{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:GTUMGJIXNHOMTWXJOYMOHRQEYM","short_pith_number":"pith:GTUMGJIX","canonical_record":{"source":{"id":"2410.24175","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-31T17:42:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9a6114e2161189829cb32fbb4936407741f4b32a0ab0f4024945bd4276819339","abstract_canon_sha256":"39fce633b3c410a689c4488cb5bbe6aa7aa40317b5af37496130ac5e095b48c5"},"schema_version":"1.0"},"canonical_sha256":"34e8c3251769dcc9dae97618e3c604c3001d92263e4cdc7c3ffb4c9c308e77bc","source":{"kind":"arxiv","id":"2410.24175","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.24175","created_at":"2026-07-05T10:55:25Z"},{"alias_kind":"arxiv_version","alias_value":"2410.24175v2","created_at":"2026-07-05T10:55:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.24175","created_at":"2026-07-05T10:55:25Z"},{"alias_kind":"pith_short_12","alias_value":"GTUMGJIXNHOM","created_at":"2026-07-05T10:55:25Z"},{"alias_kind":"pith_short_16","alias_value":"GTUMGJIXNHOMTWXJ","created_at":"2026-07-05T10:55:25Z"},{"alias_kind":"pith_short_8","alias_value":"GTUMGJIX","created_at":"2026-07-05T10:55:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:GTUMGJIXNHOMTWXJOYMOHRQEYM","target":"record","payload":{"canonical_record":{"source":{"id":"2410.24175","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-31T17:42:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9a6114e2161189829cb32fbb4936407741f4b32a0ab0f4024945bd4276819339","abstract_canon_sha256":"39fce633b3c410a689c4488cb5bbe6aa7aa40317b5af37496130ac5e095b48c5"},"schema_version":"1.0"},"canonical_sha256":"34e8c3251769dcc9dae97618e3c604c3001d92263e4cdc7c3ffb4c9c308e77bc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:55:25.444277Z","signature_b64":"jNrVLmP7Y6Cs9RJDN8vYtOPQo6/bIcbIv6oBoB4OhxbTPqApHWdX5mB/NLvekdXGw56DgR5w9HgFeOo6aucTBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"34e8c3251769dcc9dae97618e3c604c3001d92263e4cdc7c3ffb4c9c308e77bc","last_reissued_at":"2026-07-05T10:55:25.443802Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:55:25.443802Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.24175","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":"2026-07-05T10:55:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fMuKvFnLfsC2AM10HEqnQ6fWu1YBbRu11L7GJi5Ilh5J2fIZ9K/y7daZnEcsKDBYKuw4GgKQKXUS1ot3752gDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T08:53:30.849396Z"},"content_sha256":"c20b09da0d35ee70d0079780880a4217a986649712602db1ea88adadaef181c3","schema_version":"1.0","event_id":"sha256:c20b09da0d35ee70d0079780880a4217a986649712602db1ea88adadaef181c3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:GTUMGJIXNHOMTWXJOYMOHRQEYM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Constraint Back-translation Improves Complex Instruction Following of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bin Xu, Hao Peng, Juanzi Li, Lei Hou, Xiaozhi Wang, Yunjia Qi","submitted_at":"2024-10-31T17:42:26Z","abstract_excerpt":"Large language models (LLMs) struggle to follow instructions with complex constraints in format, length, etc. Following the conventional instruction-tuning practice, previous works conduct post-training on complex instruction-response pairs generated by feeding complex instructions to advanced LLMs. However, even advanced LLMs cannot follow complex instructions well, thus limiting the quality of generated data. In this work, we find that existing datasets inherently contain implicit complex constraints and propose a novel data generation technique, constraint back-translation. Specifically, we"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.24175","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/2410.24175/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":"2026-07-05T10:55:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5nQtjGNVdgEgEOmgdm4mhAD9au5jaK9FYJdu/LdHqql+KtXsSPTV4wi3fwfe7LiV12CTKKLGI0SpF7S5PX1+AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T08:53:30.849769Z"},"content_sha256":"1c082877f098ccca825a5613d707bcc40163a594197003526cb65de79406f853","schema_version":"1.0","event_id":"sha256:1c082877f098ccca825a5613d707bcc40163a594197003526cb65de79406f853"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GTUMGJIXNHOMTWXJOYMOHRQEYM/bundle.json","state_url