{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LKEH3FG5JMEMLA6N2CCBHTBCLT","short_pith_number":"pith:LKEH3FG5","schema_version":"1.0","canonical_sha256":"5a887d94dd4b08c583cdd08413cc225cdb63f132ea7c40dcbb69d2bc6a5cb804","source":{"kind":"arxiv","id":"2606.07502","version":1},"attestation_state":"computed","paper":{"title":"Your UnEmbedding Matrix is Secretly a Feature Lens for Text Embeddings","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Cong Li, Heng Cui, Rui Yan, Songhao Wu, Yuxuan Liu, Zhongxin Chen","submitted_at":"2026-06-05T17:54:32Z","abstract_excerpt":"Large language models exhibit impressive zero-shot capabilities across a wide range of downstream tasks. However, they struggle to function as off-the-shelf embedding models, leading to suboptimal performance on massive text embedding benchmarks. In this paper, we identify a potential cause underlying this deficiency. Our motivation stems from an unexpected observation: text embeddings tend to align with frequent but uninformative tokens when projected onto the vocabulary space. We argue that this excessive expression of high-frequency tokens suppresses the model's ability to capture nuanced s"},"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":"2606.07502","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-05T17:54:32Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"87e11d527bbe2ab13605fd3d065be2e2b1dcacf44a570d553b135239541bf7b1","abstract_canon_sha256":"9aab13584c9958f9c6d2f4e119b2657a22e0a7e7913f4f7f1b47175d8e8c5a0c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-08T01:05:31.368291Z","signature_b64":"qZ7cnCFq3xo14WX4ox4Zpt0cFE1eUlJVVA+OQk5Rtg4z9suJ2GFeT4MMCQrLvVsMdUgq3KIdUbkkIzl/2dIsDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a887d94dd4b08c583cdd08413cc225cdb63f132ea7c40dcbb69d2bc6a5cb804","last_reissued_at":"2026-06-08T01:05:31.367184Z","signature_status":"signed_v1","first_computed_at":"2026-06-08T01:05:31.367184Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Your UnEmbedding Matrix is Secretly a Feature Lens for Text Embeddings","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Cong Li, Heng Cui, Rui Yan, Songhao Wu, Yuxuan Liu, Zhongxin Chen","submitted_at":"2026-06-05T17:54:32Z","abstract_excerpt":"Large language models exhibit impressive zero-shot capabilities across a wide range of downstream tasks. However, they struggle to function as off-the-shelf embedding models, leading to suboptimal performance on massive text embedding benchmarks. In this paper, we identify a potential cause underlying this deficiency. Our motivation stems from an unexpected observation: text embeddings tend to align with frequent but uninformative tokens when projected onto the vocabulary space. We argue that this excessive expression of high-frequency tokens suppresses the model's ability to capture nuanced s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.07502","kind":"arxiv","version":1},"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/2606.07502/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":"2606.07502","created_at":"2026-06-08T01:05:31.367544+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.07502v1","created_at":"2026-06-08T01:05:31.367544+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.07502","created_at":"2026-06-08T01:05:31.367544+00:00"},{"alias_kind":"pith_short_12","alias_value":"LKEH3FG5JMEM","created_at":"2026-06-08T01:05:31.367544+00:00"},{"alias_kind":"pith_short_16","alias_value":"LKEH3FG5JMEMLA6N","created_at":"2026-06-08T01:05:31.367544+00:00"},{"alias_kind":"pith_short_8","alias_value":"LKEH3FG5","created_at":"2026-06-08T01:05:31.367544+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LKEH3FG5JMEMLA6N2CCBHTBCLT","json":"https://pith.science/pith/LKEH3FG5JMEMLA6N2CCBHTBCLT.json","graph_json":"https://pith.science/api/pith-number/LKEH3FG5JMEMLA6N2CCBHTBCLT/graph.json","events_json":"https://pith.science/api/pith-number/LKEH3FG5JMEMLA6N2CCBHTBCLT/events.json","paper":"https://pith.science/paper/LKEH3FG5"},"agent_actions":{"view_html":"https://pith.science/pith/LKEH3FG5JMEMLA6N2CCBHTBCLT","download_json":"https://pith.science/pith/LKEH3FG5JMEMLA6N2CCBHTBCLT.json","view_paper":"https://pith.science/paper/LKEH3FG5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.07502&json=true","fetch_graph":"https://pith.science/api/pith-number/LKEH3FG5JMEMLA6N2CCBHTBCLT/graph.json","fetch_events":"https://pith.science/api/pith-number/LKEH3FG5JMEMLA6N2CCBHTBCLT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LKEH3FG5JMEMLA6N2CCBHTBCLT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LKEH3FG5JMEMLA6N2CCBHTBCLT/action/storage_attestation","attest_author":"https://pith.science/pith/LKEH3FG5JMEMLA6N2CCBHTBCLT/action/author_attestation","sign_citation":"https://pith.science/pith/LKEH3FG5JMEMLA6N2CCBHTBCLT/action/citation_signature","submit_replication":"https://pith.science/pith/LKEH3FG5JMEMLA6N2CCBHTBCLT/action/replication_record"}},"created_at":"2026-06-08T01:05:31.367544+00:00","updated_at":"2026-06-08T01:05:31.367544+00:00"}