{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:K2V7VBKOYVM74S773Q3UACZFVH","short_pith_number":"pith:K2V7VBKO","canonical_record":{"source":{"id":"2507.07883","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-10T16:06:02Z","cross_cats_sorted":[],"title_canon_sha256":"5d6b1e1ee2c21e9613384fb2343a2239e45f049b0e60fb1af7f209e5e9aec8f2","abstract_canon_sha256":"09365882287b51343c481f76a7e0b59f15f79acf8e8c449ab8522ab2ca540458"},"schema_version":"1.0"},"canonical_sha256":"56abfa854ec559fe4bffdc37400b25a9e4e544fb8ebce01062c4bf1c82578102","source":{"kind":"arxiv","id":"2507.07883","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.07883","created_at":"2026-07-05T11:37:45Z"},{"alias_kind":"arxiv_version","alias_value":"2507.07883v3","created_at":"2026-07-05T11:37:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.07883","created_at":"2026-07-05T11:37:45Z"},{"alias_kind":"pith_short_12","alias_value":"K2V7VBKOYVM7","created_at":"2026-07-05T11:37:45Z"},{"alias_kind":"pith_short_16","alias_value":"K2V7VBKOYVM74S77","created_at":"2026-07-05T11:37:45Z"},{"alias_kind":"pith_short_8","alias_value":"K2V7VBKO","created_at":"2026-07-05T11:37:45Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:K2V7VBKOYVM74S773Q3UACZFVH","target":"record","payload":{"canonical_record":{"source":{"id":"2507.07883","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-10T16:06:02Z","cross_cats_sorted":[],"title_canon_sha256":"5d6b1e1ee2c21e9613384fb2343a2239e45f049b0e60fb1af7f209e5e9aec8f2","abstract_canon_sha256":"09365882287b51343c481f76a7e0b59f15f79acf8e8c449ab8522ab2ca540458"},"schema_version":"1.0"},"canonical_sha256":"56abfa854ec559fe4bffdc37400b25a9e4e544fb8ebce01062c4bf1c82578102","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:37:45.701083Z","signature_b64":"UI0miev666lKL1KakxEpZgZJeIXh51bcm4G6ltCG8KGPxqg4nYHHiyCnWb1Ln+JrStq51dFZDTXCUKljdEQCAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"56abfa854ec559fe4bffdc37400b25a9e4e544fb8ebce01062c4bf1c82578102","last_reissued_at":"2026-07-05T11:37:45.700374Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:37:45.700374Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.07883","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:37:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"z+imNxN3fphw57VNigmubwwIrXsEvd+jUZd5BdPSc6tAssx5qEii6Mb/SufxH3G1FbMR9ppVZS6KEyohc7tpCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T04:53:38.234763Z"},"content_sha256":"694868bd4ce93a9447d7d32543fc7f66c47c4d93ec684d2a704597e3c1ce1d1d","schema_version":"1.0","event_id":"sha256:694868bd4ce93a9447d7d32543fc7f66c47c4d93ec684d2a704597e3c1ce1d1d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:K2V7VBKOYVM74S773Q3UACZFVH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SAMO: A Lightweight Sharpness-Aware Approach for Multi-Task Optimization with Joint Global-Local Perturbation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Gokul Ram Subramani, Hao Ban, Kaiyi Ji","submitted_at":"2025-07-10T16:06:02Z","abstract_excerpt":"Multi-task learning (MTL) enables a joint model to capture commonalities across multiple tasks, reducing computation costs and improving data efficiency. However, a major challenge in MTL optimization is task conflicts, where the task gradients differ in direction or magnitude, limiting model performance compared to single-task counterparts. Sharpness-aware minimization (SAM) minimizes task loss while simultaneously reducing the sharpness of the loss landscape. Our empirical observations show that SAM effectively mitigates task conflicts in MTL. Motivated by these findings, we explore integrat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.07883","kind":"arxiv","version":3},"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/2507.07883/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":"2026-07-05T11:37:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DmscZz220oiDdf5HZxy/lvaQEgbN23yktucJYasPufDbUj04hTT1qLxRnc9O29/yeQcvX7ec7jcJ3mjTK+CxAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T04:53:38.235314Z"},"content_sha256":"54a7490d37f5f2b3cae99a95c37f7a06e3f0b995d5bd62d425d5c87f985b26e3","schema_version":"1.0","event_id":"sha256:54a7490d37f5f2b3cae99a95c37f7a06e3f0b995d5bd62d425d5c87f985b26e3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/K2V7VBKOYVM74S773Q3UACZFVH/bundle.json","state_url":"https://pith