{"id":"https://openalex.org/W7123249017","doi":"https://doi.org/10.48550/arxiv.2601.05770","title":"Weights to Code: Extracting Interpretable Algorithms from the Discrete Transformer","display_name":"Weights to Code: Extracting Interpretable Algorithms from the Discrete Transformer","publication_year":2026,"publication_date":"2026-01-09","ids":{"openalex":"https://openalex.org/W7123249017","doi":"https://doi.org/10.48550/arxiv.2601.05770"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2601.05770","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2601.05770","pdf_url":"https://arxiv.org/pdf/2601.05770","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2601.05770","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5122818881","display_name":"Yifan Zhang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yifan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101671513","display_name":"Wei Bi","orcid":"https://orcid.org/0000-0002-0308-4217","observed_orcids":["https://orcid.org/0000-0002-0308-4217"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bi, Wei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122771883","display_name":"Kechi Zhang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Kechi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114245528","display_name":"Dongming Jin","orcid":"https://orcid.org/0009-0002-8164-6227","observed_orcids":["https://orcid.org/0009-0002-8164-6227","https://orcid.org/0000-0002-5451-2816"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jin, Dongming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101713885","display_name":"Jie Fu","orcid":"https://orcid.org/0000-0002-4494-843X","observed_orcids":["https://orcid.org/0000-0002-4494-843X","https://orcid.org/0000-0003-1496-3919"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fu, Jie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133033244","display_name":"Zhi Jin","orcid":"https://orcid.org/0000-0003-1087-226X","observed_orcids":["https://orcid.org/0000-0003-1087-226X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jin, Zhi","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":true,"cited_by_count":2,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.5264000296592712,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.5264000296592712,"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/T11975","display_name":"Evolutionary Algorithms and Applications","score":0.3882000148296356,"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/T10260","display_name":"Software Engineering Research","score":0.008100000210106373,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/interpretable-algorithms","display_name":"interpretable algorithms","score":0.6320000290870667},{"id":"https://openalex.org/keywords/symbolic-regression","display_name":"symbolic regression","score":0.9399999976158142},{"id":"https://openalex.org/keywords/hypothesis-testing","display_name":"hypothesis testing","score":0.8769999742507935},{"id":"https://openalex.org/keywords/inductive-bias","display_name":"inductive bias","score":0.5139999985694885},{"id":"https://openalex.org/keywords/transformer-interpretability","display_name":"transformer interpretability","score":0.7570000290870667},{"id":"https://openalex.org/keywords/mathematical-logic","display_name":"mathematical logic","score":0.40799999237060547}],"concepts":[{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7103999853134155},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.6507999897003174},{"id":"https://openalex.org/C2781067378","wikidata":"https://www.wikidata.org/wiki/Q17027399","display_name":"Interpretability","level":2,"score":0.6495000123977661},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6258000135421753},{"id":"https://openalex.org/C160145156","wikidata":"https://www.wikidata.org/wiki/Q778586","display_name":"Executable","level":2,"score":0.4772999882698059},{"id":"https://openalex.org/C34736171","wikidata":"https://www.wikidata.org/wiki/Q918333","display_name":"Preprocessor","level":2,"score":0.38499999046325684},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.357699990272522},{"id":"https://openalex.org/C65620979","wikidata":"https://www.wikidata.org/wiki/Q7661176","display_name":"Symbolic data analysis","level":2,"score":0.33959999680519104},{"id":"https://openalex.org/C126980161","wikidata":"https://www.wikidata.org/wiki/Q863783","display_name":"Simulated