{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:V2TOX4R4MOSZKMXIDCPDL6CRLQ","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":"6077328156e20fa8f57c9e7f3b28babbd24d00491a0819119b9cad73dd3ac555","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-16T20:14:55Z","title_canon_sha256":"617c2926ad5d67d6115e561e479e6701a0845161df79cffa01509c9d3e84265a"},"schema_version":"1.0","source":{"id":"2507.12612","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.12612","created_at":"2026-06-05T01:15:14Z"},{"alias_kind":"arxiv_version","alias_value":"2507.12612v3","created_at":"2026-06-05T01:15:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.12612","created_at":"2026-06-05T01:15:14Z"},{"alias_kind":"pith_short_12","alias_value":"V2TOX4R4MOSZ","created_at":"2026-06-05T01:15:14Z"},{"alias_kind":"pith_short_16","alias_value":"V2TOX4R4MOSZKMXI","created_at":"2026-06-05T01:15:14Z"},{"alias_kind":"pith_short_8","alias_value":"V2TOX4R4","created_at":"2026-06-05T01:15:14Z"}],"graph_snapshots":[{"event_id":"sha256:ce741693773b03e21669454833b27c5ee014133bb0ccc6d4a619e8a791cf4206","target":"graph","created_at":"2026-06-05T01:15:14Z","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/2507.12612/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Supervised fine-tuning performance for large language models depends strongly on how training budget is distributed across a heterogeneous set of tasks. In practice, mixtures are often fixed using simple heuristics (e.g., uniform or size-proportional sampling) that ignore task interactions, which can hurt transfer and waste budget on redundant sources. We introduce TaskPGM, a framework for learning continuous task mixtures via an energy-based model over tasks. Tasks form the nodes of a Markov random field: unary potentials capture per-task utility, and pairwise potentials encode inter-task rel","authors_text":"Ganesh Ramakrishnan, Krishnateja Killamsetty, Nikhil Shivakumar Nayak, Parth Pratim Chatterjee, Prateek Chanda, Saral Sureka","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-16T20:14:55Z","title":"Learning What Matters: Probabilistic Task Selection via Mutual Information for Model Finetuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.12612","kind":"arxiv","version":3},"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":"sha256:05d713222840ee38a673d2f6f0b868de9fe4a77fc9b9960fe39cf959d7812caf","target":"record","created_at":"2026-06-05T01:15:14Z","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":"6077328156e20fa8f57c9e7f3b28babbd24d00491a0819119b9cad73dd3ac555","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-16T20:14:55Z","title_canon_sha256":"617c2926ad5d67d6115e561e479e6701a0845161df79cffa01509c9d3e84265a"},"schema_version":"1.0","source":{"id":"2507.12612","kind":"arxiv","version":3}},"canonical_sha256":"aea6ebf23c63a59532e8189e35f8515c0cfbbef005e5f4b9cc087d0227ff4bed","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"aea6ebf23c63a59532e8189e35f8515c0cfbbef005e5f4b9cc087d0227ff4bed","first_computed_at":"2026-06-05T01:15:14.858885Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-05T01:15:14.858885Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"GlSbO89V6rMcXy52bVT5LfvgQcOo6DX5INyDdXg1x6r/0oLPHMMTrygM+q3HQk6k0yp7+hAstrPTu0fOUUH+Cg==","signature_status":"signed_v1","signed_at":"2026-06-05T01:15:14.859652Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.12612","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:05d713222840ee38a673d2f6f0b868de9fe4a77fc9b9960fe39cf959d7812caf","sha256:ce741693773b03e21669454833b27c5ee014133bb0ccc6d4a619e8a791cf4206"],"state_sha256":"34459951083f953bb8717b51b7ad6ada1fb3ca61c826974239c910f24486e29b"}