{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:JIG42SMCP32VSHYHG4V4U3COF4","short_pith_number":"pith:JIG42SMC","canonical_record":{"source":{"id":"2507.19077","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-25T08:59:30Z","cross_cats_sorted":[],"title_canon_sha256":"ebd4cb6d6e4e914a05343d690bde1ea7dc754bed75cbc00ce77a45090ef0a476","abstract_canon_sha256":"aa936cf18dc5dddb9f1e1a327c0f39a0c6d28243658b0b9325e06ca73855c101"},"schema_version":"1.0"},"canonical_sha256":"4a0dcd49827ef5591f07372bca6c4e2f0efaacd9dd07f1f14b4d1987e1053430","source":{"kind":"arxiv","id":"2507.19077","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.19077","created_at":"2026-07-05T11:43:19Z"},{"alias_kind":"arxiv_version","alias_value":"2507.19077v1","created_at":"2026-07-05T11:43:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.19077","created_at":"2026-07-05T11:43:19Z"},{"alias_kind":"pith_short_12","alias_value":"JIG42SMCP32V","created_at":"2026-07-05T11:43:19Z"},{"alias_kind":"pith_short_16","alias_value":"JIG42SMCP32VSHYH","created_at":"2026-07-05T11:43:19Z"},{"alias_kind":"pith_short_8","alias_value":"JIG42SMC","created_at":"2026-07-05T11:43:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:JIG42SMCP32VSHYHG4V4U3COF4","target":"record","payload":{"canonical_record":{"source":{"id":"2507.19077","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-25T08:59:30Z","cross_cats_sorted":[],"title_canon_sha256":"ebd4cb6d6e4e914a05343d690bde1ea7dc754bed75cbc00ce77a45090ef0a476","abstract_canon_sha256":"aa936cf18dc5dddb9f1e1a327c0f39a0c6d28243658b0b9325e06ca73855c101"},"schema_version":"1.0"},"canonical_sha256":"4a0dcd49827ef5591f07372bca6c4e2f0efaacd9dd07f1f14b4d1987e1053430","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:43:19.239646Z","signature_b64":"5iejOoT3EbFKmEKXUtGBnhwPklWBABDyMV996WWxcRPewBdDE4+31IAJyawhr/wEcNSOWWkFBau1WP77u3EfCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4a0dcd49827ef5591f07372bca6c4e2f0efaacd9dd07f1f14b4d1987e1053430","last_reissued_at":"2026-07-05T11:43:19.239239Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:43:19.239239Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.19077","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":"2026-07-05T11:43:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cCSK/hNwYJJs1QLgN3P+WsSNLMM8jL/jea/kl6qBXsNiSiDrk7mO7G69Q79kpKgLQFYuVyfUWEfASxvvT6FRCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T06:40:59.657717Z"},"content_sha256":"68a7a868a011151b14bb703f7258f8004c5d4e70377ad58c4c77c4a4a2331a27","schema_version":"1.0","event_id":"sha256:68a7a868a011151b14bb703f7258f8004c5d4e70377ad58c4c77c4a4a2331a27"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:JIG42SMCP32VSHYHG4V4U3COF4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Multi-Task Dense Prediction Fine-Tuning with Mixture of Fine-Grained Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Duo Su, Xi Ye, Yangyang Xu","submitted_at":"2025-07-25T08:59:30Z","abstract_excerpt":"Multi-task learning (MTL) for dense prediction has shown promising results but still faces challenges in balancing shared representations with task-specific specialization. In this paper, we introduce a novel Fine-Grained Mixture of Experts (FGMoE) architecture that explores MoE-based MTL models through a combination of three key innovations and fine-tuning. First, we propose intra-task experts that partition along intermediate hidden dimensions of MLPs, enabling finer decomposition of task information while maintaining parameter efficiency. Second, we introduce shared experts that consolidate"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.19077","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/2507.19077/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":"2026-07-05T11:43:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+zITLaZ6ZfHySEjMFGOFDceftOoNBSTLsUp93R/OrjO0ypXF7QjKZpCzq8I3XCIpg19S7FFDzRPc0zMqPu4kAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T06:40:59.658233Z"},"content_sha256":"635808d178c07d556229ef83b4860340f2e85d2191f27e3b899733638c1438b7","schema_version":"1.0","event_id":"sha256:635808d178c07d556229ef83b4860340f2e85d2191f27e3b899733638c1438b7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JIG42SMCP32VSHYHG4V4U3COF4/bundle.json","state_url":"https://pith.science