{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:DN3ZR3HKHEQ53TZO2B3X3B24XF","short_pith_number":"pith:DN3ZR3HK","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"},"canonical_sha256":"1b7798ecea3921ddcf2ed0777d875cb95351abf069d75a5d21ee855cdfb2a073","source":{"kind":"arxiv","id":"2605.04638","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.04638","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"arxiv_version","alias_value":"2605.04638v2","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.04638","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"pith_short_12","alias_value":"DN3ZR3HKHEQ5","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"pith_short_16","alias_value":"DN3ZR3HKHEQ53TZO","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"pith_short_8","alias_value":"DN3ZR3HK","created_at":"2026-06-02T02:04:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:DN3ZR3HKHEQ53TZO2B3X3B24XF","target":"record","payload":{"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"},"canonical_sha256":"1b7798ecea3921ddcf2ed0777d875cb95351abf069d75a5d21ee855cdfb2a073","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"},"source_kind":"arxiv","source_id":"2605.04638","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T02:04:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"r3YNamIE1UvvWRBGWlFYlAL6BCEpg0zITgLuij3lIEGSeDCYtOgjIkC1kGS0skmXdV/i/B39tiGfyYXMdnLuCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T14:24:26.206296Z"},"content_sha256":"eee46cef8c75198430ef4d50240c2bb7c731d6078396bab82b4a94e4284e5816","schema_version":"1.0","event_id":"sha256:eee46cef8c75198430ef4d50240c2bb7c731d6078396bab82b4a94e4284e5816"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:DN3ZR3HKHEQ53TZO2B3X3B24XF","target":"graph","payload":{"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"},"verdict_id":"1b622d72-ca05-42c5-821b-9dce87edf581"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T02:04:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PCaO87ZezIJR0UynM6YWHdvcGgRcNtz4giJGtVBx+fjPEXfq2jQ2GUy+ghFF2vroEPOIV7WBU4tR8sp3mLdXAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T14:24:26.206795Z"},"content_sha256":"8883a541b0c067df36c94205abb33e3c130acba7b7c864b8beb0d505d4fafa9b","schema_version":"1.0","event_id":"sha256:8883a541b0c067df36c94205abb33e3c130acba7b7c864b8beb0d505d4fafa9b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF/bundle.json","state_url":"https://pith.science/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-04T14:24:26Z","links":{"resolver":"https://pith.science/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF","bundle":"https://pith.science/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF/bundle.json","state":"https://pith.science/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DN3ZR3HKHEQ53TZO2B3X3B24XF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:DN3ZR3HKHEQ53TZO2B3X3B24XF","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"2ada27bb035079668dcfbf065e13d15f9c1df373dee8c18b021806b2e0a9d7f6","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-06T08:30:40Z","title_canon_sha256":"12cd684de46cdae552f55d885654333db37a55a8636da04b7d1454e861fc95db"},"schema_version":"1.0","source":{"id":"2605.04638","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.04638","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"arxiv_version","alias_value":"2605.04638v2","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.04638","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"pith_short_12","alias_value":"DN3ZR3HKHEQ5","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"pith_short_16","alias_value":"DN3ZR3HKHEQ53TZO","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"pith_short_8","alias_value":"DN3ZR3HK","created_at":"2026-06-02T02:04:18Z"}],"graph_snapshots":[{"event_id":"sha256:8883a541b0c067df36c94205abb33e3c130acba7b7c864b8beb0d505d4fafa9b","target":"graph","created_at":"2026-06-02T02:04:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Gradients with respect to semantics-preserving embeddings quantify uncertainty in LLM free-form generation."}],"snapshot_sha256":"b199756e70e5f6958ffc2cd25eea21c5a56bbe9be50f0ca9b102ff2cad61b665"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"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"}],"endpoint":"/pith/2605.04638/integrity.json","findings":[],"snapshot_sha256":"239a208e4ba727ad979e7ac9dd3bbe762f8df2bbb0b07db41e13bbe60715342d","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Mingda Li, Rundong Lv, Ting Liu, Weinan Zhang, Xinyu Li","cross_cats":["cs.AI"],"headline":"Gradients with respect to semantics-preserving embeddings quantify uncertainty in LLM free-form generation.","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-06T08:30:40Z","title":"Gradients with Respect to Semantics Preserving Embeddings Tell the Uncertainty of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.04638","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-08T16:19:59.980650Z","id":"1b622d72-ca05-42c5-821b-9dce87edf581","model_set":{"reader":"grok-4.3"},"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","pith_extraction_headline":"Gradients with respect to semantics-preserving embeddings quantify uncertainty in LLM free-form generation.","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.","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."}},"verdict_id":"1b622d72-ca05-42c5-821b-9dce87edf581"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:eee46cef8c75198430ef4d50240c2bb7c731d6078396bab82b4a94e4284e5816","target":"record","created_at":"2026-06-02T02:04:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"2ada27bb035079668dcfbf065e13d15f9c1df373dee8c18b021806b2e0a9d7f6","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-06T08:30:40Z","title_canon_sha256":"12cd684de46cdae552f55d885654333db37a55a8636da04b7d1454e861fc95db"},"schema_version":"1.0","source":{"id":"2605.04638","kind":"arxiv","version":2}},"canonical_sha256":"1b7798ecea3921ddcf2ed0777d875cb95351abf069d75a5d21ee855cdfb2a073","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1b7798ecea3921ddcf2ed0777d875cb95351abf069d75a5d21ee855cdfb2a073","first_computed_at":"2026-06-02T02:04:18.608618Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T02:04:18.608618Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wU5h9TLYz/MvWlcowAlqqX6tMQBvaiS2Z1xj4RlPKfp314JqwY5IthJM0MTdFL/Muyk7CxXhiRHrgUwN/BZvBA==","signature_status":"signed_v1","signed_at":"2026-06-02T02:04:18.609275Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.04638","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:eee46cef8c75198430ef4d50240c2bb7c731d6078396bab82b4a94e4284e5816","sha256:8883a541b0c067df36c94205abb33e3c130acba7b7c864b8beb0d505d4fafa9b"],"state_sha256":"b3136f37d51a5927ecb8293e392de141be6f6a984e3b988735aaf4b421976225"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zxgf6/twbwEwMQZ3n8iYsV6SPSp0xLw4eQPRvmXsw1l+Ctto8agtXXsx4IB9dV3vHhDpTG81uRtX6zQBFeltCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-04T14:24:26.209045Z","bundle_sha256":"6773b584d7a107c693bcdd7cba1cc959f061657faa7c58c9486c8a8571fef2bb"}}