{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:564EZXKGUPKRV4JDUSBCDLKXQC","short_pith_number":"pith:564EZXKG","canonical_record":{"source":{"id":"2607.05978","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-07T08:10:57Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9e1c01e5265874562dfe8ec85f7a3745017c1c98519e81e6050f393d5160dd57","abstract_canon_sha256":"beff817f40b0f1cb965820d8a515a73cc65a2b58b55424d354b4563287f74c18"},"schema_version":"1.0"},"canonical_sha256":"efb84cdd46a3d51af123a48221ad57808078ca959da8420501a98fc58e3642f7","source":{"kind":"arxiv","id":"2607.05978","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.05978","created_at":"2026-07-08T01:18:52Z"},{"alias_kind":"arxiv_version","alias_value":"2607.05978v1","created_at":"2026-07-08T01:18:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.05978","created_at":"2026-07-08T01:18:52Z"},{"alias_kind":"pith_short_12","alias_value":"564EZXKGUPKR","created_at":"2026-07-08T01:18:52Z"},{"alias_kind":"pith_short_16","alias_value":"564EZXKGUPKRV4JD","created_at":"2026-07-08T01:18:52Z"},{"alias_kind":"pith_short_8","alias_value":"564EZXKG","created_at":"2026-07-08T01:18:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:564EZXKGUPKRV4JDUSBCDLKXQC","target":"record","payload":{"canonical_record":{"source":{"id":"2607.05978","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-07T08:10:57Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9e1c01e5265874562dfe8ec85f7a3745017c1c98519e81e6050f393d5160dd57","abstract_canon_sha256":"beff817f40b0f1cb965820d8a515a73cc65a2b58b55424d354b4563287f74c18"},"schema_version":"1.0"},"canonical_sha256":"efb84cdd46a3d51af123a48221ad57808078ca959da8420501a98fc58e3642f7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-08T01:18:52.305172Z","signature_b64":"YTbMMPnUx7A8SXt7v5gQPlW9mzU0JL7jD4mONViMDDZfx0ZRGaUwPniIhvYxKGAMEOq5gQO5Q3PPhKOU+TBfBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"efb84cdd46a3d51af123a48221ad57808078ca959da8420501a98fc58e3642f7","last_reissued_at":"2026-07-08T01:18:52.304713Z","signature_status":"signed_v1","first_computed_at":"2026-07-08T01:18:52.304713Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.05978","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":"2026-07-08T01:18:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"d51tcZHkoRJwNAU9duxX5E9jOOrWJqC5zVmMAJ/4Y+bzrpuKdog1K94Jt4Z6ea/aCcYAzlynnC3y25dX0DM7CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T12:34:17.686573Z"},"content_sha256":"ca9a89d0bebad928d8e8859783ed06fe237d26540622a6e93bee48fdacbbbfc1","schema_version":"1.0","event_id":"sha256:ca9a89d0bebad928d8e8859783ed06fe237d26540622a6e93bee48fdacbbbfc1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:564EZXKGUPKRV4JDUSBCDLKXQC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Propose and Attend: Training-free MLLM Grounding Confidence via Multi-Token Localized Attention","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Avi Ben Cohen, Daniel Shalam, Emanuel Ben Baruch, Tal Remez","submitted_at":"2026-07-07T08:10:57Z","abstract_excerpt":"Multimodal large language models can emit localized predictions, bounding boxes for objects and temporal windows for video and audio events, but they hallucinate these regions prolifically. The model's own token log-probabilities are nearly uninformative: they conflate grounding quality with input ambiguity, and coordinate tokens become near-deterministic once the model commits. We propose Multi-Token Localized Attention (MTLA): a training-free, post-hoc score that measures how strongly a prediction's tokens attend to the region they claim. Prior attention-based detectors, which sum attention "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.05978","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/2607.05978/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":"2026-07-08T01:18:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DKKCPDQ1/dPANK9qhu/NmEmdaTCphHeb97fcmH4ISplXJQFiJNrbI2yWuyd1ynbeW+qH7GyN/UK/O0HsJEWfAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T12:34:17.687183Z"},"content_sha256":"b27b8542970d894e8b607ffa837b392b9c9c21d53470dcdecc6beb46f361b686","schema_version":"1.0","event_id":"sha256:b27b8542970d894e8b607ffa837b392b9c9c21d53470dcdecc6beb46f361b686"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/564EZXKGUPKRV4JDUSBCDLKXQC/bundle.json","state_url":"https://pith.science/pith/564EZXKGUPKRV4JDUSBCDLKXQC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/564EZXKGUPKRV4JDUSBCDLKXQC/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":"2026-08-05T12:34:17Z","links":{"resolver":"https://pith.science/pith/564EZXKGUPKRV4JDUSBCDLKXQC","bundle":"https://pith.science/pith/564EZXKGUPKRV4JDUSBCDLKXQC/bundle.json","state":"https://pith.science/pith/564EZXKGUPKRV4JDUSBCDLKXQC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/564EZXKGUPKRV4JDUSBCDLKXQC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:564EZXKGUPKRV4JDUSBCDLKXQC","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":"beff817f40b0f1cb965820d8a515a73cc65a2b58b55424d354b4563287f74c18","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-07T08:10:57Z","title_canon_sha256":"9e1c01e5265874562dfe8ec85f7a3745017c1c98519e81e6050f393d5160dd57"},"schema_version":"1.0","source":{"id":"2607.05978","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.05978","created_at":"2026-07-08T01:18:52Z"},{"alias_kind":"arxiv_version","alias_value":"2607.05978v1","created_at":"2026-07-08T01:18:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.05978","created_at":"2026-07-08T01:18:52Z"},{"alias_kind":"pith_short_12","alias_value":"564EZXKGUPKR","created_at":"2026-07-08T01:18:52Z"},{"alias_kind":"pith_short_16","alias_value":"564EZXKGUPKRV4JD","created_at":"2026-07-08T01:18:52Z"},{"alias_kind":"pith_short_8","alias_value":"564EZXKG","created_at":"2026-07-08T01:18:52Z"}],"graph_snapshots":[{"event_id":"sha256:b27b8542970d894e8b607ffa837b392b9c9c21d53470dcdecc6beb46f361b686","target":"graph","created_at":"2026-07-08T01:18:52Z","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/2607.05978/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal large language models can emit localized predictions, bounding boxes for objects and temporal windows for video and audio events, but they hallucinate these regions prolifically. The model's own token log-probabilities are nearly uninformative: they conflate grounding quality with input ambiguity, and coordinate tokens become near-deterministic once the model commits. We propose Multi-Token Localized Attention (MTLA): a training-free, post-hoc score that measures how strongly a prediction's tokens attend to the region they claim. Prior attention-based detectors, which sum attention ","authors_text":"Avi Ben Cohen, Daniel Shalam, Emanuel Ben Baruch, Tal Remez","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-07T08:10:57Z","title":"Propose and Attend: Training-free MLLM Grounding Confidence via Multi-Token Localized Attention"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.05978","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:ca9a89d0bebad928d8e8859783ed06fe237d26540622a6e93bee48fdacbbbfc1","target":"record","created_at":"2026-07-08T01:18:52Z","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":"beff817f40b0f1cb965820d8a515a73cc65a2b58b55424d354b4563287f74c18","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-07T08:10:57Z","title_canon_sha256":"9e1c01e5265874562dfe8ec85f7a3745017c1c98519e81e6050f393d5160dd57"},"schema_version":"1.0","source":{"id":"2607.05978","kind":"arxiv","version":1}},"canonical_sha256":"efb84cdd46a3d51af123a48221ad57808078ca959da8420501a98fc58e3642f7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"efb84cdd46a3d51af123a48221ad57808078ca959da8420501a98fc58e3642f7","first_computed_at":"2026-07-08T01:18:52.304713Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-08T01:18:52.304713Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YTbMMPnUx7A8SXt7v5gQPlW9mzU0JL7jD4mONViMDDZfx0ZRGaUwPniIhvYxKGAMEOq5gQO5Q3PPhKOU+TBfBA==","signature_status":"signed_v1","signed_at":"2026-07-08T01:18:52.305172Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.05978","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ca9a89d0bebad928d8e8859783ed06fe237d26540622a6e93bee48fdacbbbfc1","sha256:b27b8542970d894e8b607ffa837b392b9c9c21d53470dcdecc6beb46f361b686"],"state_sha256":"49b7556c6e699fb8240e601c069b63e4a84c3c3435d25a6636133462f589b458"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XPRZBNDX8SWJeGP5IwSNSsyafdFMd9/OaT4bBSmbjjpFp/36OnYnMsxyNOWln8EI249qdWJ4OUg8SVOZwWs+Dw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T12:34:17.694093Z","bundle_sha256":"942899ef7eab63f00730f8b424f88e3bc0d58579315951e73d2160c868328c43"}}