{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:TWC7UP34FNVON73AMMDYRTTUPB","short_pith_number":"pith:TWC7UP34","canonical_record":{"source":{"id":"2605.13839","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-13T17:58:32Z","cross_cats_sorted":[],"title_canon_sha256":"0c3d912d52ea26cf8cfbc5482f2cc7d0cb63bfe9edd78b5532797e889e5f7692","abstract_canon_sha256":"2e823e305e915b0bf42db7b3bc647741a3e8de05c623468c4f13b128f471c999"},"schema_version":"1.0"},"canonical_sha256":"9d85fa3f7c2b6ae6ff60630788ce74787c796f0d7b88a5e1978956ebbc177c1d","source":{"kind":"arxiv","id":"2605.13839","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13839","created_at":"2026-05-18T02:44:09Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13839v1","created_at":"2026-05-18T02:44:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13839","created_at":"2026-05-18T02:44:09Z"},{"alias_kind":"pith_short_12","alias_value":"TWC7UP34FNVO","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"TWC7UP34FNVON73A","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"TWC7UP34","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:TWC7UP34FNVON73AMMDYRTTUPB","target":"record","payload":{"canonical_record":{"source":{"id":"2605.13839","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-13T17:58:32Z","cross_cats_sorted":[],"title_canon_sha256":"0c3d912d52ea26cf8cfbc5482f2cc7d0cb63bfe9edd78b5532797e889e5f7692","abstract_canon_sha256":"2e823e305e915b0bf42db7b3bc647741a3e8de05c623468c4f13b128f471c999"},"schema_version":"1.0"},"canonical_sha256":"9d85fa3f7c2b6ae6ff60630788ce74787c796f0d7b88a5e1978956ebbc177c1d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:44:09.567435Z","signature_b64":"PDhi5bnfy0KRF2p9NgvvoghHqF8jIQqfJwSn2fpYPi4uHrdN1v+GLLoKfXkKpEN7ZoxVNxG3eVbPgY/KxrZEDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9d85fa3f7c2b6ae6ff60630788ce74787c796f0d7b88a5e1978956ebbc177c1d","last_reissued_at":"2026-05-18T02:44:09.566975Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:44:09.566975Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.13839","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":"2026-05-18T02:44:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iET4Lep99Td5iGKMTv3pnPSx21PDjTCE8NMj+rPP2fZOkqneqSBCGdi3JLyiIxdtrZeeKvlJlBmH7U8fKFsSDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T01:39:00.013294Z"},"content_sha256":"bfaae79339dbe006978733084dbae74b0fb0e92abaa8f1cee908659506b98c03","schema_version":"1.0","event_id":"sha256:bfaae79339dbe006978733084dbae74b0fb0e92abaa8f1cee908659506b98c03"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:TWC7UP34FNVON73AMMDYRTTUPB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Good Agentic Friends Do Not Just Give Verbal Advice: They Can Update Your Weights","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Multi-agent LLMs can collaborate by mapping sender activations directly into transient low-rank weight updates on the receiver instead of passing text messages.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Huan Wang, Jian Wang, Kai Wang, Wenrui Bao, Yuzhang Shang, Zhangyang Wang","submitted_at":"2026-05-13T17:58:32Z","abstract_excerpt":"Multi-agent LLM systems usually collaborate by exchanging natural-language messages. This interface is simple and interpretable, but it forces each sender's intermediate computation to be serialized into tokens and then reprocessed by the receiver, thereby increasing the generated-token cost, prefill overhead, and KV-cache memory. We study an alternative communication interface: instead of appending a sender's message to the receiver's context, compile the sender's hidden states into a transient, receiver-specific weight perturbation. We introduce TFlow (Thought Flow), a weight-space communica"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"With three Qwen3-4B agents, TFlow improves over a standalone receiver by up to 8.5 accuracy points across five benchmarks while reducing processed tokens by up to 32.69%. Compared with a text-based three-agent baseline, it reduces total processed tokens by up to 83.27% and the wall-clock inference time by up to 4.6×, while maintaining competitive accuracy on four of five benchmarks.