{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:UFG4QA677KTFAZC2MQV7JAVO2O","short_pith_number":"pith:UFG4QA67","canonical_record":{"source":{"id":"2607.06223","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-07T12:47:23Z","cross_cats_sorted":[],"title_canon_sha256":"1d04a696dc868d1021a5d8e0971178174440e58885a13a9a8b6099aaef6ad31f","abstract_canon_sha256":"890039b03550cd0ab2397efff3035261f9254ae0ba9fc8bd2c4fd958bc554463"},"schema_version":"1.0"},"canonical_sha256":"a14dc803dffaa650645a642bf482aed385e240e16749b38d3ec7ae2387c002e3","source":{"kind":"arxiv","id":"2607.06223","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.06223","created_at":"2026-07-08T01:19:18Z"},{"alias_kind":"arxiv_version","alias_value":"2607.06223v1","created_at":"2026-07-08T01:19:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.06223","created_at":"2026-07-08T01:19:18Z"},{"alias_kind":"pith_short_12","alias_value":"UFG4QA677KTF","created_at":"2026-07-08T01:19:18Z"},{"alias_kind":"pith_short_16","alias_value":"UFG4QA677KTFAZC2","created_at":"2026-07-08T01:19:18Z"},{"alias_kind":"pith_short_8","alias_value":"UFG4QA67","created_at":"2026-07-08T01:19:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:UFG4QA677KTFAZC2MQV7JAVO2O","target":"record","payload":{"canonical_record":{"source":{"id":"2607.06223","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-07T12:47:23Z","cross_cats_sorted":[],"title_canon_sha256":"1d04a696dc868d1021a5d8e0971178174440e58885a13a9a8b6099aaef6ad31f","abstract_canon_sha256":"890039b03550cd0ab2397efff3035261f9254ae0ba9fc8bd2c4fd958bc554463"},"schema_version":"1.0"},"canonical_sha256":"a14dc803dffaa650645a642bf482aed385e240e16749b38d3ec7ae2387c002e3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-08T01:19:18.269157Z","signature_b64":"c9hl67ZLKos/hYoQQPv+3KyNzG9wRAWVvslInsgVuwMiWQ448IB/OELsfnWteRJUAoEeLtnkuLCggTGZGbBQAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a14dc803dffaa650645a642bf482aed385e240e16749b38d3ec7ae2387c002e3","last_reissued_at":"2026-07-08T01:19:18.268744Z","signature_status":"signed_v1","first_computed_at":"2026-07-08T01:19:18.268744Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.06223","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":"2026-07-08T01:19:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qiIB2uUKma4LwghQ0YVDaZPB+lh5VTfqHBJ2GJ31+3aWEw01jJryS/UZcPtt8Sb5YlcIRwu+mtX2UtxqFUGPCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-02T05:04:17.147523Z"},"content_sha256":"7605594a863be615ac36f6544958314f1cf5b891f9e933c8cf7c96f1a548b671","schema_version":"1.0","event_id":"sha256:7605594a863be615ac36f6544958314f1cf5b891f9e933c8cf7c96f1a548b671"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:UFG4QA677KTFAZC2MQV7JAVO2O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Information Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn LLM Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Fan Xu, Haoxiang Zhang, Jiaxin Ding, Luoyi Fu, Shiqing Gao, Xinbing Wang, Xin Ding, Yijun Zhang, Yule Xie","submitted_at":"2026-07-07T12:47:23Z","abstract_excerpt":"Reinforcement learning has become a promising paradigm for improving large language model (LLM) agents on long-horizon search tasks, where the agent must make a sequence of intermediate decisions before receiving a final outcome. However, existing methods still face a key limitation: the rollout budget is often allocated without explicitly assessing the utility of intermediate states. As a result, substantial computation may be spent on low-value states, even though different branches can vary drastically in their informativeness. In this paper, we propose Information Gain-based Rollout Policy"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.06223","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/2607.06223/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":"2026-07-08T01:19:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UyYNE6u2bwT+mjyZZL+rlFWd2sO7uayoGonlbq+j6MMpXiu1wh64TlkNGjp+dr6LAr1oT/WAWy/uTrb4wB9QAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-02T05:04:17.148055Z"},"content_sha256":"24461b9dde5eff64ec2120270654b768df5f59f4e50c5fbc77707bab305de84d","schema_version":"1.0","event_id":"sha256:24461b9dde5eff64ec2120270654b768df5f59f4e50c5fbc77707bab305de84d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UFG4QA677KTFAZC2MQV7JAVO2O/bundle.json","state_url":"https