":{"openalex":"https://openalex.org/W7163711728","doi":"https://doi.org/10.48550/arxiv.2606.06350"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.06350","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.06350","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.06350","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137950002","display_name":"Zhihao Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Zhihao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138013527","display_name":"Linhai Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Linhai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137998143","display_name":"Taiyi Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Taiyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138004430","display_name":"Runcong Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Runcong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137988197","display_name":"Peter Andrews","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Andrews, Peter","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137973017","display_name":"Cesare Aloisi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Aloisi, Cesare","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138004003","display_name":"Yulan He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Yulan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.6197999715805054,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.6197999715805054,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.1476999968290329,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.03579999879002571,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/rubric","display_name":"Rubric","score":0.8889999985694885},{"id":"https://openalex.org/keywords/grading","display_name":"Grading (engineering)","score":0.7731000185012817},{"id":"https://openalex.org/keywords/judgement","display_name":"Judgement","score":0.5320000052452087},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.42890000343322754},{"id":"https://openalex.org/keywords/intervention","display_name":"Intervention (counseling)","score":0.36800000071525574}],"concepts":[{"id":"https://openalex.org/C111640148","wikidata":"https://www.wikidata.org/wiki/Q847349","display_name":"Rubric","level":2,"score":0.8889999985694885},{"id":"https://openalex.org/C2777286243","wikidata":"https://www.wikidata.org/wiki/Q5591926","display_name":"Grading (engineering)","level":2,"score":0.7731000185012817},{"id":"https://openalex.org/C2776548248","wikidata":"https://www.wikidata.org/wiki/Q12621536","display_name":"Judgement","level":2,"score":0.5320000052452087},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.51910001039505},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5163000226020813},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.43459999561309814},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.42890000343322754},{"id":"https://openalex.org/C2780665704","wikidata":"https://www.wikidata.org/wiki/Q959298","display_name":"Intervention (counseling)","level":2,"score":0.36800000071525574},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.349700003862381},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.29989999532699585},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.29589998722076416},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.2720000147819519}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.06350","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.06350","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.06350","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.06350","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"score":0.4602282643318176,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reliable":[0],"rubric":[1,110],"grading":[2,49,136],"requires":[3],"more":[4,76],"than":[5],"accurate":[6],"score":[7],"prediction.":[8],"Each":[9],"judgement":[10],"must":[11],"be":[12],"grounded":[13],"in":[14,40],"the":[15,21,55,59,93,115],"mark":[16,61,95],"scheme":[17],"and":[18,26,96,146,153],"evidence":[19],"from":[20,108],"student":[22],"answer.":[23],"Existing":[24],"credit-assignment":[25],"intervention":[27],"methods,":[28],"primarily":[29],"designed":[30],"for":[31,74],"self-contained":[32],"reasoning":[33,50,84],"tasks":[34],"such":[35],"as":[36],"mathematics":[37],"reasoning,":[38],"struggle":[39],"this":[41],"setting":[42],"because":[43],"they":[44],"do":[45],"not":[46],"identify":[47],"where":[48],"goes":[51],"wrong":[52],"or":[53],"how":[54],"model's":[56],"belief":[57,91,124],"about":[58],"final":[60,94],"changes":[62],"during":[63],"reasoning.":[64],"We":[65],"propose":[66],"Evidence-Diagnosed":[67],"Intervention":[68],"Training":[69],"(EDIT),":[70],"a":[71,109],"two-phase":[72],"framework":[73],"training":[75],"rubric-faithful":[77],"LLM":[78],"graders.":[79],"First,":[80],"EDIT-SFT":[81],"locates":[82],"problematic":[83],"steps":[85,105],"using":[86],"internal":[87],"model":[88],"signals:":[89],"posterior":[90],"over":[92],"input-grounding":[97],"scores.":[98],"It":[99],"then":[100],"revises":[101],"only":[102],"these":[103,164],"local":[104],"with":[106,117,156],"help":[107],"checklist.":[111],"Second,":[112],"EDIT-RL":[113],"calibrates":[114],"grader":[116],"belief-guided":[118],"reward":[119],"shaping,":[120],"penalising":[121],"large":[122],"harmful":[123],"drifts":[125],"while":[126],"still":[127],"allowing":[128],"helpful":[129],"exploration.":[130],"Experiments":[131],"on":[132,150],"two":[133],"real-world,":[134],"multi-subject":[135],"benchmarks":[137],"demonstrate":[138],"that":[139,160],"EDIT":[140],"consistently":[141],"outperforms":[142],"strong":[143],"supervised":[144],"fine-tuning":[145],"reinforcement":[147],"learning":[148],"baselines":[149],"both":[151],"in-domain":[152],"out-of-domain":[154],"splits,":[155],"ablation":[157],"studies":[158],"confirming":[159],"internal-state":[161],"diagnostics":[162],"drive":[163],"gains.":[165]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-06T00:00:00"}