{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:QYQ3QVAK5IBKL5FZAAQZZV2O2V","short_pith_number":"pith:QYQ3QVAK","canonical_record":{"source":{"id":"2606.11204","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-04-22T12:12:25Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"800c6a22d4710706f7c33a89e9c6f31b61a5bdac765885dcc4f012f17aeef764","abstract_canon_sha256":"514e50839ad0a719e0a07c9a78ae29054d00bc223aaa61a1fbc28e80a6bdb93b"},"schema_version":"1.0"},"canonical_sha256":"8621b8540aea02a5f4b900219cd74ed56a19d86da0f30a54209124a67e07dbdc","source":{"kind":"arxiv","id":"2606.11204","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.11204","created_at":"2026-06-11T00:08:06Z"},{"alias_kind":"arxiv_version","alias_value":"2606.11204v1","created_at":"2026-06-11T00:08:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.11204","created_at":"2026-06-11T00:08:06Z"},{"alias_kind":"pith_short_12","alias_value":"QYQ3QVAK5IBK","created_at":"2026-06-11T00:08:06Z"},{"alias_kind":"pith_short_16","alias_value":"QYQ3QVAK5IBKL5FZ","created_at":"2026-06-11T00:08:06Z"},{"alias_kind":"pith_short_8","alias_value":"QYQ3QVAK","created_at":"2026-06-11T00:08:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:QYQ3QVAK5IBKL5FZAAQZZV2O2V","target":"record","payload":{"canonical_record":{"source":{"id":"2606.11204","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-04-22T12:12:25Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"800c6a22d4710706f7c33a89e9c6f31b61a5bdac765885dcc4f012f17aeef764","abstract_canon_sha256":"514e50839ad0a719e0a07c9a78ae29054d00bc223aaa61a1fbc28e80a6bdb93b"},"schema_version":"1.0"},"canonical_sha256":"8621b8540aea02a5f4b900219cd74ed56a19d86da0f30a54209124a67e07dbdc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-11T00:08:06.365411Z","signature_b64":"/Yl+IQhbs61Y3jTTGCjy+yzqiNbkW4nDii0Ehi91jaUIw8ibERipSdDY6ETzdcKMerPs5WMcJn1M3pJm06Q2CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8621b8540aea02a5f4b900219cd74ed56a19d86da0f30a54209124a67e07dbdc","last_reissued_at":"2026-06-11T00:08:06.364153Z","signature_status":"signed_v1","first_computed_at":"2026-06-11T00:08:06.364153Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.11204","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-11T00:08:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dH5+sfZ31+JGtwMVqMNT/zceI4K+VFdvaFq6cf5wu6REI94gX+vX+zeQgXo3VoogrDv8F73iRGElvvoUAL6TAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T10:43:22.119308Z"},"content_sha256":"ca914d8aba1db0fe2a590ffdf1bbc06494438ec24674cb8c9605b4173a26ed6e","schema_version":"1.0","event_id":"sha256:ca914d8aba1db0fe2a590ffdf1bbc06494438ec24674cb8c9605b4173a26ed6e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:QYQ3QVAK5IBKL5FZAAQZZV2O2V","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Benchmarking Large Language Models for Safety Data Extraction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Jonas Grill, S\\\"oren Berlinger, Thomas Bayer","submitted_at":"2026-04-22T12:12:25Z","abstract_excerpt":"Accurate extraction of structured information from Safety Data Sheets (SDS) remains challenging in industrial safety due to heterogeneous document formats and the limitations of traditional rule-based methods. This study benchmarks state-of-the-art Large Language Models (LLMs) for automated SDS data extraction, comparing text-based and multimodal processing pipelines. We systematically evaluate four models: Gemini 1.5 Pro, GPT-4o, Claude 3.7 Sonnet, and Llama 3.1-70B, across three prompting strategies: zero-shot, few-shot, and chain-of-thought. The evaluation framework assessed accuracy, laten"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.11204","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/2606.11204/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-11T00:08:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lqSeTrv/wAgA+JA4kOsZjvzhPX4XAlzjJywwAtLC0psJh2Abg6ipiDIzSL4ZXjbn0DRz7wZEC0IMzHJZ9olKCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T10:43:22.119697Z"},"content_sha256":"2550192a46289f732d9a538aaf68fbedb67d40c6dcf3ae4cd03f1e8defbe1c97","schema_version":"1.0","event_id":"sha256:2550192a46289f732d9a538aaf68fbedb67d40c6dcf3ae4cd03f1e8defbe1c97"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QYQ3QVAK5IBKL5FZAAQZZV2O2V/bundle.json","state_url":"https://pith.science/pith/QYQ3QVAK5IBKL5FZAAQZZV2O2V/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QYQ3QVAK5IBKL5FZAAQZZV2O2V/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":"2026-07-06T10:43:22Z","links":{"resolver":"https://pith.science/pith/QYQ3QVAK5IBKL5FZAAQZZV2O2V","bundle":"https://pith.science/pith/QYQ3QVAK5IBKL5FZAAQZZV2O2V/bundle.json","state":"https://pith.science/pith/QYQ3QVAK5IBKL5FZAAQZZV2O2V/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QYQ3QVAK5IBKL5FZAAQZZV2O2V/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:QYQ3QVAK5IBKL5FZAAQZZV2O2V","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":"514e50839ad0a719e0a07c9a78ae29054d00bc223aaa61a1fbc28e80a6bdb93b","cross_cats_sorted":["cs.IR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-04-22T12:12:25Z","title_canon_sha256":"800c6a22d4710706f7c33a89e9c6f31b61a5bdac765885dcc4f012f17aeef764"},"schema_version":"1.0","source":{"id":"2606.11204","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.11204","created_at":"2026-06-11T00:08:06Z"},{"alias_kind":"arxiv_version","alias_value":"2606.11204v1","created_at":"2026-06-11T00:08:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.11204","created_at":"2026-06-11T00:08:06Z"},{"alias_kind":"pith_short_12","alias_value":"QYQ3QVAK5IBK","created_at":"2026-06-11T00:08:06Z"},{"alias_kind":"pith_short_16","alias_value":"QYQ3QVAK5IBKL5FZ","created_at":"2026-06-11T00:08:06Z"},{"alias_kind":"pith_short_8","alias_value":"QYQ3QVAK","created_at":"2026-06-11T00:08:06Z"}],"graph_snapshots":[{"event_id":"sha256:2550192a46289f732d9a538aaf68fbedb67d40c6dcf3ae4cd03f1e8defbe1c97","target":"graph","created_at":"2026-06-11T00:08:06Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2606.11204/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Accurate extraction of structured information from Safety Data Sheets (SDS) remains challenging in industrial safety due to heterogeneous document formats and the limitations of traditional rule-based methods. This study benchmarks state-of-the-art Large Language Models (LLMs) for automated SDS data extraction, comparing text-based and multimodal processing pipelines. We systematically evaluate four models: Gemini 1.5 Pro, GPT-4o, Claude 3.7 Sonnet, and Llama 3.1-70B, across three prompting strategies: zero-shot, few-shot, and chain-of-thought. The evaluation framework assessed accuracy, laten","authors_text":"Jonas Grill, S\\\"oren Berlinger, Thomas Bayer","cross_cats":["cs.IR"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-04-22T12:12:25Z","title":"Benchmarking Large Language Models for Safety Data Extraction"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.11204","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ca914d8aba1db0fe2a590ffdf1bbc06494438ec24674cb8c9605b4173a26ed6e","target":"record","created_at":"2026-06-11T00:08:06Z","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":"514e50839ad0a719e0a07c9a78ae29054d00bc223aaa61a1fbc28e80a6bdb93b","cross_cats_sorted":["cs.IR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-04-22T12:12:25Z","title_canon_sha256":"800c6a22d4710706f7c33a89e9c6f31b61a5bdac765885dcc4f012f17aeef764"},"schema_version":"1.0","source":{"id":"2606.11204","kind":"arxiv","version":1}},"canonical_sha256":"8621b8540aea02a5f4b900219cd74ed56a19d86da0f30a54209124a67e07dbdc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8621b8540aea02a5f4b900219cd74ed56a19d86da0f30a54209124a67e07dbdc","first_computed_at":"2026-06-11T00:08:06.364153Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-11T00:08:06.364153Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/Yl+IQhbs61Y3jTTGCjy+yzqiNbkW4nDii0Ehi91jaUIw8ibERipSdDY6ETzdcKMerPs5WMcJn1M3pJm06Q2CA==","signature_status":"signed_v1","signed_at":"2026-06-11T00:08:06.365411Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.11204","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ca914d8aba1db0fe2a590ffdf1bbc06494438ec24674cb8c9605b4173a26ed6e","sha256:2550192a46289f732d9a538aaf68fbedb67d40c6dcf3ae4cd03f1e8defbe1c97"],"state_sha256":"ce8c9b810590ce2d49ddd43142ccebc565a374db7ddc357701dcd11aa2f9e2f2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"i3RTihogmPw8op5EqRaJKA6M/DfttANk2kLKqEtq5naOFoP1N75ZUIEHGHu1y0AtB68DT7nNslQ/3y461kB3CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T10:43:22.121642Z","bundle_sha256":"e477dba4474fbee88ff8a09bb931218ddcc282887676758f54481eb1b64a56ea"}}