{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QMULW2BFINQOFFZUF5MARGKMUE","short_pith_number":"pith:QMULW2BF","schema_version":"1.0","canonical_sha256":"8328bb68254360e297342f5808994ca107f1fbb2cb7ea0640efe08ac04a239ed","source":{"kind":"arxiv","id":"2501.09706","version":3},"attestation_state":"computed","paper":{"title":"Domain Adaptation of Foundation LLMs for e-Commerce","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Christian Herold, Dominika Basaj, Michael Kozielski, Patrycja Cieplicka, Pavel Petrushkov, Seyyed Hadi Hashemi, Shahram Khadivi, Tala Bazazo, Yannick Versley","submitted_at":"2025-01-16T17:58:32Z","abstract_excerpt":"We present the e-Llama models: 8 billion and 70 billion parameter large language models that are adapted towards the e-commerce domain. These models are meant as foundation models with deep knowledge about e-commerce, that form a base for instruction- and fine-tuning. The e-Llama models are obtained by continuously pretraining the Llama 3.1 base models on 1 trillion tokens of domain-specific data.\n  We discuss our approach and motivate our choice of hyperparameters with a series of ablation studies. To quantify how well the models have been adapted to the e-commerce domain, we define and imple"},"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":"2501.09706","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-16T17:58:32Z","cross_cats_sorted":[],"title_canon_sha256":"6504b651071953d2ed082c8e894d9899ae99fbf7f331c334f3ef8305f0bced4c","abstract_canon_sha256":"26f72df39cc80239fa6e1d47bff6a925ba8f30af67d5c9c10b279b39b0549fe5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:09:06.637413Z","signature_b64":"g2o88MJ7zON8Kysm50SkgK6OvQG7f1yLsM9Am/F5FLP3V/ZBOhOsykpj6PX8Ut7afCCZ+wDZnizoDCJEazz4CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8328bb68254360e297342f5808994ca107f1fbb2cb7ea0640efe08ac04a239ed","last_reissued_at":"2026-07-05T11:09:06.636888Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:09:06.636888Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Domain Adaptation of Foundation LLMs for e-Commerce","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Christian Herold, Dominika Basaj, Michael Kozielski, Patrycja Cieplicka, Pavel Petrushkov, Seyyed Hadi Hashemi, Shahram Khadivi, Tala Bazazo, Yannick Versley","submitted_at":"2025-01-16T17:58:32Z","abstract_excerpt":"We present the e-Llama models: 8 billion and 70 billion parameter large language models that are adapted towards the e-commerce domain. These models are meant as foundation models with deep knowledge about e-commerce, that form a base for instruction- and fine-tuning. The e-Llama models are obtained by continuously pretraining the Llama 3.1 base models on 1 trillion tokens of domain-specific data.\n  We discuss our approach and motivate our choice of hyperparameters with a series of ablation studies. To quantify how well the models have been adapted to the e-commerce domain, we define and imple"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.09706","kind":"arxiv","version":3},"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/2501.09706/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":"2501.09706","created_at":"2026-07-05T11:09:06.636950+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.09706v3","created_at":"2026-07-05T11:09:06.636950+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.09706","created_at":"2026-07-05T11:09:06.636950+00:00"},{"alias_kind":"pith_short_12","alias_value":"QMULW2BFINQO","created_at":"2026-07-05T11:09:06.636950+00:00"},{"alias_kind":"pith_short_16","alias_value":"QMULW2BFINQOFFZU","created_at":"2026-07-05T11:09:06.636950+00:00"},{"alias_kind":"pith_short_8","alias_value":"QMULW2BF","created_at":"2026-07-05T11:09:06.636950+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.06173","citing_title":"Retina-RAG: Retrieval-Augmented Vision-Language Modeling for Joint Retinal Diagnosis and Clinical Report Generation","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06173","citing_title":"Retina-RAG: Retrieval-Augmented Vision-Language Modeling for Joint Retinal Diagnosis and Clinical Report Generation","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QMULW2BFINQOFFZUF5MARGKMUE","json":"https://pith.science/pith/QMULW2BFINQOFFZUF5MARGKMUE.json","graph_json":"https://pith.science/api/pith-number/QMULW2BFINQOFFZUF5MARGKMUE/graph.json","events_json":"https://pith.science/api/pith-number/QMULW2BFINQOFFZUF5MARGKMUE/events.json","paper":"https://pith.science/paper/QMULW2BF"},"agent_actions":{"view_html":"https://pith.science/pith/QMULW2BFINQOFFZUF5MARGKMUE","download_json":"https://pith.science/pith/QMULW2BFINQOFFZUF5MARGKMUE.json","view_paper":"https://pith.science/paper/QMULW2BF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.09706&json=true","fetch_graph":"https://pith.science/api/pith-number/QMULW2BFINQOFFZUF5MARGKMUE/graph.json","fetch_events":"https://pith.science/api/pith-number/QMULW2BFINQOFFZUF5MARGKMUE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QMULW2BFINQOFFZUF5MARGKMUE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QMULW2BFINQOFFZUF5MARGKMUE/action/storage_attestation","attest_author":"https://pith.science/pith/QMULW2BFINQOFFZUF5MARGKMUE/action/author_attestation","sign_citation":"https://pith.science/pith/QMULW2BFINQOFFZUF5MARGKMUE/action/citation_signature","submit_replication":"https://pith.science/pith/QMULW2BFINQOFFZUF5MARGKMUE/action/replication_record"}},"created_at":"2026-07-05T11:09:06.636950+00:00","updated_at":"2026-07-05T11:09:06.636950+00:00"}