{"id":"https://openalex.org/W7225611908","doi":"https://doi.org/10.5281/zenodo.24073661","title":"Subconscious Latent Offloading: Single-Shot System-RAM Steering Across Quantized and Cross-Scale Language Models","display_name":"Subconscious Latent Offloading: Single-Shot System-RAM Steering Across Quantized and Cross-Scale Language Models","publication_year":2026,"publication_date":"2026-11-09","ids":{"openalex":"https://openalex.org/W7225611908","doi":"https://doi.org/10.5281/zenodo.24073661"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.24073661","doi":"https://doi.org/10.5281/zenodo.24073661","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24073661","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.5281/zenodo.24073661","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5161305155","display_name":"Pratham Solanki","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Solanki, Pratham","raw_affiliation_strings":["Independent AI Systems Researcher"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Independent AI Systems Researcher","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5161305155"],"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.8306000232696533,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.8306000232696533,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10181","display_name":"Natural Language Processing Techniques","score":0.0560000017285347,"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":0.03669999912381172,"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":[],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7775999903678894},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.47679999470710754},{"id":"https://openalex.org/C12186640","wikidata":"https://www.wikidata.org/wiki/Q6815743","display_name":"Memory model","level":3,"score":0.4025999903678894},{"id":"https://openalex.org/C46900642","wikidata":"https://www.wikidata.org/wiki/Q2647","display_name":"Huffman coding","level":3,"score":0.3847000002861023},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.3734000027179718},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.36800000071525574},{"id":"https://openalex.org/C205711294","wikidata":"https://www.wikidata.org/wiki/Q176953","display_name":"Rendering (computer graphics)","level":2,"score":0.3605000078678131},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.3549000024795532},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.35359999537467957},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.3346000015735626},{"id":"https://openalex.org/C61224824","wikidata":"https://www.wikidata.org/wiki/Q2260434","display_name":"Mixture model","level":2,"score":0.3222000002861023},{"id":"https://openalex.org/C193702766","wikidata":"https://www.wikidata.org/wiki/Q1414548","display_name":"Concurrency","level":2,"score":0.32100000977516174},{"id":"https://openalex.org/C64543145","wikidata":"https://www.wikidata.org/wiki/Q162942","display_name":"Intersection (aeronautics)","level":2,"score":0.3172000050544739},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3118000030517578},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.3082999885082245},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.3001999855041504},{"id":"https://openalex.org/C75165309","wikidata":"https://www.wikidata.org/wiki/Q2258979","display_name":"Search engine indexing","level":2,"score":0.30000001192092896},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.29820001125335693},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2939999997615814},{"id":"https://openalex.org/C153338461","wikidata":"https://www.wikidata.org/wiki/Q2651","display_name":"Arithmetic coding","level":4,"score":0.2906999886035919},{"id":"https://openalex.org/C189783530","wikidata":"https://www.wikidata.org/wiki/Q352090","display_name":"CPU cache","level":3,"score":0.2856000065803528},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.27970001101493835},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.2793999910354614},{"id":"https://openalex.org/C2778514511","wikidata":"https://www.wikidata.org/wiki/Q1374194","display_name":"Programmer","level":2,"score":0.2761000096797943},{"id":"https://openalex.org/C96851999","wikidata":"https://www.wikidata.org/wiki/Q848369","display_name":"Subconscious","level":3,"score":0.27379998564720154},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.2703000009059906},{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.26759999990463257},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.2660999894142151},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.25780001282691956},{"id":"https://openalex.org/C189930140","wikidata":"https://www.wikidata.org/wiki/Q1112878","display_name":"CAS