{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ONGBXOJ2VZOTL5MY6ZDIEYZUDH","short_pith_number":"pith:ONGBXOJ2","canonical_record":{"source":{"id":"2501.08248","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-14T16:38:33Z","cross_cats_sorted":["cs.AI","cs.IR","cs.LG"],"title_canon_sha256":"dbd5d1d5f0043cb49678218db7ebf71d35a6f3000babdddfee67ea4be32bf07a","abstract_canon_sha256":"86477bd1e29a14ee6d92df0fca193c49cd984279cbcdb496d66f817084cc6045"},"schema_version":"1.0"},"canonical_sha256":"734c1bb93aae5d35f598f64682633419e25aeba1bd9bdb38087a325880b0e0a0","source":{"kind":"arxiv","id":"2501.08248","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.08248","created_at":"2026-07-05T11:18:17Z"},{"alias_kind":"arxiv_version","alias_value":"2501.08248v3","created_at":"2026-07-05T11:18:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.08248","created_at":"2026-07-05T11:18:17Z"},{"alias_kind":"pith_short_12","alias_value":"ONGBXOJ2VZOT","created_at":"2026-07-05T11:18:17Z"},{"alias_kind":"pith_short_16","alias_value":"ONGBXOJ2VZOTL5MY","created_at":"2026-07-05T11:18:17Z"},{"alias_kind":"pith_short_8","alias_value":"ONGBXOJ2","created_at":"2026-07-05T11:18:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ONGBXOJ2VZOTL5MY6ZDIEYZUDH","target":"record","payload":{"canonical_record":{"source":{"id":"2501.08248","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-14T16:38:33Z","cross_cats_sorted":["cs.AI","cs.IR","cs.LG"],"title_canon_sha256":"dbd5d1d5f0043cb49678218db7ebf71d35a6f3000babdddfee67ea4be32bf07a","abstract_canon_sha256":"86477bd1e29a14ee6d92df0fca193c49cd984279cbcdb496d66f817084cc6045"},"schema_version":"1.0"},"canonical_sha256":"734c1bb93aae5d35f598f64682633419e25aeba1bd9bdb38087a325880b0e0a0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:17.087713Z","signature_b64":"1sbIiyEs/z+7KXX6yvnZfsRQDlW981ckA+0c9zsQBMLW3hcbfflVvvx7wNI+/ddzNYE8YTKFtKTgnkvGXyIXAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"734c1bb93aae5d35f598f64682633419e25aeba1bd9bdb38087a325880b0e0a0","last_reissued_at":"2026-07-05T11:18:17.087170Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:17.087170Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.08248","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:18:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Fg9d/uxpPFAATds/E9n5sLELJEHlikjOF35TaW+/a24RqjraoKQQ/N4T3Ur+IEy/0y4p/cHH1R99W9gpFqtJDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T00:52:52.634873Z"},"content_sha256":"dba340a9cc6d1dfadcd44cff55b32106061b708cbdffe6d869405f0ea747996f","schema_version":"1.0","event_id":"sha256:dba340a9cc6d1dfadcd44cff55b32106061b708cbdffe6d869405f0ea747996f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ONGBXOJ2VZOTL5MY6ZDIEYZUDH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Eliciting In-context Retrieval and Reasoning for Long-context Large Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.IR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Benjamin Han, Navdeep Jaitly, Shay B. Cohen, Varun Embar, Yifu Qiu, Yizhe Zhang","submitted_at":"2025-01-14T16:38:33Z","abstract_excerpt":"Recent advancements in long-context language models (LCLMs) promise to transform Retrieval-Augmented Generation (RAG) by simplifying pipelines. With their expanded context windows, LCLMs can process entire knowledge bases and perform retrieval and reasoning directly -- a capability we define as In-Context Retrieval and Reasoning (ICR^2). However, existing benchmarks like LOFT often overestimate LCLM performance by providing overly simplified contexts. To address this, we introduce ICR^2, a benchmark that evaluates LCLMs in more realistic scenarios by including confounding passages retrieved wi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.08248","kind":"arxiv","version":3},"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/2501.08248/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":"2026-07-05T11:18:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LTR3aIaO6hbYh4HgvG31WoDyr6oLCnsNHiMGe2SyVsKZXK3M5Y20mW5Mg1sKD3xeoPhQYzuV+iUjpGMMwZjdAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T00:52:52.635275Z"},"content_sha256":"b99d9c691466ef7da50bd43930b4fa978a96746b9cd30253d054a633cc67cbab","schema_version":"1.0","event_id":"sha256:b99d9c691466ef7da50bd43930b4fa978a96746b9cd30253d054a633cc67cbab"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ONGBXOJ2VZOTL5MY6ZDIEYZUDH/bundle.json","state_url":"https://pith.science