":"https://pith.science/pith/GTUMGJIXNHOMTWXJOYMOHRQEYM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GTUMGJIXNHOMTWXJOYMOHRQEYM/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":"2026-07-06T08:53:30Z","links":{"resolver":"https://pith.science/pith/GTUMGJIXNHOMTWXJOYMOHRQEYM","bundle":"https://pith.science/pith/GTUMGJIXNHOMTWXJOYMOHRQEYM/bundle.json","state":"https://pith.science/pith/GTUMGJIXNHOMTWXJOYMOHRQEYM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GTUMGJIXNHOMTWXJOYMOHRQEYM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GTUMGJIXNHOMTWXJOYMOHRQEYM","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":"39fce633b3c410a689c4488cb5bbe6aa7aa40317b5af37496130ac5e095b48c5","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-31T17:42:26Z","title_canon_sha256":"9a6114e2161189829cb32fbb4936407741f4b32a0ab0f4024945bd4276819339"},"schema_version":"1.0","source":{"id":"2410.24175","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.24175","created_at":"2026-07-05T10:55:25Z"},{"alias_kind":"arxiv_version","alias_value":"2410.24175v2","created_at":"2026-07-05T10:55:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.24175","created_at":"2026-07-05T10:55:25Z"},{"alias_kind":"pith_short_12","alias_value":"GTUMGJIXNHOM","created_at":"2026-07-05T10:55:25Z"},{"alias_kind":"pith_short_16","alias_value":"GTUMGJIXNHOMTWXJ","created_at":"2026-07-05T10:55:25Z"},{"alias_kind":"pith_short_8","alias_value":"GTUMGJIX","created_at":"2026-07-05T10:55:25Z"}],"graph_snapshots":[{"event_id":"sha256:1c082877f098ccca825a5613d707bcc40163a594197003526cb65de79406f853","target":"graph","created_at":"2026-07-05T10:55:25Z","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/2410.24175/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) struggle to follow instructions with complex constraints in format, length, etc. Following the conventional instruction-tuning practice, previous works conduct post-training on complex instruction-response pairs generated by feeding complex instructions to advanced LLMs. However, even advanced LLMs cannot follow complex instructions well, thus limiting the quality of generated data. In this work, we find that existing datasets inherently contain implicit complex constraints and propose a novel data generation technique, constraint back-translation. Specifically, we","authors_text":"Bin Xu, Hao Peng, Juanzi Li, Lei Hou, Xiaozhi Wang, Yunjia Qi","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-31T17:42:26Z","title":"Constraint Back-translation Improves Complex Instruction Following of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.24175","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:c20b09da0d35ee70d0079780880a4217a986649712602db1ea88adadaef181c3","target":"record","created_at":"2026-07-05T10:55:25Z","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":"39fce633b3c410a689c4488cb5bbe6aa7aa40317b5af37496130ac5e095b48c5","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-31T17:42:26Z","title_canon_sha256":"9a6114e2161189829cb32fbb4936407741f4b32a0ab0f4024945bd4276819339"},"schema_version":"1.0","source":{"id":"2410.24175","kind":"arxiv","version":2}},"canonical_sha256":"34e8c3251769dcc9dae97618e3c604c3001d92263e4cdc7c3ffb4c9c308e77bc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"34e8c3251769dcc9dae97618e3c604c3001d92263e4cdc7c3ffb4c9c308e77bc","first_computed_at":"2026-07-05T10:55:25.443802Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:55:25.443802Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"jNrVLmP7Y6Cs9RJDN8vYtOPQo6/bIcbIv6oBoB4OhxbTPqApHWdX5mB/NLvekdXGw56DgR5w9HgFeOo6aucTBw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:55:25.444277Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.24175","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c20b09da0d35ee70d0079780880a4217a986649712602db1ea88adadaef181c3","sha256:1c082877f098ccca825a5613d707bcc40163a594197003526cb65de79406f853"],"state_sha256":"58b607b8a3f1e64378e47e675a2dce4ffb4a0cd8ee2b6d887cee2f7b3ea0f6e2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"a1nUVwbRcw6QnebUrR1qQUT3cyxQSCn6OEQWGB1TFk9ER5ifxccsTsfNymY0nYFdv9dhCj+cZT0TLc08reFyAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T08:53:30.851697Z","bundle_sha256":"c4da6e2374578f8f11eb8ed3d1e0a65cd41042b16f23401eb49197043d3906a1"}}