{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:HVWT2AGUB6OLPX45H75AX3W526","short_pith_number":"pith:HVWT2AGU","canonical_record":{"source":{"id":"2504.12687","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-17T06:29:28Z","cross_cats_sorted":[],"title_canon_sha256":"d8375f47dea016117bdf21589f8097376ce6e400c3b8117a2b402d0caba89352","abstract_canon_sha256":"ce67d1a45b15a860e14c7cb4dfc7f426f883d41bbb25bf9072779c8ab1558c90"},"schema_version":"1.0"},"canonical_sha256":"3d6d3d00d40f9cb7df9d3ffa0beeddd7a2e424b2307e18fe042da32c448b326a","source":{"kind":"arxiv","id":"2504.12687","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.12687","created_at":"2026-07-05T10:50:20Z"},{"alias_kind":"arxiv_version","alias_value":"2504.12687v1","created_at":"2026-07-05T10:50:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.12687","created_at":"2026-07-05T10:50:20Z"},{"alias_kind":"pith_short_12","alias_value":"HVWT2AGUB6OL","created_at":"2026-07-05T10:50:20Z"},{"alias_kind":"pith_short_16","alias_value":"HVWT2AGUB6OLPX45","created_at":"2026-07-05T10:50:20Z"},{"alias_kind":"pith_short_8","alias_value":"HVWT2AGU","created_at":"2026-07-05T10:50:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:HVWT2AGUB6OLPX45H75AX3W526","target":"record","payload":{"canonical_record":{"source":{"id":"2504.12687","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-17T06:29:28Z","cross_cats_sorted":[],"title_canon_sha256":"d8375f47dea016117bdf21589f8097376ce6e400c3b8117a2b402d0caba89352","abstract_canon_sha256":"ce67d1a45b15a860e14c7cb4dfc7f426f883d41bbb25bf9072779c8ab1558c90"},"schema_version":"1.0"},"canonical_sha256":"3d6d3d00d40f9cb7df9d3ffa0beeddd7a2e424b2307e18fe042da32c448b326a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:50:20.632411Z","signature_b64":"uEKDM6LSf1ocQhBr/BS5KmQrrcKu3uKzKTOlYvkDUBn1MsqQ/4ftFEXtUdUzbKmlHmqpj2vSKZkM9TPuVoMfCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3d6d3d00d40f9cb7df9d3ffa0beeddd7a2e424b2307e18fe042da32c448b326a","last_reissued_at":"2026-07-05T10:50:20.631888Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:50:20.631888Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2504.12687","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":"2026-07-05T10:50:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"F0U/N8uf6LfbbI75KUOwtYvy84TOYTtGJzfpDkuDZA8t2kTU3LoTlJpLG5zw19pb8FOTLu8tCxQiHVotZBF0Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T10:42:09.140851Z"},"content_sha256":"b3ebfe5e4f709a01f867eacc6dbff9aafd1abe113582f1b12326ed59566b0b43","schema_version":"1.0","event_id":"sha256:b3ebfe5e4f709a01f867eacc6dbff9aafd1abe113582f1b12326ed59566b0b43"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:HVWT2AGUB6OLPX45H75AX3W526","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Data-efficient LLM Fine-tuning for Code Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Sheng-Jun Huang, Weijie Lv, Xuan Xia","submitted_at":"2025-04-17T06:29:28Z","abstract_excerpt":"Large language models (LLMs) have demonstrated significant potential in code generation tasks. However, there remains a performance gap between open-source and closed-source models. To address this gap, existing approaches typically generate large amounts of synthetic data for fine-tuning, which often leads to inefficient training. In this work, we propose a data selection strategy in order to improve the effectiveness and efficiency of training for code-based LLMs. By prioritizing data complexity and ensuring that the sampled subset aligns with the distribution of the original dataset, our sa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.12687","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/2504.12687/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":"2026-07-05T10:50:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SBn44nJHTYUiIZbwMEKfsW3sul/xUq7O1wMh+0nCoHJKXC2IInEYKYft6boYmC9VATU5T7MuDSFfvPpREx+WBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T10:42:09.141354Z"},"content_sha256":"e9b749335d7b9527d990158792db5d01e8d011610d5e38461449393a38467789","schema_version":"1.0","event_id":"sha256:e9b749335d7b9527d990158792db5d01e8d011610d5e38461449393a38467789"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HVWT2AGUB6OLPX45H75AX3W526/bundle.json","state_