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That a learned parameter generator, trained once, can map arbitrary sender activations into effective, stable, receiver-specific LoRA perturbations for every new query without overfitting or degrading generation quality.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"TFlow enables multi-agent LLMs to collaborate via transient low-rank LoRA perturbations derived from sender activations, yielding up to 8.5 accuracy gains and 83% token reduction versus text-based baselines on Qwen3-4B models.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Multi-agent LLMs can collaborate by mapping sender activations directly into transient low-rank weight updates on the receiver instead of passing text messages.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"a612ec25da6fa95aef53f882f09dc8efb546770245bd71d1e72fa60790ec43e2"},"source":{"id":"2605.13839","kind":"arxiv","version":1},"verdict":{"id":"b63380da-48f3-4603-80fa-d63178e0ec61","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T18:57:09.200161Z","strongest_claim":"With three Qwen3-4B agents, TFlow improves over a standalone receiver by up to 8.5 accuracy points across five benchmarks while reducing processed tokens by up to 32.69%. Compared with a text-based three-agent baseline, it reduces total processed tokens by up to 83.27% and the wall-clock inference time by up to 4.6×, while maintaining competitive accuracy on four of five benchmarks.","one_line_summary":"TFlow enables multi-agent LLMs to collaborate via transient low-rank LoRA perturbations derived from sender activations, yielding up to 8.5 accuracy gains and 83% token reduction versus text-based baselines on Qwen3-4B models.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That a learned parameter generator, trained once, can map arbitrary sender activations into effective, stable, receiver-specific LoRA perturbations for every new query without overfitting or degrading generation quality.","pith_extraction_headline":"Multi-agent LLMs can collaborate by mapping sender activations directly into transient low-rank weight updates on the receiver instead of passing text messages."},"references":{"count":58,"sample":[{"doi":"","year":2023,"title":"Li, G., H. Hammoud, H. Itani, et al. Camel: Communicative agents for\" mind\" exploration of large language model society. InNeurIPS. 2023","work_id":"4d095bca-8e26-4f15-9003-bac7eafebb9c","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Hong, S., M. Zhuge, J. Chen, et al. Metagpt: Meta programming for a multi-agent collaborative framework. InICLR. 2023","work_id":"fbd9997d-1af3-4426-97df-90a48d7c0816","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Wu, Q., G. Bansal, J. Zhang, et al. AutoGen: Enabling next-gen LLM applications via multi- agent conversations. InCOLM. 2024","work_id":"544eb61c-a62f-4511-8b40-6b43090e3ef8","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Du, Y ., S. Li, A. Torralba, et al. Improving factuality and reasoning in language models through multiagent debate. InICML. 2024","work_id":"15f47422-cc65-4cb2-8f36-9938690c4dd3","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Liang, T., Z. He, W. Jiao, et al. Encouraging divergent thinking in large language models through multi-agent debate. InEMNLP. 2024","work_id":"e50ee9c8-7446-4b89-8f32-1e1571001335","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":58,"snapshot_sha256":"0ccd6656dfff0bfa309637122014ef8a4d5e70e610e52c34a145e15d75496ffa","internal_anchors":11},"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":"b63380da-48f3-4603-80fa-d63178e0ec61"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T02:44:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DeSy2ANk07BRnnviPRm+YJXpngc+Pl/M91+1rxk12e5PH+5wI0ungDPi7wlPWEdUp5ZgMZKfsgnc5dxEF22iCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T01:39:00.014450Z"},"content_sha256":"4013bb589444ce5c4f4e5290de21075e8133b899956aab6c242fa232a91aba5f","schema_version":"1.0","event_id":"sha256:4013bb589444ce5c4f4e5290de21075e8133b899956aab6c242fa232a91aba5f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TWC7UP34FNVON73AMMDYRTTUPB/bundle.json","state_url":"https://pith.science/pith/TWC7UP34FNVON73AMMDYRTTUPB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TWC7UP34FNVON73AMMDYRTTUPB/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":"2026-05-26T01:39:00Z","links":{"resolver":"https://pith.science/pith/TWC7UP34FNVON73AMMDYRTTUPB","bundle":"https://pith.science/pith/TWC7UP34FNVON73AMMDYRTTUPB/bundle.json","state":"https://pith.science/pith/TWC7UP34FNVON73AMMDYRTTUPB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TWC7UP34FNVON73AMMDYRTTUPB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:TWC7UP34FNVON73AMMDYRTTUPB","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":"2e823e305e915b0bf42db7b3bc647741a3e8de05c623468c4f13b128f471c999","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-13T17:58:32Z","title_canon_sha256":"0c3d912d52ea26cf8cfbc5482f2cc7d0cb63bfe9edd78b5532797e889e5f7692"},"schema_version":"1.0","source":{"id":"2605.13839","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13839","created_at":"2026-05-18T02:44:09Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13839v1","created_at":"2026-05-18T02:44:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13839","created_at":"2026-05-18T02:44:09Z"},{"alias_kind":"pith_short_12","alias_value":"TWC7UP34FNVO","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"TWC7UP34FNVON73A","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"TWC7UP34","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:4013bb589444ce5c4f4e5290de21075e8133b899956aab6c242fa232a91aba5f","target":"graph","created_at":"2026-05-18T02:44:09Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"With three Qwen3-4B agents, TFlow improves over a standalone receiver by up to 8.5 accuracy points across five benchmarks while reducing processed tokens by up to 32.69%. Compared with a text-based three-agent baseline, it reduces total processed tokens by up to 83.27% and the wall-clock inference time by up to 4.6×, while maintaining competitive accuracy on four of five benchmarks."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That a learned parameter generator, trained once, can map arbitrary sender activations into effective, stable, receiver-specific LoRA perturbations for every new query without overfitting or degrading generation quality."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"TFlow enables multi-agent LLMs to collaborate via transient low-rank LoRA perturbations derived from sender activations, yielding up to 8.5 accuracy gains and 83% token reduction versus text-based baselines on Qwen3-4B models."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Multi-agent LLMs can collaborate by mapping sender activations directly into transient low-rank weight updates on the receiver instead of passing text messages."}],"snapshot_sha256":"a612ec25da6fa95aef53f882f09dc8efb546770245bd71d1e72fa60790ec43e2"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Multi-agent LLM systems usually collaborate by exchanging natural-language messages. This interface is simple and interpretable, but it forces each sender's intermediate computation to be serialized into tokens and then reprocessed by the receiver, thereby increasing the generated-token cost, prefill overhead, and KV-cache memory. We study an alternative communication interface: instead of appending a sender's message to the receiver's context, compile the sender's hidden states into a transient, receiver-specific weight perturbation. We introduce TFlow (Thought Flow), a weight-space communica","authors_text":"Huan Wang, Jian Wang, Kai Wang, Wenrui Bao, Yuzhang Shang, Zhangyang Wang","cross_cats":[],"headline":"Multi-agent LLMs can collaborate by mapping sender activations directly into transient low-rank weight updates on the receiver instead of passing text messages.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-13T17:58:32Z","title":"Good Agentic Friends Do Not Just Give Verbal Advice: They Can Update Your Weights"},"references":{"count":58,"internal_anchors":11,"resolved_work":58,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Li, G., H. Hammoud, H. Itani, et al. Camel: Communicative agents for\" mind\" exploration of large language model society. InNeurIPS. 