.science/pith/K2V7VBKOYVM74S773Q3UACZFVH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/K2V7VBKOYVM74S773Q3UACZFVH/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":"2026-08-04T04:53:38Z","links":{"resolver":"https://pith.science/pith/K2V7VBKOYVM74S773Q3UACZFVH","bundle":"https://pith.science/pith/K2V7VBKOYVM74S773Q3UACZFVH/bundle.json","state":"https://pith.science/pith/K2V7VBKOYVM74S773Q3UACZFVH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/K2V7VBKOYVM74S773Q3UACZFVH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:K2V7VBKOYVM74S773Q3UACZFVH","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":"09365882287b51343c481f76a7e0b59f15f79acf8e8c449ab8522ab2ca540458","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-10T16:06:02Z","title_canon_sha256":"5d6b1e1ee2c21e9613384fb2343a2239e45f049b0e60fb1af7f209e5e9aec8f2"},"schema_version":"1.0","source":{"id":"2507.07883","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.07883","created_at":"2026-07-05T11:37:45Z"},{"alias_kind":"arxiv_version","alias_value":"2507.07883v3","created_at":"2026-07-05T11:37:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.07883","created_at":"2026-07-05T11:37:45Z"},{"alias_kind":"pith_short_12","alias_value":"K2V7VBKOYVM7","created_at":"2026-07-05T11:37:45Z"},{"alias_kind":"pith_short_16","alias_value":"K2V7VBKOYVM74S77","created_at":"2026-07-05T11:37:45Z"},{"alias_kind":"pith_short_8","alias_value":"K2V7VBKO","created_at":"2026-07-05T11:37:45Z"}],"graph_snapshots":[{"event_id":"sha256:54a7490d37f5f2b3cae99a95c37f7a06e3f0b995d5bd62d425d5c87f985b26e3","target":"graph","created_at":"2026-07-05T11:37:45Z","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/2507.07883/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multi-task learning (MTL) enables a joint model to capture commonalities across multiple tasks, reducing computation costs and improving data efficiency. However, a major challenge in MTL optimization is task conflicts, where the task gradients differ in direction or magnitude, limiting model performance compared to single-task counterparts. Sharpness-aware minimization (SAM) minimizes task loss while simultaneously reducing the sharpness of the loss landscape. Our empirical observations show that SAM effectively mitigates task conflicts in MTL. Motivated by these findings, we explore integrat","authors_text":"Gokul Ram Subramani, Hao Ban, Kaiyi Ji","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-10T16:06:02Z","title":"SAMO: A Lightweight Sharpness-Aware Approach for Multi-Task Optimization with Joint Global-Local Perturbation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.07883","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":"sha256:694868bd4ce93a9447d7d32543fc7f66c47c4d93ec684d2a704597e3c1ce1d1d","target":"record","created_at":"2026-07-05T11:37:45Z","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":"09365882287b51343c481f76a7e0b59f15f79acf8e8c449ab8522ab2ca540458","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-10T16:06:02Z","title_canon_sha256":"5d6b1e1ee2c21e9613384fb2343a2239e45f049b0e60fb1af7f209e5e9aec8f2"},"schema_version":"1.0","source":{"id":"2507.07883","kind":"arxiv","version":3}},"canonical_sha256":"56abfa854ec559fe4bffdc37400b25a9e4e544fb8ebce01062c4bf1c82578102","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"56abfa854ec559fe4bffdc37400b25a9e4e544fb8ebce01062c4bf1c82578102","first_computed_at":"2026-07-05T11:37:45.700374Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:37:45.700374Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"UI0miev666lKL1KakxEpZgZJeIXh51bcm4G6ltCG8KGPxqg4nYHHiyCnWb1Ln+JrStq51dFZDTXCUKljdEQCAA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:37:45.701083Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.07883","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:694868bd4ce93a9447d7d32543fc7f66c47c4d93ec684d2a704597e3c1ce1d1d","sha256:54a7490d37f5f2b3cae99a95c37f7a06e3f0b995d5bd62d425d5c87f985b26e3"],"state_sha256":"8451e3642906fad7cde0e3015dafeb5de7731d5dd1d902ad2377a44366c2c0c1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GZ7RuQXH1xhaowpdtW6H1Tb0lmblwtKqqrJ43QAiGy4gvYC0JcVm4SWttChQE2givWvvCkXRiphTgG6JNamRBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T04:53:38.239182Z","bundle_sha256":"978252861bf7eb0095058d1a6999936cb47bb92eafdfb1bb5674805a76ab38d3"}}