latency","level":4,"score":0.2554999887943268},{"id":"https://openalex.org/C100279451","wikidata":"https://www.wikidata.org/wiki/Q372193","display_name":"Perplexity","level":3,"score":0.25459998846054077},{"id":"https://openalex.org/C164155591","wikidata":"https://www.wikidata.org/wiki/Q2067766","display_name":"Satisfiability modulo theories","level":2,"score":0.25380000472068787},{"id":"https://openalex.org/C70388272","wikidata":"https://www.wikidata.org/wiki/Q5968558","display_name":"IBM","level":2,"score":0.2515000104904175}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.5281/zenodo.24073661","doi":"https://doi.org/10.5281/zenodo.24073661","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24073661","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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"},{"id":"doi:10.5281/zenodo.24088952","doi":"https://doi.org/10.5281/zenodo.24088952","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24088952","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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":"doi:10.5281/zenodo.24073661","doi":"https://doi.org/10.5281/zenodo.24073661","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24073661","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Deploying":[0],"reasoning-capable":[1],"Large":[2],"Language":[3],"Models":[4],"(LLMs)":[5],"on":[6,27],"consumer":[7],"edge":[8],"GPUs":[9],"is":[10],"fundamentally":[11],"constrained":[12],"by":[13,37,221,231],"VRAM":[14],"capacity":[15],"and":[16,95,127,175,182,196,228],"PCIe/System-RAM":[17],"memory":[18],"bandwidth.":[19],"Standard":[20],"layer-offloading":[21],"engines":[22],"execute":[23],"offloaded":[24],"CPU/RAM":[25],"layers":[26,73],"every":[28],"generated":[29],"token,":[30],"reducing":[31,176,218],"autoregressive":[32],"decoding":[33],"to":[34,111,158,173,180],"2\u20135":[35],"tokens/s.Inspired":[36],"a":[38,57,66,132,149,184,189,197],"modular":[39],"cognitive":[40],"paradigm\u2014\u201cThe":[41],"brain":[42],"does":[43],"the":[44,46],"thinking;":[45],"mouth":[47],"speaks":[48],"it\u201d\u2014we":[49],"introduce":[50],"Subconscious":[51,76],"Latent":[52],"Offloading,":[53],"an":[54,159,206],"architecture":[55],"where":[56,148],"fast,":[58],"self-contained":[59],"GPU":[60,93,106,163,191],"model":[61,192],"(the":[62,75],"Active":[63],"Mouth)":[64],"invokes":[65],"memory-mapped":[67],"System-RAM":[68,152,199],"slice":[69,153,200],"of":[70],"middle":[71],"reasoning":[72,98],"L12..19":[74],"Brain)":[77],"strictly":[78],"once":[79],"per":[80],"query":[81],"during":[82],"prompt":[83],"prefill.":[84],"By":[85],"over-lapping":[86],"this":[87],"single-shot":[88],"CPU":[89,229],"pass":[90],"asynchronously":[91],"behind":[92],"prefill":[94],"hybrid":[96],"symbolic-continuous":[97],"cycles":[99],"(Caveman":[100],"Anchors":[101],"+":[102,213],"Soft-Token":[103],"Superposition),":[104],"effective":[105],"blocking":[107],"wait":[108],"time":[109],"drops":[110],"8\u201312":[112],"microseconds":[113],"(0.008\u20130.012":[114],"ms).":[115],"We":[116],"evaluate":[117],"two":[118],"distinct":[119],"operating":[120],"regimes":[121],"across":[122],"10":[123],"rigorous":[124],"distributed":[125],"systems":[126],"concurrency":[128],"benchmarks":[129],"(P01\u2013P10):":[130],"(1)":[131],"Zero-Calibration":[133],"Same-Model":[134],"Precision":[135],"Bridge":[136],"(Wup":[137],"=I,":[138],"Wdown":[139],"=":[140,194,202],"I,":[141],"zero":[142,145],"calibration":[143],"prompts,":[144],"trained":[146],"parameters),":[147],"crisp":[150],"FP16":[151,211],"restores":[154],"middle-layer":[155],"arithmetic":[156],"fidelity":[157],"aggressively":[160],"low-bit":[161],"quantized":[162],"model,":[164],"jumping":[165],"zero-shot":[166],"accuracy":[167],"from":[168,178],"10%":[169],"(0%":[170],"unbridged":[171],"Caveman)":[172],"60%":[174],"perplexity":[177],"9.05":[179],"5.75;":[181],"(2)":[183],"Cross-Scale":[185],"Subspace":[186],"Translator":[187],"bridging":[188],"1.5B":[190],"(d":[193,201],"1536)":[195],"7B":[198],"3584)":[203],"compressed":[204],"via":[205],"Intra-Layer":[207],"Mixed-Precision":[208],"Split":[209],"(1%":[210],"outliers":[212],"99%":[214],"packed":[215],"INT4":[216],"bulk),":[217],"RAM":[219],"footprint":[220],"3.48\u00d7":[222],"(2816":[223],"MB":[224],"\u2192":[225,235],"808":[226],"MB)":[227],"latency":[230],"2.69\u00d7":[232],"(125.0":[233],"ms":[234],"45.9":[236],"ms)":[237],"at":[238],"0.9993":[239],"cosine":[240],"fidelity.":[241]},"counts_by_year":[],"updated_date":"2026-11-10T13:44:29.811591","created_date":"2026-11-10T00:00:00"}