{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:GVMIA2JHN2VOKD3CV7IRN3AQHW","short_pith_number":"pith:GVMIA2JH","canonical_record":{"source":{"id":"2409.09086","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-11T12:44:12Z","cross_cats_sorted":["cs.AI","cs.CV","cs.DC","cs.PF"],"title_canon_sha256":"0348812eed2677ca0be5ab687c817adc27ea85a56b1207c7e575d53031f29d98","abstract_canon_sha256":"9b5160d8006bb63e5db282ecbe50f7a75d51d9d1373640eb980c9a9e915df132"},"schema_version":"1.0"},"canonical_sha256":"35588069276eaae50f62afd116ec103da820cc85664172cd2c8c5f9e624527f2","source":{"kind":"arxiv","id":"2409.09086","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.09086","created_at":"2026-07-05T09:07:04Z"},{"alias_kind":"arxiv_version","alias_value":"2409.09086v1","created_at":"2026-07-05T09:07:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.09086","created_at":"2026-07-05T09:07:04Z"},{"alias_kind":"pith_short_12","alias_value":"GVMIA2JHN2VO","created_at":"2026-07-05T09:07:04Z"},{"alias_kind":"pith_short_16","alias_value":"GVMIA2JHN2VOKD3C","created_at":"2026-07-05T09:07:04Z"},{"alias_kind":"pith_short_8","alias_value":"GVMIA2JH","created_at":"2026-07-05T09:07:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:GVMIA2JHN2VOKD3CV7IRN3AQHW","target":"record","payload":{"canonical_record":{"source":{"id":"2409.09086","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-11T12:44:12Z","cross_cats_sorted":["cs.AI","cs.CV","cs.DC","cs.PF"],"title_canon_sha256":"0348812eed2677ca0be5ab687c817adc27ea85a56b1207c7e575d53031f29d98","abstract_canon_sha256":"9b5160d8006bb63e5db282ecbe50f7a75d51d9d1373640eb980c9a9e915df132"},"schema_version":"1.0"},"canonical_sha256":"35588069276eaae50f62afd116ec103da820cc85664172cd2c8c5f9e624527f2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:07:04.735309Z","signature_b64":"YcU22mGiLUl7i1AvEmsE3rsPZ15Dh12Cb0xETRsBXoViQ8wTcNkh0r/OUqBWoEyjQ5sbP8clOFXtyP8tQSAKDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"35588069276eaae50f62afd116ec103da820cc85664172cd2c8c5f9e624527f2","last_reissued_at":"2026-07-05T09:07:04.734918Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:07:04.734918Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2409.09086","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":"2026-07-05T09:07:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wIC7x+/sl7rE4Squ6aQW4wN6+W0zDaIkdDkR6kmoJf2HoTY6ZFwxp27KAazvBE1nVXzlCM4JYz+bQ3T1JlJoAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T10:44:18.744230Z"},"content_sha256":"ba9b17c000f8035f1b5e8e4afa1ef6cfe5ff06f9d10930b7144e374b0a4991b9","schema_version":"1.0","event_id":"sha256:ba9b17c000f8035f1b5e8e4afa1ef6cfe5ff06f9d10930b7144e374b0a4991b9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:GVMIA2JHN2VOKD3CV7IRN3AQHW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Inf-MLLM: Efficient Streaming Inference of Multimodal Large Language Models on a Single GPU","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.DC","cs.PF"],"primary_cat":"cs.LG","authors_text":"Jieru Zhao, Minyi Guo, Qihao Jin, Wenchao Ding, Zhenyu Ning","submitted_at":"2024-09-11T12:44:12Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) are distinguished by their multimodal comprehensive ability and widely used in many real-world applications including GPT-4o, autonomous driving and robotics. Despite their impressive performance, the multimodal inputs always incur long context. The inference under long context requires caching massive Key and Value states (KV cache) of previous tokens, which introduces high latency and excessive memory consumption. Due to this reason, it is challenging to deploy streaming inference of MLLMs on edge devices, which largely constrains the power and usage "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.09086","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/2409.09086/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":"2026-07-05T09:07:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"z9QMDHcm7zR3ENbASoCEZZC989Vq81qzP3JMlKb7NM4KQuK9+1Jlf/KPxZEVj/OGQV4AnN8hiC7hdcPYNdjvDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T10:44:18.744792Z"},"content_sha256":"7a03880c8c5ff9bc736093760787c05d2becac3870511d1d345ad3c7566dc77b","schema_version":"1.0","event_id":"sha256:7a03880c8c5ff9bc736093760787c05d2becac3870511d1d345ad3c7566dc77b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GVMIA2JHN2VOKD3CV7IRN3AQHW/bundle.json","state_url":"https://pith.science/pith/GVMIA2JHN2VOKD3CV7IRN3AQHW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GVMIA2JHN2VOKD3CV7IRN3AQHW/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":"2026-08-11T10:44:18Z","links":{"resolver":"https://pith.science/pith/GVMIA2JHN2VOKD3CV7IRN3AQHW","bundle":"https://pith.science/pith/GVMIA2JHN2VOKD3CV7IRN3AQHW/bundle.json","state":"https://pith.science/pith/GVMIA2JHN2VOKD3CV7IRN3AQHW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GVMIA2JHN2VOKD3CV7IRN3AQHW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GVMIA2JHN2VOKD3CV7IRN3AQHW","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":"9b5160d8006bb63e5db282ecbe50f7a75d51d9d1373640eb980c9a9e915df132","cross_cats_sorted":["cs.AI","cs.CV","cs.DC","cs.PF"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-11T12:44:12Z","title_canon_sha256":"0348812eed2677ca0be5ab687c817adc27ea85a56b1207c7e575d53031f29d98"},"schema_version":"1.0","source":{"id":"2409.09086","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.09086","created_at":"2026-07-05T09:07:04Z"},{"alias_kind":"arxiv_version","alias_value":"2409.09086v1","created_at":"2026-07-05T09:07:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.09086","created_at":"2026-07-05T09:07:04Z"},{"alias_kind":"pith_short_12","alias_value":"GVMIA2JHN2VO","created_at":"2026-07-05T09:07:04Z"},{"alias_kind":"pith_short_16","alias_value":"GVMIA2JHN2VOKD3C","created_at":"2026-07-05T09:07:04Z"},{"alias_kind":"pith_short_8","alias_value":"GVMIA2JH","created_at":"2026-07-05T09:07:04Z"}],"graph_snapshots":[{"event_id":"sha256:7a03880c8c5ff9bc736093760787c05d2becac3870511d1d345ad3c7566dc77b","target":"graph","created_at":"2026-07-05T09:07:04Z","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/2409.09086/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal Large Language Models (MLLMs) are distinguished by their multimodal comprehensive ability and widely used in many real-world applications including GPT-4o, autonomous driving and robotics. Despite their impressive performance, the multimodal inputs always incur long context. The inference under long context requires caching massive Key and Value states (KV cache) of previous tokens, which introduces high latency and excessive memory consumption. Due to this reason, it is challenging to deploy streaming inference of MLLMs on edge devices, which largely constrains the power and usage ","authors_text":"Jieru Zhao, Minyi Guo, Qihao Jin, Wenchao Ding, Zhenyu Ning","cross_cats":["cs.AI","cs.CV","cs.DC","cs.PF"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-11T12:44:12Z","title":"Inf-MLLM: Efficient Streaming Inference of Multimodal Large Language Models on a Single GPU"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.09086","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:ba9b17c000f8035f1b5e8e4afa1ef6cfe5ff06f9d10930b7144e374b0a4991b9","target":"record","created_at":"2026-07-05T09:07:04Z","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":"9b5160d8006bb63e5db282ecbe50f7a75d51d9d1373640eb980c9a9e915df132","cross_cats_sorted":["cs.AI","cs.CV","cs.DC","cs.PF"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-11T12:44:12Z","title_canon_sha256":"0348812eed2677ca0be5ab687c817adc27ea85a56b1207c7e575d53031f29d98"},"schema_version":"1.0","source":{"id":"2409.09086","kind":"arxiv","version":1}},"canonical_sha256":"35588069276eaae50f62afd116ec103da820cc85664172cd2c8c5f9e624527f2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"35588069276eaae50f62afd116ec103da820cc85664172cd2c8c5f9e624527f2","first_computed_at":"2026-07-05T09:07:04.734918Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:07:04.734918Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YcU22mGiLUl7i1AvEmsE3rsPZ15Dh12Cb0xETRsBXoViQ8wTcNkh0r/OUqBWoEyjQ5sbP8clOFXtyP8tQSAKDg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:07:04.735309Z","signed_message":"canonical_sha256_bytes"},"source_id":"2409.09086","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ba9b17c000f8035f1b5e8e4afa1ef6cfe5ff06f9d10930b7144e374b0a4991b9","sha256:7a03880c8c5ff9bc736093760787c05d2becac3870511d1d345ad3c7566dc77b"],"state_sha256":"9b485d7e0d49c7697a757a301cf83d5edc640b4b03a16f310526cc1dad9da7d7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fnk8dvgjt2OHlz3b9f1iG9FlQNKCOWt7woG/y4sSKcb/loYYTXX1ci3/oWZyVOREDizP1rRCEumxmfUhSCdLAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T10:44:18.749642Z","bundle_sha256":"b7dbd1daa1cad99b63cd7f987a04a185d6a41aa11e6b7ad01da149da6acd1853"}}