{"id":"https://openalex.org/W4416034340","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.861","title":"Hopscotch: Discovering and Skipping Redundancies in Language Models","display_name":"Hopscotch: Discovering and Skipping Redundancies in Language Models","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034340","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.861"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.861","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.861","pdf_url":"https://aclanthology.org/2025.findings-emnlp.861.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-emnlp.861.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5030343262","display_name":"Mustafa Eyceoz","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mustafa Eyceoz","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020990010","display_name":"N. Nayak","orcid":"https://orcid.org/0000-0002-9588-3533"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nikhil Shivakumar Nayak","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100599815","display_name":"Hao Wang","orcid":"https://orcid.org/0000-0002-7308-938X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hao Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016596228","display_name":"Ligong Han","orcid":"https://orcid.org/0000-0003-3166-0848"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ligong Han","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5022716943","display_name":"Akash Srivastava","orcid":"https://orcid.org/0000-0002-5959-3926"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Akash Srivastava","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":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.29815066,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"15904","last_page":"15913"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.27810001373291016,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.27810001373291016,"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":0.23240000009536743,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.054099999368190765,"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/language-model","display_name":"Language model","score":0.4172999858856201},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.3386000096797943},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.28209999203681946},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.27379998564720154},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.26809999346733093}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6596999764442444},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5151000022888184},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4846000075340271},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4172999858856201},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.3386000096797943},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.2969000041484833},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.28209999203681946},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.27379998564720154},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.26809999346733093},{"id":"https://openalex.org/C129792486","wikidata":"https://www.wikidata.org/wiki/Q1050419","display_name":"Language identification","level":3,"score":0.25839999318122864}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.861","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.861","pdf_url":"https://aclanthology.org/2025.findings-emnlp.861.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.861","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.861","pdf_url":"https://aclanthology.org/2025.findings-emnlp.861.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416034340.pdf","grobid_xml":"https://content.openalex.org/works/W4416034340.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Modern":[0],"causal":[1],"language":[2],"models":[3],"stack":[4],"many":[5],"attention":[6,31,67,81,121],"blocks":[7,14,32,48],"to":[8,36,41,49,53,66,91,105],"improve":[9],"performance,":[10],"but":[11],"not":[12,84],"all":[13],"are":[15],"necessary":[16],"for":[17],"every":[18],"task.We":[19],"propose":[20],"Hopscotch,":[21],"a":[22,37,112],"simple":[23],"yet":[24],"effective":[25],"method":[26],"that":[27],"identifies":[28],"and":[29,39,51,68,96,107],"skips":[30],"with":[33,99],"least":[34],"contributions":[35],"task":[38],"adapts":[40],"preserve":[42],"output":[43],"quality.Hopscotch":[44],"jointly":[45],"optimizes":[46],"which":[47],"skip":[50],"how":[52],"scale":[54],"the":[55,58],"outputs":[56],"of":[57],"remaining":[59],"layers.By":[60],"introducing":[61],"lightweight,":[62],"trainable":[63],"scaling":[64],"parameters":[65],"MLP":[69],"blocks,":[70],"it":[71],"mitigates":[72],"distribution":[73],"shifts":[74],"in":[75,115],"hidden":[76],"states":[77],"caused":[78],"by":[79],"removing":[80],"blocks.Hopscotch":[82],"does":[83],"modify":[85],"model":[86,101],"weights":[87],"or":[88,93],"require":[89],"access":[90],"pretraining":[92],"instruction-tuning":[94],"data,":[95],"is":[97],"compatible":[98],"existing":[100],"compression":[102],"techniques.When":[103],"applied":[104],"Llama-3.1-8B":[106],"Qwen2.5-7B,Hopscotch":[108],"achieves":[109],"less":[110],"than":[111],"2%":[113],"drop":[114],"performance":[116],"even":[117],"after":[118],"skipping":[119],"four":[120],"blocks.":[122]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-11-08T00:00:00"}