{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:YMRDBEDMMNCEHDMYHLB2AXLRXY","short_pith_number":"pith:YMRDBEDM","canonical_record":{"source":{"id":"2309.09958","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-09-18T17:30:46Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"2567e7dedbef3a20c7e8bac2b28ec009be7db053543f68c9887658072614d6bb","abstract_canon_sha256":"b9fe87f797d685fdcad77dd4f6d8ced98e5d83468de301632e28f0f98bc75239"},"schema_version":"1.0"},"canonical_sha256":"c32230906c6344438d983ac3a05d71be0f2e2ad22fa5cc72c108aac40a86e30a","source":{"kind":"arxiv","id":"2309.09958","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.09958","created_at":"2026-07-05T06:51:47Z"},{"alias_kind":"arxiv_version","alias_value":"2309.09958v1","created_at":"2026-07-05T06:51:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.09958","created_at":"2026-07-05T06:51:47Z"},{"alias_kind":"pith_short_12","alias_value":"YMRDBEDMMNCE","created_at":"2026-07-05T06:51:47Z"},{"alias_kind":"pith_short_16","alias_value":"YMRDBEDMMNCEHDMY","created_at":"2026-07-05T06:51:47Z"},{"alias_kind":"pith_short_8","alias_value":"YMRDBEDM","created_at":"2026-07-05T06:51:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:YMRDBEDMMNCEHDMYHLB2AXLRXY","target":"record","payload":{"canonical_record":{"source":{"id":"2309.09958","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-09-18T17:30:46Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"2567e7dedbef3a20c7e8bac2b28ec009be7db053543f68c9887658072614d6bb","abstract_canon_sha256":"b9fe87f797d685fdcad77dd4f6d8ced98e5d83468de301632e28f0f98bc75239"},"schema_version":"1.0"},"canonical_sha256":"c32230906c6344438d983ac3a05d71be0f2e2ad22fa5cc72c108aac40a86e30a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:51:47.145438Z","signature_b64":"70q/xT12SyQD+6cEM2JgEll69YzHfrHHcLoDRERH8r0mWlK7NvkM8vw/FbWJLAtqsspiFBxYhCJPILS+hoPpBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c32230906c6344438d983ac3a05d71be0f2e2ad22fa5cc72c108aac40a86e30a","last_reissued_at":"2026-07-05T06:51:47.144925Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:51:47.144925Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2309.09958","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":"2026-07-05T06:51:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0PsGviHa7oySW638zqxZtwQttGzo36xuoi0IYq9jxzkHnr8kc6d9dem5FSVGMiVRtIjmaIFROzP6fBV1CJ4vDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-09T06:38:43.493812Z"},"content_sha256":"c0246e1b2c72d7114381919be12ecd9541bff1ced8603528abe41adafb7225f3","schema_version":"1.0","event_id":"sha256:c0246e1b2c72d7114381919be12ecd9541bff1ced8603528abe41adafb7225f3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:YMRDBEDMMNCEHDMYHLB2AXLRXY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"An Empirical Study of Scaling Instruct-Tuned Large Multimodal Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Chunyuan Li, Haotian Liu, Jianfeng Gao, Jianwei Yang, Yadong Lu, Yelong Shen","submitted_at":"2023-09-18T17:30:46Z","abstract_excerpt":"Visual instruction tuning has recently shown encouraging progress with open-source large multimodal models (LMM) such as LLaVA and MiniGPT-4. However, most existing studies of open-source LMM are performed using models with 13B parameters or smaller. In this paper we present an empirical study of scaling LLaVA up to 33B and 65B/70B, and share our findings from our explorations in image resolution, data mixing and parameter-efficient training methods such as LoRA/QLoRA. These are evaluated by their impact on the multi-modal and language capabilities when completing real-world tasks in the wild."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.09958","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/2309.09958/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":"2026-07-05T06:51:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"b9RgL+JiltwGT9OOh0X+00WmL9kRi6Bo/iPiI+tdi29Cdo/gjT+SfDEkeWKm2IVr7TweWt1rMfGHhheV6CSqDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-09T06:38:43.494739Z"},"content_sha256":"ae68be38b097e966b6df724cbea0669264b5494859dda4344559363e45518b1a","schema_version":"1.0","event_id":"sha256:ae68be38b097e966b6df724cbea0669264b5494859dda4344559363e45518b1a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YMRDBEDMMNCEHDMYHLB2AXLRXY/bundle.json","state_url":"https://pith.science/pith/YMRDBEDMMNCEHDMYHLB2AXLRXY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YMRDBEDMMNCEHDMYHLB2AXLRXY/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":"2026-07-09T06:38:43Z","links":{"resolver":"https://pith.science/pith/YMRDBEDMMNCEHDMYHLB2AXLRXY","bundle":"https://pith.science/pith/YMRDBEDMMNCEHDMYHLB2AXLRXY/bundle.json","state":"https://pith.science/pith/YMRDBEDMMNCEHDMYHLB2AXLRXY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YMRDBEDMMNCEHDMYHLB2AXLRXY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:YMRDBEDMMNCEHDMYHLB2AXLRXY","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":"b9fe87f797d685fdcad77dd4f6d8ced98e5d83468de301632e28f0f98bc75239","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-09-18T17:30:46Z","title_canon_sha256":"2567e7dedbef3a20c7e8bac2b28ec009be7db053543f68c9887658072614d6bb"},"schema_version":"1.0","source":{"id":"2309.09958","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.09958","created_at":"2026-07-05T06:51:47Z"},{"alias_kind":"arxiv_version","alias_value":"2309.09958v1","created_at":"2026-07-05T06:51:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.09958","created_at":"2026-07-05T06:51:47Z"},{"alias_kind":"pith_short_12","alias_value":"YMRDBEDMMNCE","created_at":"2026-07-05T06:51:47Z"},{"alias_kind":"pith_short_16","alias_value":"YMRDBEDMMNCEHDMY","created_at":"2026-07-05T06:51:47Z"},{"alias_kind":"pith_short_8","alias_value":"YMRDBEDM","created_at":"2026-07-05T06:51:47Z"}],"graph_snapshots":[{"event_id":"sha256:ae68be38b097e966b6df724cbea0669264b5494859dda4344559363e45518b1a","target":"graph","created_at":"2026-07-05T06:51:47Z","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/2309.09958/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Visual instruction tuning has recently shown encouraging progress with open-source large multimodal models (LMM) such as LLaVA and MiniGPT-4. However, most existing studies of open-source LMM are performed using models with 13B parameters or smaller. In this paper we present an empirical study of scaling LLaVA up to 33B and 65B/70B, and share our findings from our explorations in image resolution, data mixing and parameter-efficient training methods such as LoRA/QLoRA. These are evaluated by their impact on the multi-modal and language capabilities when completing real-world tasks in the wild.","authors_text":"Chunyuan Li, Haotian Liu, Jianfeng Gao, Jianwei Yang, Yadong Lu, Yelong Shen","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-09-18T17:30:46Z","title":"An Empirical Study of Scaling Instruct-Tuned Large Multimodal Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.09958","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:c0246e1b2c72d7114381919be12ecd9541bff1ced8603528abe41adafb7225f3","target":"record","created_at":"2026-07-05T06:51:47Z","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":"b9fe87f797d685fdcad77dd4f6d8ced98e5d83468de301632e28f0f98bc75239","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-09-18T17:30:46Z","title_canon_sha256":"2567e7dedbef3a20c7e8bac2b28ec009be7db053543f68c9887658072614d6bb"},"schema_version":"1.0","source":{"id":"2309.09958","kind":"arxiv","version":1}},"canonical_sha256":"c32230906c6344438d983ac3a05d71be0f2e2ad22fa5cc72c108aac40a86e30a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c32230906c6344438d983ac3a05d71be0f2e2ad22fa5cc72c108aac40a86e30a","first_computed_at":"2026-07-05T06:51:47.144925Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:51:47.144925Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"70q/xT12SyQD+6cEM2JgEll69YzHfrHHcLoDRERH8r0mWlK7NvkM8vw/FbWJLAtqsspiFBxYhCJPILS+hoPpBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T06:51:47.145438Z","signed_message":"canonical_sha256_bytes"},"source_id":"2309.09958","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c0246e1b2c72d7114381919be12ecd9541bff1ced8603528abe41adafb7225f3","sha256:ae68be38b097e966b6df724cbea0669264b5494859dda4344559363e45518b1a"],"state_sha256":"4b9f1959afd004be43273da162e4eb287047e4ef91329988f69b00536b597fec"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1bQwyI/g/2seFQPkwM8gu8YSb9X/nfXAqXfaeKrDcOXz4qI5ny9zIMIg6uWZpNc3hcWIKaEOUvdjv1VPwEMGAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-09T06:38:43.498283Z","bundle_sha256":"59a4e43ee792125fccb968d993b8a71b9442252416761bb737cebfc1c9481356"}}