{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:GG7FYD47ZKO3LFCQZROI4COYS7","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":"a82ffd82a0536972002cb9ca9654977ea4614509a446ff2d2f0b67c3e5a55c3d","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-17T21:12:57Z","title_canon_sha256":"a3fb1db05d509012c129c654c927052477246c39695b5ccb7ea44c1f37dad723"},"schema_version":"1.0","source":{"id":"2502.12325","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.12325","created_at":"2026-07-05T10:15:53Z"},{"alias_kind":"arxiv_version","alias_value":"2502.12325v1","created_at":"2026-07-05T10:15:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.12325","created_at":"2026-07-05T10:15:53Z"},{"alias_kind":"pith_short_12","alias_value":"GG7FYD47ZKO3","created_at":"2026-07-05T10:15:53Z"},{"alias_kind":"pith_short_16","alias_value":"GG7FYD47ZKO3LFCQ","created_at":"2026-07-05T10:15:53Z"},{"alias_kind":"pith_short_8","alias_value":"GG7FYD47","created_at":"2026-07-05T10:15:53Z"}],"graph_snapshots":[{"event_id":"sha256:36d1143da447bc7b4cbb39ca0acf31073abd5e1a553b96b8bc2313ca47be459b","target":"graph","created_at":"2026-07-05T10:15:53Z","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/2502.12325/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Training large language models (LLMs) for different inference constraints is computationally expensive, limiting control over efficiency-accuracy trade-offs. Moreover, once trained, these models typically process tokens uniformly, regardless of their complexity, leading to static and inflexible behavior. In this paper, we introduce a post-training optimization framework, DynaMoE, that adapts a pre-trained dense LLM to a token-difficulty-driven Mixture-of-Experts model with minimal fine-tuning cost. This adaptation makes the model dynamic, with sensitivity control to customize the balance betwe","authors_text":"Devang Naik, Kumari Nishu, Mahyar Najibi, Maxwell Horton, Mehrdad Farajtabar, Minsik Cho, Moin Nabi, Sachin Mehta, Samira Abnar","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-17T21:12:57Z","title":"From Dense to Dynamic: Token-Difficulty Driven MoEfication of Pre-Trained LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.12325","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":"sha256:4e5858d79b565d3e166d5137bc739c1d87afcf432efd3e1a7b87e0e097da52d6","target":"record","created_at":"2026-07-05T10:15:53Z","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":"a82ffd82a0536972002cb9ca9654977ea4614509a446ff2d2f0b67c3e5a55c3d","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-17T21:12:57Z","title_canon_sha256":"a3fb1db05d509012c129c654c927052477246c39695b5ccb7ea44c1f37dad723"},"schema_version":"1.0","source":{"id":"2502.12325","kind":"arxiv","version":1}},"canonical_sha256":"31be5c0f9fca9db59450cc5c8e09d897fd5377328a615ca058e5d51135ad4441","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"31be5c0f9fca9db59450cc5c8e09d897fd5377328a615ca058e5d51135ad4441","first_computed_at":"2026-07-05T10:15:53.639034Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:15:53.639034Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uu60F3QnaPx9B+l13aqXiTVsxpQdlWzr5UdUmy7HcSqvgB/a45SehHqtCU/E68ILxFR+d2FsGveysdEn5FmuCw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:15:53.639543Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.12325","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4e5858d79b565d3e166d5137bc739c1d87afcf432efd3e1a7b87e0e097da52d6","sha256:36d1143da447bc7b4cbb39ca0acf31073abd5e1a553b96b8bc2313ca47be459b"],"state_sha256":"b13638531bb51d2a1f25bffc1779292da2abbab5a303c3c3545f6b11bb03183a"}