{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:BCX2TBY2CQO62JATQJSDCM3ZZI","short_pith_number":"pith:BCX2TBY2","canonical_record":{"source":{"id":"2502.06814","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-04T22:20:20Z","cross_cats_sorted":["cs.AI","cs.GR"],"title_canon_sha256":"3ee3614c7786ac604a08fa817a5f04235655122aa9588a09d8fa6ea75b9eda58","abstract_canon_sha256":"1817f6c2f20aa223875bdbf63607314eb20ddb13e8c983c9e5bcd61bc7798549"},"schema_version":"1.0"},"canonical_sha256":"08afa9871a141ded24138264313379ca0b83bb941bf3da7719dd77a39bc4ff14","source":{"kind":"arxiv","id":"2502.06814","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.06814","created_at":"2026-07-05T11:09:07Z"},{"alias_kind":"arxiv_version","alias_value":"2502.06814v2","created_at":"2026-07-05T11:09:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.06814","created_at":"2026-07-05T11:09:07Z"},{"alias_kind":"pith_short_12","alias_value":"BCX2TBY2CQO6","created_at":"2026-07-05T11:09:07Z"},{"alias_kind":"pith_short_16","alias_value":"BCX2TBY2CQO62JAT","created_at":"2026-07-05T11:09:07Z"},{"alias_kind":"pith_short_8","alias_value":"BCX2TBY2","created_at":"2026-07-05T11:09:07Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:BCX2TBY2CQO62JATQJSDCM3ZZI","target":"record","payload":{"canonical_record":{"source":{"id":"2502.06814","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-04T22:20:20Z","cross_cats_sorted":["cs.AI","cs.GR"],"title_canon_sha256":"3ee3614c7786ac604a08fa817a5f04235655122aa9588a09d8fa6ea75b9eda58","abstract_canon_sha256":"1817f6c2f20aa223875bdbf63607314eb20ddb13e8c983c9e5bcd61bc7798549"},"schema_version":"1.0"},"canonical_sha256":"08afa9871a141ded24138264313379ca0b83bb941bf3da7719dd77a39bc4ff14","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:09:07.960822Z","signature_b64":"XVfiJKqkM4KenyDn7qbIrEOXXIfzkXe9gGWzrio6Uk4RkGsTZIsiIS03sz26hYNGYVoOvwYjvt/1KKURFk+3DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"08afa9871a141ded24138264313379ca0b83bb941bf3da7719dd77a39bc4ff14","last_reissued_at":"2026-07-05T11:09:07.960362Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:09:07.960362Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.06814","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:09:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+Gbu3Q679o6as+5JrCdRa+Qm74+dXMW30FgRP7mtIE7x8ZzhiBN3fkj6NmvSRjWxy4F39J8aaqaKjKcHHusCBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T02:53:03.362615Z"},"content_sha256":"d2cfd0bebd2771a3caa41ed6175b7465047fd89bff238070f2cd04199ff06bc1","schema_version":"1.0","event_id":"sha256:d2cfd0bebd2771a3caa41ed6175b7465047fd89bff238070f2cd04199ff06bc1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:BCX2TBY2CQO62JATQJSDCM3ZZI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Diffusion Instruction Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.GR"],"primary_cat":"cs.LG","authors_text":"Amrutha Saseendran, Chen Jin, Philip Teare, Ryutaro Tanno, Tom Diethe","submitted_at":"2025-02-04T22:20:20Z","abstract_excerpt":"We introduce Lavender, a simple supervised fine-tuning (SFT) method that boosts the performance of advanced vision-language models (VLMs) by leveraging state-of-the-art image generation models such as Stable Diffusion. Specifically, Lavender aligns the text-vision attention in the VLM transformer with the equivalent used by Stable Diffusion during SFT, instead of adapting separate encoders. This alignment enriches the model's visual understanding and significantly boosts performance across in- and out-of-distribution tasks. Lavender requires just 0.13 million training examples, 2.5% of typical"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.06814","kind":"arxiv","version":2},"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/2502.06814/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":"2026-07-05T11:09:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kXS8nM3AYhXNiF5176JZaoNcb7MoHCghAGGnHwqsV150bZEtxCjhCVPn0FuJpLGF4eKnw3m63YTa3MiO6PAxAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T02:53:03.363122Z"},"content_sha256":"1965860d681fed950374d45b3149c1e3ef4df49d70a489b28c7e85a4354a6801","schema_version":"1.0","event_id":"sha256:1965860d681fed950374d45b3149c1e3ef4df49d70a489b28c7e85a4354a6801"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BCX2TBY2CQO62JATQJSDCM3ZZI/bundle.json","state_url":"https://pith.science/pith