{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:WTGS2XBT6O667PVFMJRJZEHTRJ","short_pith_number":"pith:WTGS2XBT","canonical_record":{"source":{"id":"2505.16944","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-22T17:31:10Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"196023b7dfd8f46eed5a838bd61f6e6e62848577f42d1c1cf624d6cebc9e9074","abstract_canon_sha256":"8cd0de66b6fa56f28768ae3d50ab383e1b0d7d79c494bfb69b3434591d437c4a"},"schema_version":"1.0"},"canonical_sha256":"b4cd2d5c33f3bdefbea562629c90f38a53739f5c30b496a5c41bf01dc9236820","source":{"kind":"arxiv","id":"2505.16944","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.16944","created_at":"2026-07-05T11:07:43Z"},{"alias_kind":"arxiv_version","alias_value":"2505.16944v1","created_at":"2026-07-05T11:07:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.16944","created_at":"2026-07-05T11:07:43Z"},{"alias_kind":"pith_short_12","alias_value":"WTGS2XBT6O66","created_at":"2026-07-05T11:07:43Z"},{"alias_kind":"pith_short_16","alias_value":"WTGS2XBT6O667PVF","created_at":"2026-07-05T11:07:43Z"},{"alias_kind":"pith_short_8","alias_value":"WTGS2XBT","created_at":"2026-07-05T11:07:43Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:WTGS2XBT6O667PVFMJRJZEHTRJ","target":"record","payload":{"canonical_record":{"source":{"id":"2505.16944","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-22T17:31:10Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"196023b7dfd8f46eed5a838bd61f6e6e62848577f42d1c1cf624d6cebc9e9074","abstract_canon_sha256":"8cd0de66b6fa56f28768ae3d50ab383e1b0d7d79c494bfb69b3434591d437c4a"},"schema_version":"1.0"},"canonical_sha256":"b4cd2d5c33f3bdefbea562629c90f38a53739f5c30b496a5c41bf01dc9236820","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:07:43.639968Z","signature_b64":"sef8Unam8JfzblQgQGD8YJUyai4J3AJlys9dxue37Xsi5jmxAc3hi07JmdqDve6UeaR1UXGLHHS8La6+WhU9Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b4cd2d5c33f3bdefbea562629c90f38a53739f5c30b496a5c41bf01dc9236820","last_reissued_at":"2026-07-05T11:07:43.639431Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:07:43.639431Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.16944","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":"2026-07-05T11:07:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tcuur8BcY1KMiU3CScVKutSRD9YOQgw5WwweooFnHmfjc09jvuJyJI8Zbymiml2K/FAYryAq4fRnahVse8KnBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T07:15:12.749126Z"},"content_sha256":"390053e13b80136715465b5bfd709933ae6f1ad32d6f7a70d820c69be8759547","schema_version":"1.0","event_id":"sha256:390053e13b80136715465b5bfd709933ae6f1ad32d6f7a70d820c69be8759547"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:WTGS2XBT6O667PVFMJRJZEHTRJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"AGENTIF: Benchmarking Instruction Following of Large Language Models in Agentic Scenarios","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Amy Xin, Bin Xu, Hao Peng, Juanzi Li, Lei Hou, Xiaozhi Wang, Youfeng Liu, Yunjia Qi","submitted_at":"2025-05-22T17:31:10Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated advanced capabilities in real-world agentic applications. Growing research efforts aim to develop LLM-based agents to address practical demands, introducing a new challenge: agentic scenarios often involve lengthy instructions with complex constraints, such as extended system prompts and detailed tool specifications. While adherence to such instructions is crucial for agentic applications, whether LLMs can reliably follow them remains underexplored. In this paper, we introduce AgentIF, the first benchmark for systematically evaluating LLM instruct"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.16944","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/2505.16944/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":"2026-07-05T11:07:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"c0SAGGpgtWJeM4myBqZIlThg3TranGvxPRpwIvlaI96ZGqwLunhwV7xtlpeZNl18mnOz/oe9txoY+sdkAH10BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T07:15:12.749829Z"},"content_sha256":"463018bdf1321a597ab03b5de5bfaeaebc08fab01b17e20c3ab803b997aa7e98","schema_version":"1.0","event_id":"sha256:463018bdf1321a597ab03b5de5bfaeaebc08fab01b17e20c3ab803b997aa7e98"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WTGS2XBT6O667PVFMJRJZEHTRJ/bundle.json","state_url":"https