{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:DEBS22A5MFKECLCJT4VFLKIPGQ","short_pith_number":"pith:DEBS22A5","canonical_record":{"source":{"id":"2406.11256","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-17T06:47:03Z","cross_cats_sorted":[],"title_canon_sha256":"10ac2df278d4c256bd90d5eb3a8198243e98dfa173f15bb15dea842afe6330f9","abstract_canon_sha256":"ac2f4cd36845be6b1790b35146fd58ea30eb07f83d63d16d57f734968c471265"},"schema_version":"1.0"},"canonical_sha256":"19032d681d6154412c499f2a55a90f34027ec1eaf954dab549c3aec24b37b06f","source":{"kind":"arxiv","id":"2406.11256","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.11256","created_at":"2026-07-05T08:32:50Z"},{"alias_kind":"arxiv_version","alias_value":"2406.11256v1","created_at":"2026-07-05T08:32:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.11256","created_at":"2026-07-05T08:32:50Z"},{"alias_kind":"pith_short_12","alias_value":"DEBS22A5MFKE","created_at":"2026-07-05T08:32:50Z"},{"alias_kind":"pith_short_16","alias_value":"DEBS22A5MFKECLCJ","created_at":"2026-07-05T08:32:50Z"},{"alias_kind":"pith_short_8","alias_value":"DEBS22A5","created_at":"2026-07-05T08:32:50Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:DEBS22A5MFKECLCJT4VFLKIPGQ","target":"record","payload":{"canonical_record":{"source":{"id":"2406.11256","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-17T06:47:03Z","cross_cats_sorted":[],"title_canon_sha256":"10ac2df278d4c256bd90d5eb3a8198243e98dfa173f15bb15dea842afe6330f9","abstract_canon_sha256":"ac2f4cd36845be6b1790b35146fd58ea30eb07f83d63d16d57f734968c471265"},"schema_version":"1.0"},"canonical_sha256":"19032d681d6154412c499f2a55a90f34027ec1eaf954dab549c3aec24b37b06f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:32:50.944963Z","signature_b64":"MjngW560RDfRIa2WUiysNdMORBFsqca4FGKsL0wYfG1cVTNyeEIM+s34cGkFTfGsqJt8gKTfOj5iHjMmen4KAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"19032d681d6154412c499f2a55a90f34027ec1eaf954dab549c3aec24b37b06f","last_reissued_at":"2026-07-05T08:32:50.944473Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:32:50.944473Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.11256","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":"2026-07-05T08:32:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mI5/L0QekpHYyqz/0RVDtOQcIcxJGZj35HYJ/vKnMA8iOpLLUO7p7iRoOBIDLLZ5+d/YT9bw9sXzQiRbQyVoCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T01:48:43.438121Z"},"content_sha256":"2099d935dff6b10f46bd2b51f099728b3161ebd68b5ff85479d7456af6de6b73","schema_version":"1.0","event_id":"sha256:2099d935dff6b10f46bd2b51f099728b3161ebd68b5ff85479d7456af6de6b73"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:DEBS22A5MFKECLCJT4VFLKIPGQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Dynamic Data Mixing Maximizes Instruction Tuning for Mixture-of-Experts","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Daize Dong, Jiacheng Ruan, Tong Zhu, Wenliang Chen, Xiaoye Qu, Yu Cheng","submitted_at":"2024-06-17T06:47:03Z","abstract_excerpt":"Mixture-of-Experts (MoE) models have shown remarkable capability in instruction tuning, especially when the number of tasks scales. However, previous methods simply merge all training tasks (e.g. creative writing, coding, and mathematics) and apply fixed sampling weights, without considering the importance of different tasks as the model training state changes. In this way, the most helpful data cannot be effectively distinguished, leading to suboptimal model performance. To reduce the potential redundancies of datasets, we make the first attempt and propose a novel dynamic data mixture for Mo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.11256","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/2406.11256/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":"2026-07-05T08:32:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"q8UxxoAiYxW+Xln6FgU6cC2P0ZjVBEXTdAIw8DXZbyvOBdpXlq4t5sis9EablLUdVYH3alNpgEmbZdu6jTS9Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T01:48:43.438715Z"},"content_sha256":"68bdece22bb629c12d3bbdde17dad8d8089f1b7a475126f3ea2acf5d2bd9fa1d","schema_version":"1.0","event_id":"sha256:68bdece22bb629c12d3bbdde17dad8d8089f1b7a475126f3ea2acf5d2bd9fa1d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DEBS22A5MFKECLCJT4VFLKIPGQ/bundle.json","state_url":"https://pith.