annealing","level":2,"score":0.31929999589920044},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.28619998693466187},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.2623000144958496},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.25589999556541443},{"id":"https://openalex.org/C73000952","wikidata":"https://www.wikidata.org/wiki/Q17007827","display_name":"Discretization","level":2,"score":0.2556999921798706},{"id":"https://openalex.org/C1893757","wikidata":"https://www.wikidata.org/wiki/Q3653001","display_name":"Inversion (geology)","level":3,"score":0.25290000438690186},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.25200000405311584}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2601.05770","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2601.05770","pdf_url":"https://arxiv.org/pdf/2601.05770","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2601.05770","doi":"https://doi.org/10.48550/arxiv.2601.05770","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2601.05770","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2601.05770","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2601.05770","pdf_url":"https://arxiv.org/pdf/2601.05770","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7123249017.pdf","grobid_xml":"https://content.openalex.org/works/W7123249017.grobid-xml"},"referenced_works_count":12,"referenced_works":["https://openalex.org/W2547875792","https://openalex.org/W2771655537","https://openalex.org/W2970971581","https://openalex.org/W2995107071","https://openalex.org/W3102263967","https://openalex.org/W4368303491","https://openalex.org/W4385572928","https://openalex.org/W4386839891","https://openalex.org/W4396570446","https://openalex.org/W4399795128","https://openalex.org/W4404928080","https://openalex.org/W4412945263"],"related_works":["https://openalex.org/W7128544229","https://openalex.org/W7133185117","https://openalex.org/W7133187725","https://openalex.org/W4414886925","https://openalex.org/W7128803984","https://openalex.org/W4392089795","https://openalex.org/W3082067696","https://openalex.org/W4415796113","https://openalex.org/W7131868694","https://openalex.org/W4399554936"],"abstract_inverted_index":{"Algorithm":[0],"extraction":[1,112,142],"aims":[2],"to":[3,33,64,90,101,113],"synthesize":[4],"executable":[5,19],"programs":[6],"directly":[7],"from":[8,21],"models":[9],"trained":[10],"on":[11,25,106],"algorithmic":[12],"tasks,":[13,109],"enabling":[14],"de":[15],"novo":[16],"recovery":[17,51],"of":[18,52],"mechanisms":[20],"weights":[22],"without":[23],"relying":[24],"human-written":[26],"target":[27],"programs.":[28,93],"However,":[29],"applying":[30],"this":[31],"paradigm":[32],"Transformer":[34,97,135,144],"is":[35],"complicated":[36],"by":[37],"representation":[38],"entanglement":[39],"(e.g.,":[40],"superposition),":[41],"where":[42],"features":[43],"encoded":[44],"in":[45],"overlapping":[46],"directions":[47],"substantially":[48],"hinder":[49],"the":[50,57,66,95,102,133],"symbolic":[53,73,88],"expressions.":[54],"We":[55],"propose":[56],"Discrete":[58,96,134],"Transformer,":[59],"an":[60],"architecture":[61],"explicitly":[62],"designed":[63],"bridge":[65],"gap":[67],"between":[68],"continuous":[69],"representations":[70],"and":[71,87,143],"discrete":[72,108],"logic.":[74],"By":[75],"injecting":[76],"discreteness":[77],"through":[78],"temperature-annealed":[79],"sampling,":[80],"our":[81],"framework":[82],"effectively":[83],"leverages":[84],"hypothesis":[85],"testing":[86],"regression":[89],"extract":[91],"human-readable":[92],"Empirically,":[94],"achieves":[98],"performance":[99],"comparable":[100],"RNN-based":[103],"MIPS":[104],"baseline":[105],"shared":[107],"while":[110],"broadening":[111],"tasks":[114],"with":[115],"continuous-valued":[116],"intermediate":[117],"computations.":[118],"Finally,":[119],"we":[120],"show":[121],"that":[122],"architectural":[123],"inductive":[124],"biases":[125],"provide":[126],"fine-grained":[127],"control":[128],"over":[129],"synthesized":[130],"programs,":[131],"establishing":[132],"as":[136],"a":[137],"controllable":[138],"testbed":[139],"for":[140],"algorithm":[141],"interpretability.":[145]},"counts_by_year":[{"year":2026,"cited_by_count":2}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-01-13T00:00:00"}