2023","work_id":"4d095bca-8e26-4f15-9003-bac7eafebb9c","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Hong, S., M. Zhuge, J. Chen, et al. Metagpt: Meta programming for a multi-agent collaborative framework. InICLR. 2023","work_id":"fbd9997d-1af3-4426-97df-90a48d7c0816","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Wu, Q., G. Bansal, J. Zhang, et al. AutoGen: Enabling next-gen LLM applications via multi- agent conversations. InCOLM. 2024","work_id":"544eb61c-a62f-4511-8b40-6b43090e3ef8","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Du, Y ., S. Li, A. Torralba, et al. Improving factuality and reasoning in language models through multiagent debate. InICML. 2024","work_id":"15f47422-cc65-4cb2-8f36-9938690c4dd3","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Liang, T., Z. He, W. Jiao, et al. Encouraging divergent thinking in large language models through multi-agent debate. InEMNLP. 2024","work_id":"e50ee9c8-7446-4b89-8f32-1e1571001335","year":2024}],"snapshot_sha256":"0ccd6656dfff0bfa309637122014ef8a4d5e70e610e52c34a145e15d75496ffa"},"source":{"id":"2605.13839","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T18:57:09.200161Z","id":"b63380da-48f3-4603-80fa-d63178e0ec61","model_set":{"reader":"grok-4.3"},"one_line_summary":"TFlow enables multi-agent LLMs to collaborate via transient low-rank LoRA perturbations derived from sender activations, yielding up to 8.5 accuracy gains and 83% token reduction versus text-based baselines on Qwen3-4B models.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Multi-agent LLMs can collaborate by mapping sender activations directly into transient low-rank weight updates on the receiver instead of passing text messages.","strongest_claim":"With three Qwen3-4B agents, TFlow improves over a standalone receiver by up to 8.5 accuracy points across five benchmarks while reducing processed tokens by up to 32.69%. Compared with a text-based three-agent baseline, it reduces total processed tokens by up to 83.27% and the wall-clock inference time by up to 4.6×, while maintaining competitive accuracy on four of five benchmarks.","weakest_assumption":"That a learned parameter generator, trained once, can map arbitrary sender activations into effective, stable, receiver-specific LoRA perturbations for every new query without overfitting or degrading generation quality."}},"verdict_id":"b63380da-48f3-4603-80fa-d63178e0ec61"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:bfaae79339dbe006978733084dbae74b0fb0e92abaa8f1cee908659506b98c03","target":"record","created_at":"2026-05-18T02:44:09Z","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":"2e823e305e915b0bf42db7b3bc647741a3e8de05c623468c4f13b128f471c999","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-13T17:58:32Z","title_canon_sha256":"0c3d912d52ea26cf8cfbc5482f2cc7d0cb63bfe9edd78b5532797e889e5f7692"},"schema_version":"1.0","source":{"id":"2605.13839","kind":"arxiv","version":1}},"canonical_sha256":"9d85fa3f7c2b6ae6ff60630788ce74787c796f0d7b88a5e1978956ebbc177c1d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9d85fa3f7c2b6ae6ff60630788ce74787c796f0d7b88a5e1978956ebbc177c1d","first_computed_at":"2026-05-18T02:44:09.566975Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T02:44:09.566975Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"PDhi5bnfy0KRF2p9NgvvoghHqF8jIQqfJwSn2fpYPi4uHrdN1v+GLLoKfXkKpEN7ZoxVNxG3eVbPgY/KxrZEDg==","signature_status":"signed_v1","signed_at":"2026-05-18T02:44:09.567435Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.13839","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bfaae79339dbe006978733084dbae74b0fb0e92abaa8f1cee908659506b98c03","sha256:4013bb589444ce5c4f4e5290de21075e8133b899956aab6c242fa232a91aba5f"],"state_sha256":"44b996a73528b6e5f7e5cb7e60723ae38c7361982138a66f878c7ff78eaff42d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OY5/PFJ45jPeRqJJ//8xbea/vukQh811TqiXASs/OZYyu5PUNwj2m+Xg4pSlpwhzLwhW1L0ZSo6TFOjZ5kvoCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T01:39:00.019234Z","bundle_sha256":"6d457a586a92e4806504a36d60199820fb7cbe2b9a270ca257022c3f36ae0972"}}