science/pith/DEBS22A5MFKECLCJT4VFLKIPGQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DEBS22A5MFKECLCJT4VFLKIPGQ/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":"2026-08-19T01:48:43Z","links":{"resolver":"https://pith.science/pith/DEBS22A5MFKECLCJT4VFLKIPGQ","bundle":"https://pith.science/pith/DEBS22A5MFKECLCJT4VFLKIPGQ/bundle.json","state":"https://pith.science/pith/DEBS22A5MFKECLCJT4VFLKIPGQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DEBS22A5MFKECLCJT4VFLKIPGQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:DEBS22A5MFKECLCJT4VFLKIPGQ","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":"ac2f4cd36845be6b1790b35146fd58ea30eb07f83d63d16d57f734968c471265","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-17T06:47:03Z","title_canon_sha256":"10ac2df278d4c256bd90d5eb3a8198243e98dfa173f15bb15dea842afe6330f9"},"schema_version":"1.0","source":{"id":"2406.11256","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.11256","created_at":"2026-07-05T08:32:50Z"},{"alias_kind":"arxiv_version","alias_value":"2406.11256v1","created_at":"2026-07-05T08:32:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.11256","created_at":"2026-07-05T08:32:50Z"},{"alias_kind":"pith_short_12","alias_value":"DEBS22A5MFKE","created_at":"2026-07-05T08:32:50Z"},{"alias_kind":"pith_short_16","alias_value":"DEBS22A5MFKECLCJ","created_at":"2026-07-05T08:32:50Z"},{"alias_kind":"pith_short_8","alias_value":"DEBS22A5","created_at":"2026-07-05T08:32:50Z"}],"graph_snapshots":[{"event_id":"sha256:68bdece22bb629c12d3bbdde17dad8d8089f1b7a475126f3ea2acf5d2bd9fa1d","target":"graph","created_at":"2026-07-05T08:32:50Z","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/2406.11256/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Mixture-of-Experts (MoE) models have shown remarkable capability in instruction tuning, especially when the number of tasks scales. However, previous methods simply merge all training tasks (e.g. creative writing, coding, and mathematics) and apply fixed sampling weights, without considering the importance of different tasks as the model training state changes. In this way, the most helpful data cannot be effectively distinguished, leading to suboptimal model performance. To reduce the potential redundancies of datasets, we make the first attempt and propose a novel dynamic data mixture for Mo","authors_text":"Daize Dong, Jiacheng Ruan, Tong Zhu, Wenliang Chen, Xiaoye Qu, Yu Cheng","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-17T06:47:03Z","title":"Dynamic Data Mixing Maximizes Instruction Tuning for Mixture-of-Experts"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.11256","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":"sha256:2099d935dff6b10f46bd2b51f099728b3161ebd68b5ff85479d7456af6de6b73","target":"record","created_at":"2026-07-05T08:32:50Z","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":"ac2f4cd36845be6b1790b35146fd58ea30eb07f83d63d16d57f734968c471265","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-17T06:47:03Z","title_canon_sha256":"10ac2df278d4c256bd90d5eb3a8198243e98dfa173f15bb15dea842afe6330f9"},"schema_version":"1.0","source":{"id":"2406.11256","kind":"arxiv","version":1}},"canonical_sha256":"19032d681d6154412c499f2a55a90f34027ec1eaf954dab549c3aec24b37b06f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"19032d681d6154412c499f2a55a90f34027ec1eaf954dab549c3aec24b37b06f","first_computed_at":"2026-07-05T08:32:50.944473Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:32:50.944473Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MjngW560RDfRIa2WUiysNdMORBFsqca4FGKsL0wYfG1cVTNyeEIM+s34cGkFTfGsqJt8gKTfOj5iHjMmen4KAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:32:50.944963Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.11256","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2099d935dff6b10f46bd2b51f099728b3161ebd68b5ff85479d7456af6de6b73","sha256:68bdece22bb629c12d3bbdde17dad8d8089f1b7a475126f3ea2acf5d2bd9fa1d"],"state_sha256":"fcc22ad34a02f12a78d8a999c7913120a9ad38f5509d7cd03350c620ede3a4c0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VBJx9TLqPNYFaYSsjigy5LF1w6Hu1UP9Nv3Ju1t8NwQqOpXB4Qx7VcPQ6BUAXKdsn+W5sdCVaQtB269fW7V7CQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T01:48:43.444456Z","bundle_sha256":"008c079f3ff7ce98af24b9cb9cd24fe7a2bebcb3487b4985fb30d1c59077835e"}}