{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:RULL7LLRPMLZSXM7VGMQPGI3WE","short_pith_number":"pith:RULL7LLR","schema_version":"1.0","canonical_sha256":"8d16bfad717b17995d9fa99907991bb111519a6ffbc33535157bf493f5826e6e","source":{"kind":"arxiv","id":"2311.05296","version":2},"attestation_state":"computed","paper":{"title":"BeLLM: Backward Dependency Enhanced Large Language Model for Sentence Embeddings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jing Li, Xianming Li","submitted_at":"2023-11-09T11:53:52Z","abstract_excerpt":"Sentence embeddings are crucial in measuring semantic similarity. Most recent studies employed large language models (LLMs) to learn sentence embeddings. Existing LLMs mainly adopted autoregressive architecture without explicit backward dependency modeling. Therefore, we examined the effects of backward dependencies in LLMs for semantic similarity measurements. Concretely, we propose a novel model: backward dependency enhanced large language model (BeLLM). It learns sentence embeddings via transforming specific attention layers from uni- to bi-directional. We extensively experiment across vari"},"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":"2311.05296","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-09T11:53:52Z","cross_cats_sorted":[],"title_canon_sha256":"bc33efebd5d40aa2f1f7df676a0a777656d1bd4e5a67640296174925db1d235f","abstract_canon_sha256":"7e7b41cba1ba131fd5e58c450a927d98b63440c5e4cd47956b93c0d8fa008801"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:56:01.080939Z","signature_b64":"RSZ0vqjB6jpbfUs5NYdsN6CoQrZ0iyMnPdTqQT0fBx1HDHifR2wgtkDLUJLV235nOVWiaqerQYR0odgCR+rKCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8d16bfad717b17995d9fa99907991bb111519a6ffbc33535157bf493f5826e6e","last_reissued_at":"2026-07-05T07:56:01.080395Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:56:01.080395Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BeLLM: Backward Dependency Enhanced Large Language Model for Sentence Embeddings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jing Li, Xianming Li","submitted_at":"2023-11-09T11:53:52Z","abstract_excerpt":"Sentence embeddings are crucial in measuring semantic similarity. Most recent studies employed large language models (LLMs) to learn sentence embeddings. Existing LLMs mainly adopted autoregressive architecture without explicit backward dependency modeling. Therefore, we examined the effects of backward dependencies in LLMs for semantic similarity measurements. Concretely, we propose a novel model: backward dependency enhanced large language model (BeLLM). It learns sentence embeddings via transforming specific attention layers from uni- to bi-directional. We extensively experiment across vari"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.05296","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/2311.05296/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":"2311.05296","created_at":"2026-07-05T07:56:01.080461+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.05296v2","created_at":"2026-07-05T07:56:01.080461+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.05296","created_at":"2026-07-05T07:56:01.080461+00:00"},{"alias_kind":"pith_short_12","alias_value":"RULL7LLRPMLZ","created_at":"2026-07-05T07:56:01.080461+00:00"},{"alias_kind":"pith_short_16","alias_value":"RULL7LLRPMLZSXM7","created_at":"2026-07-05T07:56:01.080461+00:00"},{"alias_kind":"pith_short_8","alias_value":"RULL7LLR","created_at":"2026-07-05T07:56:01.080461+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.00389","citing_title":"CSE-SFP: Enabling Unsupervised Sentence Representation Learning via a Single Forward Pass","ref_index":18,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RULL7LLRPMLZSXM7VGMQPGI3WE","json":"https://pith.science/pith/RULL7LLRPMLZSXM7VGMQPGI3WE.json","graph_json":"https://pith.science/api/pith-number/RULL7LLRPMLZSXM7VGMQPGI3WE/graph.json","events_json":"https://pith.science/api/pith-number/RULL7LLRPMLZSXM7VGMQPGI3WE/events.json","paper":"https://pith.science/paper/RULL7LLR"},"agent_actions":{"view_html":"https://pith.science/pith/RULL7LLRPMLZSXM7VGMQPGI3WE","download_json":"https://pith.science/pith/RULL7LLRPMLZSXM7VGMQPGI3WE.json","view_paper":"https://pith.science/paper/RULL7LLR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.05296&json=true","fetch_graph":"https://pith.science/api/pith-number/RULL7LLRPMLZSXM7VGMQPGI3WE/graph.json","fetch_events":"https://pith.science/api/pith-number/RULL7LLRPMLZSXM7VGMQPGI3WE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RULL7LLRPMLZSXM7VGMQPGI3WE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RULL7LLRPMLZSXM7VGMQPGI3WE/action/storage_attestation","attest_author":"https://pith.science/pith/RULL7LLRPMLZSXM7VGMQPGI3WE/action/author_attestation","sign_citation":"https://pith.science/pith/RULL7LLRPMLZSXM7VGMQPGI3WE/action/citation_signature","submit_replication":"https://pith.science/pith/RULL7LLRPMLZSXM7VGMQPGI3WE/action/replication_record"}},"created_at":"2026-07-05T07:56:01.080461+00:00","updated_at":"2026-07-05T07:56:01.080461+00:00"}