{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:SLVOXWTKHWNEIMMX676GQ5TOCV","short_pith_number":"pith:SLVOXWTK","canonical_record":{"source":{"id":"2505.14826","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-20T18:41:34Z","cross_cats_sorted":["cs.CL","stat.ML"],"title_canon_sha256":"715567791343f2dd23fc61abc8d6f91990e349569ea5e727e3c12025fc8b17f6","abstract_canon_sha256":"650dd89e8dac1ffcb37ea59ec66ab3be390a9ed74d935fe17a3f4d57303e96f8"},"schema_version":"1.0"},"canonical_sha256":"92eaebda6a3d9a443197f7fc68766e1551ec226ec8b1d5ca62e000486912dfcd","source":{"kind":"arxiv","id":"2505.14826","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.14826","created_at":"2026-07-05T11:06:36Z"},{"alias_kind":"arxiv_version","alias_value":"2505.14826v1","created_at":"2026-07-05T11:06:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14826","created_at":"2026-07-05T11:06:36Z"},{"alias_kind":"pith_short_12","alias_value":"SLVOXWTKHWNE","created_at":"2026-07-05T11:06:36Z"},{"alias_kind":"pith_short_16","alias_value":"SLVOXWTKHWNEIMMX","created_at":"2026-07-05T11:06:36Z"},{"alias_kind":"pith_short_8","alias_value":"SLVOXWTK","created_at":"2026-07-05T11:06:36Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:SLVOXWTKHWNEIMMX676GQ5TOCV","target":"record","payload":{"canonical_record":{"source":{"id":"2505.14826","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-20T18:41:34Z","cross_cats_sorted":["cs.CL","stat.ML"],"title_canon_sha256":"715567791343f2dd23fc61abc8d6f91990e349569ea5e727e3c12025fc8b17f6","abstract_canon_sha256":"650dd89e8dac1ffcb37ea59ec66ab3be390a9ed74d935fe17a3f4d57303e96f8"},"schema_version":"1.0"},"canonical_sha256":"92eaebda6a3d9a443197f7fc68766e1551ec226ec8b1d5ca62e000486912dfcd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:36.003076Z","signature_b64":"HCONzZll3IVniwbvksPLbzJIyI7aJKgacCPVySdvY8SShCPIge5aQAV6rZWBxArnwdfvQmxAQwV86yd8W9CBDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"92eaebda6a3d9a443197f7fc68766e1551ec226ec8b1d5ca62e000486912dfcd","last_reissued_at":"2026-07-05T11:06:36.002579Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:36.002579Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.14826","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":"2026-07-05T11:06:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lUmtYGu4n5uNrLIrpBmqhnntyY/U6HGai7FUhjSmfaXv77AKG0lp+G/yfEJa/D2+cEQrFHzz8DNC7RqwVcBoDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T18:43:25.298985Z"},"content_sha256":"f9bcd6707efc92195ed27ae8956c5bfe53d8d3ea8cbfee3b99bced4a0b824ac1","schema_version":"1.0","event_id":"sha256:f9bcd6707efc92195ed27ae8956c5bfe53d8d3ea8cbfee3b99bced4a0b824ac1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:SLVOXWTKHWNEIMMX676GQ5TOCV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"FisherSFT: Data-Efficient Supervised Fine-Tuning of Language Models Using Information Gain","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Branislav Kveton, Gaurush Hiranandani, Kiran Thekumparampil, Kousha Kalantari, Rohan Deb, Shoham Sabach","submitted_at":"2025-05-20T18:41:34Z","abstract_excerpt":"Supervised fine-tuning (SFT) is a standard approach to adapting large language models (LLMs) to new domains. In this work, we improve the statistical efficiency of SFT by selecting an informative subset of training examples. Specifically, for a fixed budget of training examples, which determines the computational cost of fine-tuning, we determine the most informative ones. The key idea in our method is to select examples that maximize information gain, measured by the Hessian of the log-likelihood of the LLM. We approximate it efficiently by linearizing the LLM at the last layer using multinom"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14826","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/2505.14826/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":"2026-07-05T11:06:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"E6pKi2iun/qbCiUfV7RGJbPWCWJ2bNfP7en3K5Oa4PHzNLezW3evviaHn4k6dGc/wN5w7QuCFgJB1zSYGWaqAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T18:43:25.299538Z"},"content_sha256":"7014c4798663ffe4e194cb99633474d88d4396996530c0e001034a6fcb21b88a","schema_version":"1.0","event_id":"sha256:7014c4798663ffe4e194cb99633474d88d4396996530c0e001034a6fcb21b88a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SLVOXWTKHWNEIMMX676GQ5TOCV/bundle.json","state_url":"https://pith.science/pith/SLVOXWTKHWNEIMMX676GQ5TOCV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SLVOXWTKHWNEIMMX676GQ5TOCV/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":"2026-08-08T18:43:25Z","links":{"resolver":"https://pith.science/pith/SLVOXWTKHWNEIMMX676GQ5TOCV","bundle":"https://pith.science/pith/SLVOXWTKHWNEIMMX676GQ5TOCV/bundle.json","state":"https://pith.science/pith/SLVOXWTKHWNEIMMX676GQ5TOCV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SLVOXWTKHWNEIMMX676GQ5TOCV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:SLVOXWTKHWNEIMMX676GQ5TOCV","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":"650dd89e8dac1ffcb37ea59ec66ab3be390a9ed74d935fe17a3f4d57303e96f8","cross_cats_sorted":["cs.CL","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-20T18:41:34Z","title_canon_sha256":"715567791343f2dd23fc61abc8d6f91990e349569ea5e727e3c12025fc8b17f6"},"schema_version":"1.0","source":{"id":"2505.14826","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.14826","created_at":"2026-07-05T11:06:36Z"},{"alias_kind":"arxiv_version","alias_value":"2505.14826v1","created_at":"2026-07-05T11:06:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14826","created_at":"2026-07-05T11:06:36Z"},{"alias_kind":"pith_short_12","alias_value":"SLVOXWTKHWNE","created_at":"2026-07-05T11:06:36Z"},{"alias_kind":"pith_short_16","alias_value":"SLVOXWTKHWNEIMMX","created_at":"2026-07-05T11:06:36Z"},{"alias_kind":"pith_short_8","alias_value":"SLVOXWTK","created_at":"2026-07-05T11:06:36Z"}],"graph_snapshots":[{"event_id":"sha256:7014c4798663ffe4e194cb99633474d88d4396996530c0e001034a6fcb21b88a","target":"graph","created_at":"2026-07-05T11:06:36Z","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/2505.14826/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Supervised fine-tuning (SFT) is a standard approach to adapting large language models (LLMs) to new domains. In this work, we improve the statistical efficiency of SFT by selecting an informative subset of training examples. Specifically, for a fixed budget of training examples, which determines the computational cost of fine-tuning, we determine the most informative ones. The key idea in our method is to select examples that maximize information gain, measured by the Hessian of the log-likelihood of the LLM. We approximate it efficiently by linearizing the LLM at the last layer using multinom","authors_text":"Branislav Kveton, Gaurush Hiranandani, Kiran Thekumparampil, Kousha Kalantari, Rohan Deb, Shoham Sabach","cross_cats":["cs.CL","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-20T18:41:34Z","title":"FisherSFT: Data-Efficient Supervised Fine-Tuning of Language Models Using Information Gain"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14826","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:f9bcd6707efc92195ed27ae8956c5bfe53d8d3ea8cbfee3b99bced4a0b824ac1","target":"record","created_at":"2026-07-05T11:06:36Z","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":"650dd89e8dac1ffcb37ea59ec66ab3be390a9ed74d935fe17a3f4d57303e96f8","cross_cats_sorted":["cs.CL","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-20T18:41:34Z","title_canon_sha256":"715567791343f2dd23fc61abc8d6f91990e349569ea5e727e3c12025fc8b17f6"},"schema_version":"1.0","source":{"id":"2505.14826","kind":"arxiv","version":1}},"canonical_sha256":"92eaebda6a3d9a443197f7fc68766e1551ec226ec8b1d5ca62e000486912dfcd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"92eaebda6a3d9a443197f7fc68766e1551ec226ec8b1d5ca62e000486912dfcd","first_computed_at":"2026-07-05T11:06:36.002579Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:06:36.002579Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HCONzZll3IVniwbvksPLbzJIyI7aJKgacCPVySdvY8SShCPIge5aQAV6rZWBxArnwdfvQmxAQwV86yd8W9CBDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:06:36.003076Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.14826","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f9bcd6707efc92195ed27ae8956c5bfe53d8d3ea8cbfee3b99bced4a0b824ac1","sha256:7014c4798663ffe4e194cb99633474d88d4396996530c0e001034a6fcb21b88a"],"state_sha256":"0b958c9ef486adf003b581c90a81f5f34245dffd02c47fdf0278a019ff6465d2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/dhfySKwqmXzHaOHT01Jk4+4ZFqJ3T+YEc32dS7PKcpF4/IkcT1Y98kp/FDC9f2+gyzotS4hOOZDY209+k7rDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T18:43:25.306357Z","bundle_sha256":"4c8fa4635eea669a35c7cdd7f0fb62f1fb79874583b934af7a3e4415c5d92398"}}