/pith/ONGBXOJ2VZOTL5MY6ZDIEYZUDH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ONGBXOJ2VZOTL5MY6ZDIEYZUDH/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":"2026-08-20T00:52:52Z","links":{"resolver":"https://pith.science/pith/ONGBXOJ2VZOTL5MY6ZDIEYZUDH","bundle":"https://pith.science/pith/ONGBXOJ2VZOTL5MY6ZDIEYZUDH/bundle.json","state":"https://pith.science/pith/ONGBXOJ2VZOTL5MY6ZDIEYZUDH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ONGBXOJ2VZOTL5MY6ZDIEYZUDH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ONGBXOJ2VZOTL5MY6ZDIEYZUDH","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":"86477bd1e29a14ee6d92df0fca193c49cd984279cbcdb496d66f817084cc6045","cross_cats_sorted":["cs.AI","cs.IR","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-14T16:38:33Z","title_canon_sha256":"dbd5d1d5f0043cb49678218db7ebf71d35a6f3000babdddfee67ea4be32bf07a"},"schema_version":"1.0","source":{"id":"2501.08248","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.08248","created_at":"2026-07-05T11:18:17Z"},{"alias_kind":"arxiv_version","alias_value":"2501.08248v3","created_at":"2026-07-05T11:18:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.08248","created_at":"2026-07-05T11:18:17Z"},{"alias_kind":"pith_short_12","alias_value":"ONGBXOJ2VZOT","created_at":"2026-07-05T11:18:17Z"},{"alias_kind":"pith_short_16","alias_value":"ONGBXOJ2VZOTL5MY","created_at":"2026-07-05T11:18:17Z"},{"alias_kind":"pith_short_8","alias_value":"ONGBXOJ2","created_at":"2026-07-05T11:18:17Z"}],"graph_snapshots":[{"event_id":"sha256:b99d9c691466ef7da50bd43930b4fa978a96746b9cd30253d054a633cc67cbab","target":"graph","created_at":"2026-07-05T11:18:17Z","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/2501.08248/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advancements in long-context language models (LCLMs) promise to transform Retrieval-Augmented Generation (RAG) by simplifying pipelines. With their expanded context windows, LCLMs can process entire knowledge bases and perform retrieval and reasoning directly -- a capability we define as In-Context Retrieval and Reasoning (ICR^2). However, existing benchmarks like LOFT often overestimate LCLM performance by providing overly simplified contexts. To address this, we introduce ICR^2, a benchmark that evaluates LCLMs in more realistic scenarios by including confounding passages retrieved wi","authors_text":"Benjamin Han, Navdeep Jaitly, Shay B. Cohen, Varun Embar, Yifu Qiu, Yizhe Zhang","cross_cats":["cs.AI","cs.IR","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-14T16:38:33Z","title":"Eliciting In-context Retrieval and Reasoning for Long-context Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.08248","kind":"arxiv","version":3},"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:dba340a9cc6d1dfadcd44cff55b32106061b708cbdffe6d869405f0ea747996f","target":"record","created_at":"2026-07-05T11:18:17Z","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":"86477bd1e29a14ee6d92df0fca193c49cd984279cbcdb496d66f817084cc6045","cross_cats_sorted":["cs.AI","cs.IR","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-14T16:38:33Z","title_canon_sha256":"dbd5d1d5f0043cb49678218db7ebf71d35a6f3000babdddfee67ea4be32bf07a"},"schema_version":"1.0","source":{"id":"2501.08248","kind":"arxiv","version":3}},"canonical_sha256":"734c1bb93aae5d35f598f64682633419e25aeba1bd9bdb38087a325880b0e0a0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"734c1bb93aae5d35f598f64682633419e25aeba1bd9bdb38087a325880b0e0a0","first_computed_at":"2026-07-05T11:18:17.087170Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:18:17.087170Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1sbIiyEs/z+7KXX6yvnZfsRQDlW981ckA+0c9zsQBMLW3hcbfflVvvx7wNI+/ddzNYE8YTKFtKTgnkvGXyIXAw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:18:17.087713Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.08248","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dba340a9cc6d1dfadcd44cff55b32106061b708cbdffe6d869405f0ea747996f","sha256:b99d9c691466ef7da50bd43930b4fa978a96746b9cd30253d054a633cc67cbab"],"state_sha256":"bbf3479342c01d55773ef75fc79dfa550d56eaf0db5286847daec2f5690f7a19"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lFBrLtj9cm6NJPZquCMLY0EHnr1LcNlRzu5kXhgPqe/1KEbm5fBcE3jHek+S07iFmYu7lX/NUYc+t2JAWyQPDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-20T00:52:52.639292Z","bundle_sha256":"c64e79b07c6d0b5c062aed6c87b4df6c5ff6d2a32672dd4eb40d0e0ece24e4f3"}}