url":"https://pith.science/pith/HVWT2AGUB6OLPX45H75AX3W526/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HVWT2AGUB6OLPX45H75AX3W526/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":"2026-08-08T10:42:09Z","links":{"resolver":"https://pith.science/pith/HVWT2AGUB6OLPX45H75AX3W526","bundle":"https://pith.science/pith/HVWT2AGUB6OLPX45H75AX3W526/bundle.json","state":"https://pith.science/pith/HVWT2AGUB6OLPX45H75AX3W526/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HVWT2AGUB6OLPX45H75AX3W526/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:HVWT2AGUB6OLPX45H75AX3W526","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":"ce67d1a45b15a860e14c7cb4dfc7f426f883d41bbb25bf9072779c8ab1558c90","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-17T06:29:28Z","title_canon_sha256":"d8375f47dea016117bdf21589f8097376ce6e400c3b8117a2b402d0caba89352"},"schema_version":"1.0","source":{"id":"2504.12687","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.12687","created_at":"2026-07-05T10:50:20Z"},{"alias_kind":"arxiv_version","alias_value":"2504.12687v1","created_at":"2026-07-05T10:50:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.12687","created_at":"2026-07-05T10:50:20Z"},{"alias_kind":"pith_short_12","alias_value":"HVWT2AGUB6OL","created_at":"2026-07-05T10:50:20Z"},{"alias_kind":"pith_short_16","alias_value":"HVWT2AGUB6OLPX45","created_at":"2026-07-05T10:50:20Z"},{"alias_kind":"pith_short_8","alias_value":"HVWT2AGU","created_at":"2026-07-05T10:50:20Z"}],"graph_snapshots":[{"event_id":"sha256:e9b749335d7b9527d990158792db5d01e8d011610d5e38461449393a38467789","target":"graph","created_at":"2026-07-05T10:50:20Z","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/2504.12687/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have demonstrated significant potential in code generation tasks. However, there remains a performance gap between open-source and closed-source models. To address this gap, existing approaches typically generate large amounts of synthetic data for fine-tuning, which often leads to inefficient training. In this work, we propose a data selection strategy in order to improve the effectiveness and efficiency of training for code-based LLMs. By prioritizing data complexity and ensuring that the sampled subset aligns with the distribution of the original dataset, our sa","authors_text":"Sheng-Jun Huang, Weijie Lv, Xuan Xia","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-17T06:29:28Z","title":"Data-efficient LLM Fine-tuning for Code Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.12687","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:b3ebfe5e4f709a01f867eacc6dbff9aafd1abe113582f1b12326ed59566b0b43","target":"record","created_at":"2026-07-05T10:50:20Z","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":"ce67d1a45b15a860e14c7cb4dfc7f426f883d41bbb25bf9072779c8ab1558c90","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-17T06:29:28Z","title_canon_sha256":"d8375f47dea016117bdf21589f8097376ce6e400c3b8117a2b402d0caba89352"},"schema_version":"1.0","source":{"id":"2504.12687","kind":"arxiv","version":1}},"canonical_sha256":"3d6d3d00d40f9cb7df9d3ffa0beeddd7a2e424b2307e18fe042da32c448b326a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3d6d3d00d40f9cb7df9d3ffa0beeddd7a2e424b2307e18fe042da32c448b326a","first_computed_at":"2026-07-05T10:50:20.631888Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:50:20.631888Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uEKDM6LSf1ocQhBr/BS5KmQrrcKu3uKzKTOlYvkDUBn1MsqQ/4ftFEXtUdUzbKmlHmqpj2vSKZkM9TPuVoMfCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:50:20.632411Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.12687","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b3ebfe5e4f709a01f867eacc6dbff9aafd1abe113582f1b12326ed59566b0b43","sha256:e9b749335d7b9527d990158792db5d01e8d011610d5e38461449393a38467789"],"state_sha256":"6ea03f2e78d40e7d0072d108cf2e0e2752965fb3868ba6d46470e816b3551517"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3fzDUo0oXKogoNu7cXi/zzICmYqKt89uCHqJYXqVAXeaEFTihkR5lJd7/+qJt0gIHk7d/D7Kr50OX1oY7e7NBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T10:42:09.146416Z","bundle_sha256":"b3bc632c5f3ecd4461c907f089690c9936a31aca80797f80c6768fd4193294b6"}}