{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:W6B2FNK63BKRVHPBEUMAL6FO6H","short_pith_number":"pith:W6B2FNK6","canonical_record":{"source":{"id":"2409.10571","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-14T11:39:13Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"376781bed60877a677d5da2298703d1747514cd83a197aec3c729f6b5574778a","abstract_canon_sha256":"936f7a572978d1693087c21259cdff20b9589fe3340416145bf0c54e58f5d554"},"schema_version":"1.0"},"canonical_sha256":"b783a2b55ed8551a9de1251805f8aef1ebea69685057b856df62e377e8a84af2","source":{"kind":"arxiv","id":"2409.10571","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.10571","created_at":"2026-07-05T09:07:52Z"},{"alias_kind":"arxiv_version","alias_value":"2409.10571v1","created_at":"2026-07-05T09:07:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.10571","created_at":"2026-07-05T09:07:52Z"},{"alias_kind":"pith_short_12","alias_value":"W6B2FNK63BKR","created_at":"2026-07-05T09:07:52Z"},{"alias_kind":"pith_short_16","alias_value":"W6B2FNK63BKRVHPB","created_at":"2026-07-05T09:07:52Z"},{"alias_kind":"pith_short_8","alias_value":"W6B2FNK6","created_at":"2026-07-05T09:07:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:W6B2FNK63BKRVHPBEUMAL6FO6H","target":"record","payload":{"canonical_record":{"source":{"id":"2409.10571","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-14T11:39:13Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"376781bed60877a677d5da2298703d1747514cd83a197aec3c729f6b5574778a","abstract_canon_sha256":"936f7a572978d1693087c21259cdff20b9589fe3340416145bf0c54e58f5d554"},"schema_version":"1.0"},"canonical_sha256":"b783a2b55ed8551a9de1251805f8aef1ebea69685057b856df62e377e8a84af2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:07:52.961730Z","signature_b64":"68wcexNfNccWzvvrIut/eRBZWbeceh6teGQb1IgL0JVOHpHbMX9ro+dFPEfquyFwkFBu90c/OKVlbWNUrsnjDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b783a2b55ed8551a9de1251805f8aef1ebea69685057b856df62e377e8a84af2","last_reissued_at":"2026-07-05T09:07:52.961303Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:07:52.961303Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2409.10571","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":"2026-07-05T09:07:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YkLq2pdWT/FQ6aSYomrLz6uAVKGliFO9fhnbcUsIjAUdnJHpnAyHgLQDDnTE2UT2oKeagZRwus8lUuG2GRwxAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T11:12:05.323518Z"},"content_sha256":"4e7d4ffd225cd845c3bf97110c18388cfe73e76262d6b24b32e5b81ad86fed5f","schema_version":"1.0","event_id":"sha256:4e7d4ffd225cd845c3bf97110c18388cfe73e76262d6b24b32e5b81ad86fed5f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:W6B2FNK63BKRVHPBEUMAL6FO6H","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ASFT: Aligned Supervised Fine-Tuning through Absolute Likelihood","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jiachen Sun, Quan Fang, Ruoyu Wang, Shaowei Hua","submitted_at":"2024-09-14T11:39:13Z","abstract_excerpt":"Direct Preference Optimization (DPO) is a method for enhancing model performance by directly optimizing for the preferences or rankings of outcomes, instead of traditional loss functions. This approach has proven effective in aligning Large Language Models (LLMs) with human preferences. Despite its widespread use across various tasks, DPO has been criticized for its sensitivity to the effectiveness of Supervised Fine-Tuning (SFT) and its limitations in enabling models to learn human-preferred responses, leading to less satisfactory performance. To address these limitations, we propose Aligned "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.10571","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/2409.10571/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":"2026-07-05T09:07:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PQvqVD+RYSLdEri3geib707lc7qytBX5Y3ei6vJsbz5PPWDAOFWkkH8H2jG1urcoS9blcFNA4AQiG4L6lhX5Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T11:12:05.324154Z"},"content_sha256":"8293606702ae1b18ff27354799b72e044e0ae64764aa1ffc0f87b7fcf5f0cfa9","schema_version":"1.0","event_id":"sha256:8293606702ae1b18ff27354799b72e044e0ae64764aa1ffc0f87b7fcf5f0cfa9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/W6B2FNK63BKRVHPBEUMAL6FO6H/bundle.json","state_url":"https://pith.science/pith/W6B2FNK63BKRVHPBEUMAL6FO6H/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/W6B2FNK63BKRVHPBEUMAL6FO6H/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":"2026-07-07T11:12:05Z","links":{"resolver":"https://pith.science/pith/W6B2FNK63BKRVHPBEUMAL6FO6H","bundle":"https://pith.science/pith/W6B2FNK63BKRVHPBEUMAL6FO6H/bundle.json","state":"https://pith.science/pith/W6B2FNK63BKRVHPBEUMAL6FO6H/state.json","well_known_bundle":"https://pith.science/.well-known/pith/W6B2FNK63BKRVHPBEUMAL6FO6H/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:W6B2FNK63BKRVHPBEUMAL6FO6H","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":"936f7a572978d1693087c21259cdff20b9589fe3340416145bf0c54e58f5d554","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-14T11:39:13Z","title_canon_sha256":"376781bed60877a677d5da2298703d1747514cd83a197aec3c729f6b5574778a"},"schema_version":"1.0","source":{"id":"2409.10571","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.10571","created_at":"2026-07-05T09:07:52Z"},{"alias_kind":"arxiv_version","alias_value":"2409.10571v1","created_at":"2026-07-05T09:07:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.10571","created_at":"2026-07-05T09:07:52Z"},{"alias_kind":"pith_short_12","alias_value":"W6B2FNK63BKR","created_at":"2026-07-05T09:07:52Z"},{"alias_kind":"pith_short_16","alias_value":"W6B2FNK63BKRVHPB","created_at":"2026-07-05T09:07:52Z"},{"alias_kind":"pith_short_8","alias_value":"W6B2FNK6","created_at":"2026-07-05T09:07:52Z"}],"graph_snapshots":[{"event_id":"sha256:8293606702ae1b18ff27354799b72e044e0ae64764aa1ffc0f87b7fcf5f0cfa9","target":"graph","created_at":"2026-07-05T09:07:52Z","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/2409.10571/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Direct Preference Optimization (DPO) is a method for enhancing model performance by directly optimizing for the preferences or rankings of outcomes, instead of traditional loss functions. This approach has proven effective in aligning Large Language Models (LLMs) with human preferences. Despite its widespread use across various tasks, DPO has been criticized for its sensitivity to the effectiveness of Supervised Fine-Tuning (SFT) and its limitations in enabling models to learn human-preferred responses, leading to less satisfactory performance. To address these limitations, we propose Aligned ","authors_text":"Jiachen Sun, Quan Fang, Ruoyu Wang, Shaowei Hua","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-14T11:39:13Z","title":"ASFT: Aligned Supervised Fine-Tuning through Absolute Likelihood"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.10571","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":"sha256:4e7d4ffd225cd845c3bf97110c18388cfe73e76262d6b24b32e5b81ad86fed5f","target":"record","created_at":"2026-07-05T09:07:52Z","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":"936f7a572978d1693087c21259cdff20b9589fe3340416145bf0c54e58f5d554","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-14T11:39:13Z","title_canon_sha256":"376781bed60877a677d5da2298703d1747514cd83a197aec3c729f6b5574778a"},"schema_version":"1.0","source":{"id":"2409.10571","kind":"arxiv","version":1}},"canonical_sha256":"b783a2b55ed8551a9de1251805f8aef1ebea69685057b856df62e377e8a84af2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b783a2b55ed8551a9de1251805f8aef1ebea69685057b856df62e377e8a84af2","first_computed_at":"2026-07-05T09:07:52.961303Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:07:52.961303Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"68wcexNfNccWzvvrIut/eRBZWbeceh6teGQb1IgL0JVOHpHbMX9ro+dFPEfquyFwkFBu90c/OKVlbWNUrsnjDw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:07:52.961730Z","signed_message":"canonical_sha256_bytes"},"source_id":"2409.10571","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4e7d4ffd225cd845c3bf97110c18388cfe73e76262d6b24b32e5b81ad86fed5f","sha256:8293606702ae1b18ff27354799b72e044e0ae64764aa1ffc0f87b7fcf5f0cfa9"],"state_sha256":"7b62ccb1f4160299d54b354373c51b68f790a17cb25f77b06d2c095c97b451e4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mpHKQMgcWOzD5zyYaMT+eDpf0S63aJWdNlUdQuXfH4s+DA/XIepe0mwdYNzaXg/bJJhn/+7Qe8JRPUx1xITfBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T11:12:05.327036Z","bundle_sha256":"c1890e5591d93d37987ac254012d0508bf0c674561f28d49efb1708fdc802523"}}