{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:AXDXKVFUOYVYXCBJPGHMTCOVOJ","short_pith_number":"pith:AXDXKVFU","canonical_record":{"source":{"id":"2503.15573","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-19T11:35:57Z","cross_cats_sorted":[],"title_canon_sha256":"c1250f7d75dd767c5e0d200faf4a8822641af960dfd68a6bc71b85e33c00b22e","abstract_canon_sha256":"cade1e052ba43c4be9929b7012bf6645b27e357edbc6a2fe11e360634918c74a"},"schema_version":"1.0"},"canonical_sha256":"05c77554b4762b8b8829798ec989d5725a20314a4ca8a79e0bc67cf0ce93e0cd","source":{"kind":"arxiv","id":"2503.15573","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.15573","created_at":"2026-07-05T11:03:58Z"},{"alias_kind":"arxiv_version","alias_value":"2503.15573v2","created_at":"2026-07-05T11:03:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.15573","created_at":"2026-07-05T11:03:58Z"},{"alias_kind":"pith_short_12","alias_value":"AXDXKVFUOYVY","created_at":"2026-07-05T11:03:58Z"},{"alias_kind":"pith_short_16","alias_value":"AXDXKVFUOYVYXCBJ","created_at":"2026-07-05T11:03:58Z"},{"alias_kind":"pith_short_8","alias_value":"AXDXKVFU","created_at":"2026-07-05T11:03:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:AXDXKVFUOYVYXCBJPGHMTCOVOJ","target":"record","payload":{"canonical_record":{"source":{"id":"2503.15573","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-19T11:35:57Z","cross_cats_sorted":[],"title_canon_sha256":"c1250f7d75dd767c5e0d200faf4a8822641af960dfd68a6bc71b85e33c00b22e","abstract_canon_sha256":"cade1e052ba43c4be9929b7012bf6645b27e357edbc6a2fe11e360634918c74a"},"schema_version":"1.0"},"canonical_sha256":"05c77554b4762b8b8829798ec989d5725a20314a4ca8a79e0bc67cf0ce93e0cd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:03:58.339744Z","signature_b64":"QkMkLadRzRQvvGUevHnq3KQOge5nQaChlSiCR2oSp6eNR0BiXtQrU0C+bRkvs0YerFwkhn817LydJ9RDopI/Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"05c77554b4762b8b8829798ec989d5725a20314a4ca8a79e0bc67cf0ce93e0cd","last_reissued_at":"2026-07-05T11:03:58.339190Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:03:58.339190Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.15573","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":"2026-07-05T11:03:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YGU9K6J/veuTqwHSKUgox3hwZ3Ji9q6Ukfe2Rw9hXMv3Lnkm9dYiBUKobqAVEzvOOfTpN+TCzPzfk4z65FoxDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T22:10:56.450128Z"},"content_sha256":"8bea8b6b9806016a8f2116a4500922bf661df36ec72c819cdd8f48f907e86d8b","schema_version":"1.0","event_id":"sha256:8bea8b6b9806016a8f2116a4500922bf661df36ec72c819cdd8f48f907e86d8b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:AXDXKVFUOYVYXCBJPGHMTCOVOJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Task-Specific Data Selection for Instruction Tuning via Monosemantic Neuronal Activations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Da Ma, Gonghu Shang, Kai Yu, Lei Pan, Libo Qin, Lu Chen, Shuai Fan, Yijie Luo, Zhi Chen","submitted_at":"2025-03-19T11:35:57Z","abstract_excerpt":"Instruction tuning improves the ability of large language models (LLMs) to follow diverse human instructions, but achieving strong performance on specific target tasks remains challenging. A critical bottleneck is selecting the most relevant data to maximize task-specific performance. Existing data selection approaches include unstable influence-based methods and more stable distribution alignment methods, the latter of which critically rely on the underlying sample representation. In practice, most distribution alignment methods, from shallow features (e.g., BM25) to neural embeddings (e.g., "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.15573","kind":"arxiv","version":2},"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/2503.15573/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":"2026-07-05T11:03:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"t5qhAX5VOMkpMdJZdySCDrluxxz9GjZs+KK8s27KI+xbDpZOYS/W10c6rmfS6p2SDxHg7DsgR4KjCHXe/ECOCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T22:10:56.450630Z"},"content_sha256":"c08be139081a6a0c488ed366df7f4f3e626433b23295f29ab1c0ec855d97e447","schema_version":"1.0","event_id":"sha256:c08be139081a6a0c488ed366df7f4f3e626433b23295f29ab1c0ec855d97e447"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AXDXKVFUOYVYXCBJPGHMTCOVOJ/bundle.json","state_url":"https://pith