://pith.science/pith/UFG4QA677KTFAZC2MQV7JAVO2O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UFG4QA677KTFAZC2MQV7JAVO2O/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":"2026-08-02T05:04:17Z","links":{"resolver":"https://pith.science/pith/UFG4QA677KTFAZC2MQV7JAVO2O","bundle":"https://pith.science/pith/UFG4QA677KTFAZC2MQV7JAVO2O/bundle.json","state":"https://pith.science/pith/UFG4QA677KTFAZC2MQV7JAVO2O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UFG4QA677KTFAZC2MQV7JAVO2O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:UFG4QA677KTFAZC2MQV7JAVO2O","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":"890039b03550cd0ab2397efff3035261f9254ae0ba9fc8bd2c4fd958bc554463","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-07T12:47:23Z","title_canon_sha256":"1d04a696dc868d1021a5d8e0971178174440e58885a13a9a8b6099aaef6ad31f"},"schema_version":"1.0","source":{"id":"2607.06223","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.06223","created_at":"2026-07-08T01:19:18Z"},{"alias_kind":"arxiv_version","alias_value":"2607.06223v1","created_at":"2026-07-08T01:19:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.06223","created_at":"2026-07-08T01:19:18Z"},{"alias_kind":"pith_short_12","alias_value":"UFG4QA677KTF","created_at":"2026-07-08T01:19:18Z"},{"alias_kind":"pith_short_16","alias_value":"UFG4QA677KTFAZC2","created_at":"2026-07-08T01:19:18Z"},{"alias_kind":"pith_short_8","alias_value":"UFG4QA67","created_at":"2026-07-08T01:19:18Z"}],"graph_snapshots":[{"event_id":"sha256:24461b9dde5eff64ec2120270654b768df5f59f4e50c5fbc77707bab305de84d","target":"graph","created_at":"2026-07-08T01:19:18Z","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/2607.06223/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning has become a promising paradigm for improving large language model (LLM) agents on long-horizon search tasks, where the agent must make a sequence of intermediate decisions before receiving a final outcome. However, existing methods still face a key limitation: the rollout budget is often allocated without explicitly assessing the utility of intermediate states. As a result, substantial computation may be spent on low-value states, even though different branches can vary drastically in their informativeness. In this paper, we propose Information Gain-based Rollout Policy","authors_text":"Fan Xu, Haoxiang Zhang, Jiaxin Ding, Luoyi Fu, Shiqing Gao, Xinbing Wang, Xin Ding, Yijun Zhang, Yule Xie","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-07T12:47:23Z","title":"Information Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn LLM Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.06223","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":"sha256:7605594a863be615ac36f6544958314f1cf5b891f9e933c8cf7c96f1a548b671","target":"record","created_at":"2026-07-08T01:19:18Z","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":"890039b03550cd0ab2397efff3035261f9254ae0ba9fc8bd2c4fd958bc554463","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-07T12:47:23Z","title_canon_sha256":"1d04a696dc868d1021a5d8e0971178174440e58885a13a9a8b6099aaef6ad31f"},"schema_version":"1.0","source":{"id":"2607.06223","kind":"arxiv","version":1}},"canonical_sha256":"a14dc803dffaa650645a642bf482aed385e240e16749b38d3ec7ae2387c002e3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a14dc803dffaa650645a642bf482aed385e240e16749b38d3ec7ae2387c002e3","first_computed_at":"2026-07-08T01:19:18.268744Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-08T01:19:18.268744Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"c9hl67ZLKos/hYoQQPv+3KyNzG9wRAWVvslInsgVuwMiWQ448IB/OELsfnWteRJUAoEeLtnkuLCggTGZGbBQAQ==","signature_status":"signed_v1","signed_at":"2026-07-08T01:19:18.269157Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.06223","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7605594a863be615ac36f6544958314f1cf5b891f9e933c8cf7c96f1a548b671","sha256:24461b9dde5eff64ec2120270654b768df5f59f4e50c5fbc77707bab305de84d"],"state_sha256":"b07ad04150457b528c1dd21d7c96d0e721156a496cb0cecd69e3180f72925bdb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zH7aeSa+FfTNgtTLZxAJiewGruRls4u6PfrbStcj48cbqE2fBcNb/WTCLeeSyBUt2EYb5r9zaVXbAcH0oJdFBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-02T05:04:17.167820Z","bundle_sha256":"3b7639a60708bcc0323f6df91ec7d773cd6dcf63e64eb5d6eab850537d93d4bf"}}