{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:BSHHHTD377QA3M7LKDDRL3JU7M","short_pith_number":"pith:BSHHHTD3","canonical_record":{"source":{"id":"2606.24667","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-23T15:00:30Z","cross_cats_sorted":[],"title_canon_sha256":"d28ab6c4cb88403dafe09d64dac1d79676ac8d458f58a041155ff1b158187810","abstract_canon_sha256":"4e2ad92c7692677bdb4650fbb7f227dcfaea2e8e37eee48a11b74b5e70c1e70a"},"schema_version":"1.0"},"canonical_sha256":"0c8e73cc7bffe00db3eb50c715ed34fb09b697706e56ba3b03a4548dda284f8f","source":{"kind":"arxiv","id":"2606.24667","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.24667","created_at":"2026-06-24T01:15:38Z"},{"alias_kind":"arxiv_version","alias_value":"2606.24667v1","created_at":"2026-06-24T01:15:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.24667","created_at":"2026-06-24T01:15:38Z"},{"alias_kind":"pith_short_12","alias_value":"BSHHHTD377QA","created_at":"2026-06-24T01:15:38Z"},{"alias_kind":"pith_short_16","alias_value":"BSHHHTD377QA3M7L","created_at":"2026-06-24T01:15:38Z"},{"alias_kind":"pith_short_8","alias_value":"BSHHHTD3","created_at":"2026-06-24T01:15:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:BSHHHTD377QA3M7LKDDRL3JU7M","target":"record","payload":{"canonical_record":{"source":{"id":"2606.24667","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-23T15:00:30Z","cross_cats_sorted":[],"title_canon_sha256":"d28ab6c4cb88403dafe09d64dac1d79676ac8d458f58a041155ff1b158187810","abstract_canon_sha256":"4e2ad92c7692677bdb4650fbb7f227dcfaea2e8e37eee48a11b74b5e70c1e70a"},"schema_version":"1.0"},"canonical_sha256":"0c8e73cc7bffe00db3eb50c715ed34fb09b697706e56ba3b03a4548dda284f8f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-24T01:15:38.695843Z","signature_b64":"6vJT/tyj+U/BiwtrSMW6waySz2P054vy9FLAsl47GQAuyZwNu85Zc2JrgQ3HZ7GSy1FIzvyi6Q3sHB0IJ/9xDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0c8e73cc7bffe00db3eb50c715ed34fb09b697706e56ba3b03a4548dda284f8f","last_reissued_at":"2026-06-24T01:15:38.695446Z","signature_status":"signed_v1","first_computed_at":"2026-06-24T01:15:38.695446Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.24667","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":"2026-06-24T01:15:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WOCj87tUNtBsdboRqP57Dn4uD3OE/ioX5VeWP6xNy/TrGk083+G4yy8egB5inZRawUYmB8+C7Tb87znVcgZtAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-27T23:46:06.171538Z"},"content_sha256":"b37f429ea9dd40078712f2f70eafa246f154302b94874be86a69b8144d784d13","schema_version":"1.0","event_id":"sha256:b37f429ea9dd40078712f2f70eafa246f154302b94874be86a69b8144d784d13"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:BSHHHTD377QA3M7LKDDRL3JU7M","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DREAM: Dense Retrieval Embeddings via Autoregressive Modeling","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Yixuan Tang, Yi Yang","submitted_at":"2026-06-23T15:00:30Z","abstract_excerpt":"Dense retrieval embedding models are a fundamental component of modern retrieval-based AI systems. Most dense retrievers are trained with contrastive objectives, which require labeled positive and negative document pairs that are often costly and difficult to obtain. In this work, we investigate whether the autoregressive next-token prediction objective of a large language model (LLM) can provide supervision for dense retrieval. The intuition is simple: if a document contains information relevant to a query, conditioning on that document should make the target output easier for the LLM to pred"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.24667","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/2606.24667/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":"2026-06-24T01:15:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+Jv0cHFY5HHhQJHy2gNeA6+/k2IAcf5PtFF8CvxvR4vc7iV6cO8wDmcitO3tUGua+ZBjIJtJhg/H9L/ZdY6mAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-27T23:46:06.171908Z"},"content_sha256":"2545cee614b500b681514bd471eefc3581fe4691eb4d2b3deb3a238b24ebfb00","schema_version":"1.0","event_id":"sha256:2545cee614b500b681514bd471eefc3581fe4691eb4d2b3deb3a238b24ebfb00"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BSHHHTD377QA3M7LKDDRL3JU7M/bundle.json","state_url":"