{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:2BFLASDTLNCLBJ6EQFVGCSGYGG","short_pith_number":"pith:2BFLASDT","schema_version":"1.0","canonical_sha256":"d04ab048735b44b0a7c4816a6148d831803b5de566a52591b9f10c6de9fbd076","source":{"kind":"arxiv","id":"2109.10847","version":2},"attestation_state":"computed","paper":{"title":"Small-Bench NLP: Benchmark for small single GPU trained models in Natural Language Processing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Bhuvana Kundumani, Kamal Raj Kanakarajan, Malaikannan Sankarasubbu","submitted_at":"2021-09-22T17:18:55Z","abstract_excerpt":"Recent progress in the Natural Language Processing domain has given us several State-of-the-Art (SOTA) pretrained models which can be finetuned for specific tasks. These large models with billions of parameters trained on numerous GPUs/TPUs over weeks are leading in the benchmark leaderboards. In this paper, we discuss the need for a benchmark for cost and time effective smaller models trained on a single GPU. This will enable researchers with resource constraints experiment with novel and innovative ideas on tokenization, pretraining tasks, architecture, fine tuning methods etc. We set up Sma"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2109.10847","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-09-22T17:18:55Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"bb384d09ac67254a8b9d36c02a3142e6876287c21d8cee6979d43ee2d25ec5a9","abstract_canon_sha256":"fa081a067f661cff210963f2a28c817f85663fbb24c068f8f460ab7b9fc9dc79"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:16:54.275720Z","signature_b64":"/XxyVSMYOz+0lssTxGSegetnAVIuVAdIsENtsUEzGB7RE3uPcys03NyMPa01MFuij9Bkvvek9AHTKfU452XLCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d04ab048735b44b0a7c4816a6148d831803b5de566a52591b9f10c6de9fbd076","last_reissued_at":"2026-07-05T03:16:54.275302Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:16:54.275302Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Small-Bench NLP: Benchmark for small single GPU trained models in Natural Language Processing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Bhuvana Kundumani, Kamal Raj Kanakarajan, Malaikannan Sankarasubbu","submitted_at":"2021-09-22T17:18:55Z","abstract_excerpt":"Recent progress in the Natural Language Processing domain has given us several State-of-the-Art (SOTA) pretrained models which can be finetuned for specific tasks. These large models with billions of parameters trained on numerous GPUs/TPUs over weeks are leading in the benchmark leaderboards. In this paper, we discuss the need for a benchmark for cost and time effective smaller models trained on a single GPU. This will enable researchers with resource constraints experiment with novel and innovative ideas on tokenization, pretraining tasks, architecture, fine tuning methods etc. We set up Sma"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.10847","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2109.10847/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2109.10847","created_at":"2026-07-05T03:16:54.275370+00:00"},{"alias_kind":"arxiv_version","alias_value":"2109.10847v2","created_at":"2026-07-05T03:16:54.275370+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.10847","created_at":"2026-07-05T03:16:54.275370+00:00"},{"alias_kind":"pith_short_12","alias_value":"2BFLASDTLNCL","created_at":"2026-07-05T03:16:54.275370+00:00"},{"alias_kind":"pith_short_16","alias_value":"2BFLASDTLNCLBJ6E","created_at":"2026-07-05T03:16:54.275370+00:00"},{"alias_kind":"pith_short_8","alias_value":"2BFLASDT","created_at":"2026-07-05T03:16:54.275370+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2111.09543","citing_title":"DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2006.03654","citing_title":"DeBERTa: Decoding-enhanced BERT with Disentangled Attention","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2BFLASDTLNCLBJ6EQFVGCSGYGG","json":"https://pith.science/pith/2BFLASDTLNCLBJ6EQFVGCSGYGG.json","graph_json":"https://pith.science/api/pith-number/2BFLASDTLNCLBJ6EQFVGCSGYGG/graph.json","events_json":"https://pith.science/api/pith-number/2BFLASDTLNCLBJ6EQFVGCSGYGG/events.json","paper":"https://pith.science/paper/2BFLASDT"},"agent_actions":{"view_html":"https://pith.science/pith/2BFLASDTLNCLBJ6EQFVGCSGYGG","download_json":"https://pith.science/pith/2BFLASDTLNCLBJ6EQFVGCSGYGG.json","view_paper":"https://pith.science/paper/2BFLASDT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2109.10847&json=true","fetch_graph":"https://pith.science/api/pith-number/2BFLASDTLNCLBJ6EQFVGCSGYGG/graph.json","fetch_events":"https://pith.science/api/pith-number/2BFLASDTLNCLBJ6EQFVGCSGYGG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2BFLASDTLNCLBJ6EQFVGCSGYGG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2BFLASDTLNCLBJ6EQFVGCSGYGG/action/storage_attestation","attest_author":"https://pith.science/pith/2BFLASDTLNCLBJ6EQFVGCSGYGG/action/author_attestation","sign_citation":"https://pith.science/pith/2BFLASDTLNCLBJ6EQFVGCSGYGG/action/citation_signature","submit_replication":"https://pith.science/pith/2BFLASDTLNCLBJ6EQFVGCSGYGG/action/replication_record"}},"created_at":"2026-07-05T03:16:54.275370+00:00","updated_at":"2026-07-05T03:16:54.275370+00:00"}