/BCX2TBY2CQO62JATQJSDCM3ZZI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BCX2TBY2CQO62JATQJSDCM3ZZI/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":"2026-08-10T02:53:03Z","links":{"resolver":"https://pith.science/pith/BCX2TBY2CQO62JATQJSDCM3ZZI","bundle":"https://pith.science/pith/BCX2TBY2CQO62JATQJSDCM3ZZI/bundle.json","state":"https://pith.science/pith/BCX2TBY2CQO62JATQJSDCM3ZZI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BCX2TBY2CQO62JATQJSDCM3ZZI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:BCX2TBY2CQO62JATQJSDCM3ZZI","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":"1817f6c2f20aa223875bdbf63607314eb20ddb13e8c983c9e5bcd61bc7798549","cross_cats_sorted":["cs.AI","cs.GR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-04T22:20:20Z","title_canon_sha256":"3ee3614c7786ac604a08fa817a5f04235655122aa9588a09d8fa6ea75b9eda58"},"schema_version":"1.0","source":{"id":"2502.06814","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.06814","created_at":"2026-07-05T11:09:07Z"},{"alias_kind":"arxiv_version","alias_value":"2502.06814v2","created_at":"2026-07-05T11:09:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.06814","created_at":"2026-07-05T11:09:07Z"},{"alias_kind":"pith_short_12","alias_value":"BCX2TBY2CQO6","created_at":"2026-07-05T11:09:07Z"},{"alias_kind":"pith_short_16","alias_value":"BCX2TBY2CQO62JAT","created_at":"2026-07-05T11:09:07Z"},{"alias_kind":"pith_short_8","alias_value":"BCX2TBY2","created_at":"2026-07-05T11:09:07Z"}],"graph_snapshots":[{"event_id":"sha256:1965860d681fed950374d45b3149c1e3ef4df49d70a489b28c7e85a4354a6801","target":"graph","created_at":"2026-07-05T11:09:07Z","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/2502.06814/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce Lavender, a simple supervised fine-tuning (SFT) method that boosts the performance of advanced vision-language models (VLMs) by leveraging state-of-the-art image generation models such as Stable Diffusion. Specifically, Lavender aligns the text-vision attention in the VLM transformer with the equivalent used by Stable Diffusion during SFT, instead of adapting separate encoders. This alignment enriches the model's visual understanding and significantly boosts performance across in- and out-of-distribution tasks. Lavender requires just 0.13 million training examples, 2.5% of typical","authors_text":"Amrutha Saseendran, Chen Jin, Philip Teare, Ryutaro Tanno, Tom Diethe","cross_cats":["cs.AI","cs.GR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-04T22:20:20Z","title":"Diffusion Instruction Tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.06814","kind":"arxiv","version":2},"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:d2cfd0bebd2771a3caa41ed6175b7465047fd89bff238070f2cd04199ff06bc1","target":"record","created_at":"2026-07-05T11:09:07Z","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":"1817f6c2f20aa223875bdbf63607314eb20ddb13e8c983c9e5bcd61bc7798549","cross_cats_sorted":["cs.AI","cs.GR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-04T22:20:20Z","title_canon_sha256":"3ee3614c7786ac604a08fa817a5f04235655122aa9588a09d8fa6ea75b9eda58"},"schema_version":"1.0","source":{"id":"2502.06814","kind":"arxiv","version":2}},"canonical_sha256":"08afa9871a141ded24138264313379ca0b83bb941bf3da7719dd77a39bc4ff14","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"08afa9871a141ded24138264313379ca0b83bb941bf3da7719dd77a39bc4ff14","first_computed_at":"2026-07-05T11:09:07.960362Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:09:07.960362Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XVfiJKqkM4KenyDn7qbIrEOXXIfzkXe9gGWzrio6Uk4RkGsTZIsiIS03sz26hYNGYVoOvwYjvt/1KKURFk+3DQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:09:07.960822Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.06814","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d2cfd0bebd2771a3caa41ed6175b7465047fd89bff238070f2cd04199ff06bc1","sha256:1965860d681fed950374d45b3149c1e3ef4df49d70a489b28c7e85a4354a6801"],"state_sha256":"ca622fe4925f9e00e2efd9ebcfb392930ddd45fccdf6b15c99e3ed048d6f58a7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"r22WZFSPRgKPlQDRZR4WT8NknpPMlu//K3MaLYJ5JfzaRO6Dq1ujOxtb5nrKJPymfiq3c0tRX50IupAymXfmCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T02:53:03.368261Z","bundle_sha256":"71cfa10c9174725f2298f0b75dbcbdaaa843e7c61caf757ebda5749a7fc1e99c"}}