https://pith.science/pith/BSHHHTD377QA3M7LKDDRL3JU7M/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BSHHHTD377QA3M7LKDDRL3JU7M/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":"2026-06-27T23:46:06Z","links":{"resolver":"https://pith.science/pith/BSHHHTD377QA3M7LKDDRL3JU7M","bundle":"https://pith.science/pith/BSHHHTD377QA3M7LKDDRL3JU7M/bundle.json","state":"https://pith.science/pith/BSHHHTD377QA3M7LKDDRL3JU7M/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BSHHHTD377QA3M7LKDDRL3JU7M/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:BSHHHTD377QA3M7LKDDRL3JU7M","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":"4e2ad92c7692677bdb4650fbb7f227dcfaea2e8e37eee48a11b74b5e70c1e70a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-23T15:00:30Z","title_canon_sha256":"d28ab6c4cb88403dafe09d64dac1d79676ac8d458f58a041155ff1b158187810"},"schema_version":"1.0","source":{"id":"2606.24667","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.24667","created_at":"2026-06-24T01:15:38Z"},{"alias_kind":"arxiv_version","alias_value":"2606.24667v1","created_at":"2026-06-24T01:15:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.24667","created_at":"2026-06-24T01:15:38Z"},{"alias_kind":"pith_short_12","alias_value":"BSHHHTD377QA","created_at":"2026-06-24T01:15:38Z"},{"alias_kind":"pith_short_16","alias_value":"BSHHHTD377QA3M7L","created_at":"2026-06-24T01:15:38Z"},{"alias_kind":"pith_short_8","alias_value":"BSHHHTD3","created_at":"2026-06-24T01:15:38Z"}],"graph_snapshots":[{"event_id":"sha256:2545cee614b500b681514bd471eefc3581fe4691eb4d2b3deb3a238b24ebfb00","target":"graph","created_at":"2026-06-24T01:15:38Z","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/2606.24667/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Dense retrieval embedding models are a fundamental component of modern retrieval-based AI systems. Most dense retrievers are trained with contrastive objectives, which require labeled positive and negative document pairs that are often costly and difficult to obtain. In this work, we investigate whether the autoregressive next-token prediction objective of a large language model (LLM) can provide supervision for dense retrieval. The intuition is simple: if a document contains information relevant to a query, conditioning on that document should make the target output easier for the LLM to pred","authors_text":"Yixuan Tang, Yi Yang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-23T15:00:30Z","title":"DREAM: Dense Retrieval Embeddings via Autoregressive Modeling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.24667","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:b37f429ea9dd40078712f2f70eafa246f154302b94874be86a69b8144d784d13","target":"record","created_at":"2026-06-24T01:15:38Z","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":"4e2ad92c7692677bdb4650fbb7f227dcfaea2e8e37eee48a11b74b5e70c1e70a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-23T15:00:30Z","title_canon_sha256":"d28ab6c4cb88403dafe09d64dac1d79676ac8d458f58a041155ff1b158187810"},"schema_version":"1.0","source":{"id":"2606.24667","kind":"arxiv","version":1}},"canonical_sha256":"0c8e73cc7bffe00db3eb50c715ed34fb09b697706e56ba3b03a4548dda284f8f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0c8e73cc7bffe00db3eb50c715ed34fb09b697706e56ba3b03a4548dda284f8f","first_computed_at":"2026-06-24T01:15:38.695446Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-24T01:15:38.695446Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6vJT/tyj+U/BiwtrSMW6waySz2P054vy9FLAsl47GQAuyZwNu85Zc2JrgQ3HZ7GSy1FIzvyi6Q3sHB0IJ/9xDw==","signature_status":"signed_v1","signed_at":"2026-06-24T01:15:38.695843Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.24667","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b37f429ea9dd40078712f2f70eafa246f154302b94874be86a69b8144d784d13","sha256:2545cee614b500b681514bd471eefc3581fe4691eb4d2b3deb3a238b24ebfb00"],"state_sha256":"cf9e14861c733662f61fe4277b5f38e59ed0c9ccc35016860154682d7766a725"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZNALILcbZywvgw53tiL1WsByugEdnlPK5bXJnsZXEPzvOFbieA6qCQX3wcDpr4MHvVV7v/PFaNVwTxJOQKO8Cw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-27T23:46:06.173842Z","bundle_sha256":"1fc6f1eec24c60ac1c8b977135225045031d2fd89b92f716b6d0b8296176f5a9"}}