{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DN3ZR3HKHEQ53TZO2B3X3B24XF","short_pith_number":"pith:DN3ZR3HK","schema_version":"1.0","canonical_sha256":"1b7798ecea3921ddcf2ed0777d875cb95351abf069d75a5d21ee855cdfb2a073","source":{"kind":"arxiv","id":"2605.04638","version":2},"attestation_state":"computed","paper":{"title":"Gradients with Respect to Semantics Preserving Embeddings Tell the Uncertainty of Large Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"Gradients with respect to semantics-preserving embeddings quantify uncertainty in LLM free-form generation.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Mingda Li, Rundong Lv, Ting Liu, Weinan Zhang, Xinyu Li","submitted_at":"2026-05-06T08:30:40Z","abstract_excerpt":"Uncertainty quantification (UQ) is an important technique for ensuring the trustworthiness of LLMs, given their tendency to hallucinate. Existing state-of-the-art UQ approaches for free-form generation rely heavily on sampling, which incurs high computational cost and variance. In this work, we propose the first gradient-based UQ method for free-form generation, SemGrad, which is sampling-free and computationally efficient. Unlike prior gradient-based methods developed for classification tasks that operates in parameter space, we propose to consider gradients in semantic space. Our method buil"},"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":"2605.04638","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-06T08:30:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"12cd684de46cdae552f55d885654333db37a55a8636da04b7d1454e861fc95db","abstract_canon_sha256":"2ada27bb035079668dcfbf065e13d15f9c1df373dee8c18b021806b2e0a9d7f6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:04:18.609275Z","signature_b64":"wU5h9TLYz/MvWlcowAlqqX6tMQBvaiS2Z1xj4RlPKfp314JqwY5IthJM0MTdFL/Muyk7CxXhiRHrgUwN/BZvBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1b7798ecea3921ddcf2ed0777d875cb95351abf069d75a5d21ee855cdfb2a073","last_reissued_at":"2026-06-02T02:04:18.608618Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:04:18.608618Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Gradients with Respect to Semantics Preserving Embeddings Tell the Uncertainty of Large Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"Gradients with respect to semantics-preserving embeddings quantify uncertainty in LLM free-form generation.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Mingda Li, Rundong Lv, Ting Liu, Weinan Zhang, Xinyu Li","submitted_at":"2026-05-06T08:30:40Z","abstract_excerpt":"Uncertainty quantification (UQ) is an important technique for ensuring the trustworthiness of LLMs, given their tendency to hallucinate. Existing state-of-the-art UQ approaches for free-form generation rely heavily on sampling, which incurs high computational cost and variance. In this work, we propose the first gradient-based UQ method for free-form generation, SemGrad, which is sampling-free and computationally efficient. Unlike prior gradient-based methods developed for classification tasks that operates in parameter space, we propose to consider gradients in semantic space. Our method buil"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We propose the first gradient-based UQ method for free-form generation, SemGrad, which is sampling-free and computationally efficient. ... Experiments demonstrate that both of our methods provide efficient and effective uncertainty estimates, achieving superior performance than state-of-the-art methods, particularly in settings with multiple valid responses.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"A confident LLM maintains stable output distributions under semantically equivalent input perturbations, and the Semantic Preservation Score reliably identifies the embeddings that best capture semantics for gradient computation.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SemGrad is a gradient-based uncertainty quantification technique for free-form LLM generation that operates in semantic space using a Semantic Preservation Score to select stable embeddings.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Gradients with respect to semantics-preserving embeddings quantify uncertainty in LLM free-form generation.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"b199756e70e5f6958ffc2cd25eea21c5a56bbe9be50f0ca9b102ff2cad61b665"},"source":{"id":"2605.04638","kind":"arxiv","version":2},"verdict":{"id":"1b622d72-ca05-42c5-821b-9dce87edf581","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T16:19:59.980650Z","strongest_claim":"We propose the first gradient-based UQ method for free-form generation, SemGrad, which is sampling-free and computationally efficient. ... Experiments demonstrate that both of our methods provide efficient and effective uncertainty estimates, achieving superior performance than state-of-the-art methods, particularly in settings with multiple valid responses.","one_line_summary":"SemGrad is a gradient-based uncertainty quantification technique for free-form LLM generation that operates in semantic space using a Semantic Preservation Score to select stable embeddings.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"A confident LLM maintains stable output distributions under semantically equivalent input perturbations, and the Semantic Preservation Score reliably identifies the embeddings that best capture semantics for gradient computation.","pith_extraction_headline":"Gradients with respect to semantics-preserving embeddings quantify uncertainty in LLM free-form generation."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.04638/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T11:36:28.579702Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T22:31:19.928091Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T14:17:33.483476Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"239a208e4ba727ad979e7ac9dd3bbe762f8df2bbb0b07db41e13bbe60715342d"},"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":"2605.04638","created_at":"2026-06-02T02:04:18.608717+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.04638v2","created_at":"2026-06-02T02:04:18.608717+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.04638","created_at":"2026-06-02T02:04:18.608717+00:00"},{"alias_kind":"pith_short_12","alias_value":"DN3ZR3HKHEQ5","created_at":"2026-06-02T02:04:18.608717+00:00"},{"alias_kind":"pith_short_16","alias_value":"DN3ZR3HKHEQ53TZO","created_at":"2026-06-02T02:04:18.608717+00:00"},{"alias_kind":"pith_short_8","alias_value":"DN3ZR3HK","created_at":"2026-06-02T02:04:18.608717+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/DN3ZR3HKHEQ53TZO2B3X3B24XF","json":"https://pith.science/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF.json","graph_json":"https://pith.science/api/pith-number/DN3ZR3HKHEQ53TZO2B3X3B24XF/graph.json","events_json":"https://pith.science/api/pith-number/DN3ZR3HKHEQ53TZO2B3X3B24XF/events.json","paper":"https://pith.science/paper/DN3ZR3HK"},"agent_actions":{"view_html":"https://pith.science/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF","download_json":"https://pith.science/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF.json","view_paper":"https://pith.science/paper/DN3ZR3HK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.04638&json=true","fetch_graph":"https://pith.science/api/pith-number/DN3ZR3HKHEQ53TZO2B3X3B24XF/graph.json","fetch_events":"https://pith.science/api/pith-number/DN3ZR3HKHEQ53TZO2B3X3B24XF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF/action/storage_attestation","attest_author":"https://pith.science/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF/action/author_attestation","sign_citation":"https://pith.science/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF/action/citation_signature","submit_replication":"https://pith.science/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF/action/replication_record"}},"created_at":"2026-06-02T02:04:18.608717+00:00","updated_at":"2026-06-02T02:04:18.608717+00:00"}