://pith.science/pith/WTGS2XBT6O667PVFMJRJZEHTRJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WTGS2XBT6O667PVFMJRJZEHTRJ/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":"2026-07-07T07:15:12Z","links":{"resolver":"https://pith.science/pith/WTGS2XBT6O667PVFMJRJZEHTRJ","bundle":"https://pith.science/pith/WTGS2XBT6O667PVFMJRJZEHTRJ/bundle.json","state":"https://pith.science/pith/WTGS2XBT6O667PVFMJRJZEHTRJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WTGS2XBT6O667PVFMJRJZEHTRJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:WTGS2XBT6O667PVFMJRJZEHTRJ","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":"8cd0de66b6fa56f28768ae3d50ab383e1b0d7d79c494bfb69b3434591d437c4a","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-22T17:31:10Z","title_canon_sha256":"196023b7dfd8f46eed5a838bd61f6e6e62848577f42d1c1cf624d6cebc9e9074"},"schema_version":"1.0","source":{"id":"2505.16944","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.16944","created_at":"2026-07-05T11:07:43Z"},{"alias_kind":"arxiv_version","alias_value":"2505.16944v1","created_at":"2026-07-05T11:07:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.16944","created_at":"2026-07-05T11:07:43Z"},{"alias_kind":"pith_short_12","alias_value":"WTGS2XBT6O66","created_at":"2026-07-05T11:07:43Z"},{"alias_kind":"pith_short_16","alias_value":"WTGS2XBT6O667PVF","created_at":"2026-07-05T11:07:43Z"},{"alias_kind":"pith_short_8","alias_value":"WTGS2XBT","created_at":"2026-07-05T11:07:43Z"}],"graph_snapshots":[{"event_id":"sha256:463018bdf1321a597ab03b5de5bfaeaebc08fab01b17e20c3ab803b997aa7e98","target":"graph","created_at":"2026-07-05T11:07:43Z","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/2505.16944/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have demonstrated advanced capabilities in real-world agentic applications. Growing research efforts aim to develop LLM-based agents to address practical demands, introducing a new challenge: agentic scenarios often involve lengthy instructions with complex constraints, such as extended system prompts and detailed tool specifications. While adherence to such instructions is crucial for agentic applications, whether LLMs can reliably follow them remains underexplored. In this paper, we introduce AgentIF, the first benchmark for systematically evaluating LLM instruct","authors_text":"Amy Xin, Bin Xu, Hao Peng, Juanzi Li, Lei Hou, Xiaozhi Wang, Youfeng Liu, Yunjia Qi","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-22T17:31:10Z","title":"AGENTIF: Benchmarking Instruction Following of Large Language Models in Agentic Scenarios"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.16944","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":"sha256:390053e13b80136715465b5bfd709933ae6f1ad32d6f7a70d820c69be8759547","target":"record","created_at":"2026-07-05T11:07:43Z","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":"8cd0de66b6fa56f28768ae3d50ab383e1b0d7d79c494bfb69b3434591d437c4a","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-22T17:31:10Z","title_canon_sha256":"196023b7dfd8f46eed5a838bd61f6e6e62848577f42d1c1cf624d6cebc9e9074"},"schema_version":"1.0","source":{"id":"2505.16944","kind":"arxiv","version":1}},"canonical_sha256":"b4cd2d5c33f3bdefbea562629c90f38a53739f5c30b496a5c41bf01dc9236820","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b4cd2d5c33f3bdefbea562629c90f38a53739f5c30b496a5c41bf01dc9236820","first_computed_at":"2026-07-05T11:07:43.639431Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:07:43.639431Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"sef8Unam8JfzblQgQGD8YJUyai4J3AJlys9dxue37Xsi5jmxAc3hi07JmdqDve6UeaR1UXGLHHS8La6+WhU9Ag==","signature_status":"signed_v1","signed_at":"2026-07-05T11:07:43.639968Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.16944","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:390053e13b80136715465b5bfd709933ae6f1ad32d6f7a70d820c69be8759547","sha256:463018bdf1321a597ab03b5de5bfaeaebc08fab01b17e20c3ab803b997aa7e98"],"state_sha256":"838401b700f30bdf82a0bc5d66af2403c86e72882f363cb9df4d033a1bd9e5b0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"343K2lFFA6exkWxVea8xsPiAdnjOUQ4w/jpVxfKSq43/H3ycsdf9JAe0Iio5u7nnnsYtU19Lm7CwBzP0hvFyCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T07:15:12.754066Z","bundle_sha256":"c343afcd76239e1f3c66c68d22fb7149f8403a11380f1fd65b87808cba2ee9e1"}}