.science/pith/AXDXKVFUOYVYXCBJPGHMTCOVOJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AXDXKVFUOYVYXCBJPGHMTCOVOJ/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":"2026-08-07T22:10:56Z","links":{"resolver":"https://pith.science/pith/AXDXKVFUOYVYXCBJPGHMTCOVOJ","bundle":"https://pith.science/pith/AXDXKVFUOYVYXCBJPGHMTCOVOJ/bundle.json","state":"https://pith.science/pith/AXDXKVFUOYVYXCBJPGHMTCOVOJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AXDXKVFUOYVYXCBJPGHMTCOVOJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:AXDXKVFUOYVYXCBJPGHMTCOVOJ","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":"cade1e052ba43c4be9929b7012bf6645b27e357edbc6a2fe11e360634918c74a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-19T11:35:57Z","title_canon_sha256":"c1250f7d75dd767c5e0d200faf4a8822641af960dfd68a6bc71b85e33c00b22e"},"schema_version":"1.0","source":{"id":"2503.15573","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.15573","created_at":"2026-07-05T11:03:58Z"},{"alias_kind":"arxiv_version","alias_value":"2503.15573v2","created_at":"2026-07-05T11:03:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.15573","created_at":"2026-07-05T11:03:58Z"},{"alias_kind":"pith_short_12","alias_value":"AXDXKVFUOYVY","created_at":"2026-07-05T11:03:58Z"},{"alias_kind":"pith_short_16","alias_value":"AXDXKVFUOYVYXCBJ","created_at":"2026-07-05T11:03:58Z"},{"alias_kind":"pith_short_8","alias_value":"AXDXKVFU","created_at":"2026-07-05T11:03:58Z"}],"graph_snapshots":[{"event_id":"sha256:c08be139081a6a0c488ed366df7f4f3e626433b23295f29ab1c0ec855d97e447","target":"graph","created_at":"2026-07-05T11:03:58Z","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/2503.15573/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Instruction tuning improves the ability of large language models (LLMs) to follow diverse human instructions, but achieving strong performance on specific target tasks remains challenging. A critical bottleneck is selecting the most relevant data to maximize task-specific performance. Existing data selection approaches include unstable influence-based methods and more stable distribution alignment methods, the latter of which critically rely on the underlying sample representation. In practice, most distribution alignment methods, from shallow features (e.g., BM25) to neural embeddings (e.g., ","authors_text":"Da Ma, Gonghu Shang, Kai Yu, Lei Pan, Libo Qin, Lu Chen, Shuai Fan, Yijie Luo, Zhi Chen","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-19T11:35:57Z","title":"Task-Specific Data Selection for Instruction Tuning via Monosemantic Neuronal Activations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.15573","kind":"arxiv","version":2},"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":"sha256:8bea8b6b9806016a8f2116a4500922bf661df36ec72c819cdd8f48f907e86d8b","target":"record","created_at":"2026-07-05T11:03:58Z","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":"cade1e052ba43c4be9929b7012bf6645b27e357edbc6a2fe11e360634918c74a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-19T11:35:57Z","title_canon_sha256":"c1250f7d75dd767c5e0d200faf4a8822641af960dfd68a6bc71b85e33c00b22e"},"schema_version":"1.0","source":{"id":"2503.15573","kind":"arxiv","version":2}},"canonical_sha256":"05c77554b4762b8b8829798ec989d5725a20314a4ca8a79e0bc67cf0ce93e0cd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"05c77554b4762b8b8829798ec989d5725a20314a4ca8a79e0bc67cf0ce93e0cd","first_computed_at":"2026-07-05T11:03:58.339190Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:03:58.339190Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"QkMkLadRzRQvvGUevHnq3KQOge5nQaChlSiCR2oSp6eNR0BiXtQrU0C+bRkvs0YerFwkhn817LydJ9RDopI/Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:03:58.339744Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.15573","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8bea8b6b9806016a8f2116a4500922bf661df36ec72c819cdd8f48f907e86d8b","sha256:c08be139081a6a0c488ed366df7f4f3e626433b23295f29ab1c0ec855d97e447"],"state_sha256":"09588e65f9e0896fff90dcfe0ea69eeef8da6ae65aeb41b9e8b45c21b3c170f0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1vQHeB2OhbFW7O4yit0kci7B5D5S96NbfjQZ1Q+oZzCbzspFcKZnSPPdRTH20Le9QlZk2m4OKRFfzj7KZrvvAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T22:10:56.455850Z","bundle_sha256":"1a8d13b91ec84b6b6b67e8b3bcede0b73395fa5710b8265c77919a94052aee72"}}