/pith/JIG42SMCP32VSHYHG4V4U3COF4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JIG42SMCP32VSHYHG4V4U3COF4/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":"2026-08-16T06:40:59Z","links":{"resolver":"https://pith.science/pith/JIG42SMCP32VSHYHG4V4U3COF4","bundle":"https://pith.science/pith/JIG42SMCP32VSHYHG4V4U3COF4/bundle.json","state":"https://pith.science/pith/JIG42SMCP32VSHYHG4V4U3COF4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JIG42SMCP32VSHYHG4V4U3COF4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:JIG42SMCP32VSHYHG4V4U3COF4","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":"aa936cf18dc5dddb9f1e1a327c0f39a0c6d28243658b0b9325e06ca73855c101","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-25T08:59:30Z","title_canon_sha256":"ebd4cb6d6e4e914a05343d690bde1ea7dc754bed75cbc00ce77a45090ef0a476"},"schema_version":"1.0","source":{"id":"2507.19077","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.19077","created_at":"2026-07-05T11:43:19Z"},{"alias_kind":"arxiv_version","alias_value":"2507.19077v1","created_at":"2026-07-05T11:43:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.19077","created_at":"2026-07-05T11:43:19Z"},{"alias_kind":"pith_short_12","alias_value":"JIG42SMCP32V","created_at":"2026-07-05T11:43:19Z"},{"alias_kind":"pith_short_16","alias_value":"JIG42SMCP32VSHYH","created_at":"2026-07-05T11:43:19Z"},{"alias_kind":"pith_short_8","alias_value":"JIG42SMC","created_at":"2026-07-05T11:43:19Z"}],"graph_snapshots":[{"event_id":"sha256:635808d178c07d556229ef83b4860340f2e85d2191f27e3b899733638c1438b7","target":"graph","created_at":"2026-07-05T11:43:19Z","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/2507.19077/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multi-task learning (MTL) for dense prediction has shown promising results but still faces challenges in balancing shared representations with task-specific specialization. In this paper, we introduce a novel Fine-Grained Mixture of Experts (FGMoE) architecture that explores MoE-based MTL models through a combination of three key innovations and fine-tuning. First, we propose intra-task experts that partition along intermediate hidden dimensions of MLPs, enabling finer decomposition of task information while maintaining parameter efficiency. Second, we introduce shared experts that consolidate","authors_text":"Duo Su, Xi Ye, Yangyang Xu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-25T08:59:30Z","title":"Multi-Task Dense Prediction Fine-Tuning with Mixture of Fine-Grained Experts"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.19077","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":"sha256:68a7a868a011151b14bb703f7258f8004c5d4e70377ad58c4c77c4a4a2331a27","target":"record","created_at":"2026-07-05T11:43:19Z","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":"aa936cf18dc5dddb9f1e1a327c0f39a0c6d28243658b0b9325e06ca73855c101","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-25T08:59:30Z","title_canon_sha256":"ebd4cb6d6e4e914a05343d690bde1ea7dc754bed75cbc00ce77a45090ef0a476"},"schema_version":"1.0","source":{"id":"2507.19077","kind":"arxiv","version":1}},"canonical_sha256":"4a0dcd49827ef5591f07372bca6c4e2f0efaacd9dd07f1f14b4d1987e1053430","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4a0dcd49827ef5591f07372bca6c4e2f0efaacd9dd07f1f14b4d1987e1053430","first_computed_at":"2026-07-05T11:43:19.239239Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:43:19.239239Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"5iejOoT3EbFKmEKXUtGBnhwPklWBABDyMV996WWxcRPewBdDE4+31IAJyawhr/wEcNSOWWkFBau1WP77u3EfCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:43:19.239646Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.19077","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:68a7a868a011151b14bb703f7258f8004c5d4e70377ad58c4c77c4a4a2331a27","sha256:635808d178c07d556229ef83b4860340f2e85d2191f27e3b899733638c1438b7"],"state_sha256":"4d4c7d49c1bb4dedec9778605ee4dd29a55d081b29ec5a2f4e4894a676af01b8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KGDnu9BXlA6EbakVePb6iavOM5vgE+8a8VV4vnRrvDjIvIEQ3cRnwu5rTJLrsUVstw0dp47/qsV45b7UsSNMCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T06:40:59.662581Z","bundle_sha256":"f9847fe9cd98e8f20ca7c2cf97d1d3d2511f2e7b4d67864cfe10b7efb4d34198"}}