{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:Y6R5YAJGUGKOXMHNNROYLEOI3Q","short_pith_number":"pith:Y6R5YAJG","schema_version":"1.0","canonical_sha256":"c7a3dc0126a194ebb0ed6c5d8591c8dc3823605f8095b5819ec447639d3d1abb","source":{"kind":"arxiv","id":"2606.11652","version":1},"attestation_state":"computed","paper":{"title":"IAPO: Input Attribution-Aware Policy Optimization for Tool Use in Small Multimodal Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jiayi Tian, Liyan Tan, Yifan Yang, Zheng Zhang, Zhen Zhang","submitted_at":"2026-06-10T04:30:37Z","abstract_excerpt":"This paper investigates reinforcement learning (RL) methods for improving tool-calling capabilities in multimodal small language model (SLM) agents. While existing works have explored various reward designs to improve agentic tool-calling ability, these approaches face inherent limitations for SLM training, especially under multimodal scenarios. First, many existing methods evaluate tool use correctness through exact matching against certain ground-truth or predefined formats. However, this assumption is often unsuitable for multimodal tasks, where multiple tool use paths may be valid and anno"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.11652","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-10T04:30:37Z","cross_cats_sorted":[],"title_canon_sha256":"dc59a7556dd44e2a2a3d2e442866753f1373051c94e546e2109b1114e9f99ad9","abstract_canon_sha256":"4f2eff9a0678b8f57d4584804b90d960cd317ba2f72b3a665f42742f2755c6e6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-11T01:10:01.091181Z","signature_b64":"JLvGjIouyPfUSwxRQsbsJ3wbQaHU+NBZmG1AdR+CdvLWIsXURZ0VgPaZf8xv9+ukK/lq7KeKQp1dRBL8XFNHBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c7a3dc0126a194ebb0ed6c5d8591c8dc3823605f8095b5819ec447639d3d1abb","last_reissued_at":"2026-06-11T01:10:01.090448Z","signature_status":"signed_v1","first_computed_at":"2026-06-11T01:10:01.090448Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"IAPO: Input Attribution-Aware Policy Optimization for Tool Use in Small Multimodal Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jiayi Tian, Liyan Tan, Yifan Yang, Zheng Zhang, Zhen Zhang","submitted_at":"2026-06-10T04:30:37Z","abstract_excerpt":"This paper investigates reinforcement learning (RL) methods for improving tool-calling capabilities in multimodal small language model (SLM) agents. While existing works have explored various reward designs to improve agentic tool-calling ability, these approaches face inherent limitations for SLM training, especially under multimodal scenarios. First, many existing methods evaluate tool use correctness through exact matching against certain ground-truth or predefined formats. However, this assumption is often unsuitable for multimodal tasks, where multiple tool use paths may be valid and anno"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.11652","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/2606.11652/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.11652","created_at":"2026-06-11T01:10:01.090577+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.11652v1","created_at":"2026-06-11T01:10:01.090577+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.11652","created_at":"2026-06-11T01:10:01.090577+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y6R5YAJGUGKO","created_at":"2026-06-11T01:10:01.090577+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y6R5YAJGUGKOXMHN","created_at":"2026-06-11T01:10:01.090577+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y6R5YAJG","created_at":"2026-06-11T01:10:01.090577+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Y6R5YAJGUGKOXMHNNROYLEOI3Q","json":"https://pith.science/pith/Y6R5YAJGUGKOXMHNNROYLEOI3Q.json","graph_json":"https://pith.science/api/pith-number/Y6R5YAJGUGKOXMHNNROYLEOI3Q/graph.json","events_json":"https://pith.science/api/pith-number/Y6R5YAJGUGKOXMHNNROYLEOI3Q/events.json","paper":"https://pith.science/paper/Y6R5YAJG"},"agent_actions":{"view_html":"https://pith.science/pith/Y6R5YAJGUGKOXMHNNROYLEOI3Q","download_json":"https://pith.science/pith/Y6R5YAJGUGKOXMHNNROYLEOI3Q.json","view_paper":"https://pith.science/paper/Y6R5YAJG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.11652&json=true","fetch_graph":"https://pith.science/api/pith-number/Y6R5YAJGUGKOXMHNNROYLEOI3Q/graph.json","fetch_events":"https://pith.science/api/pith-number/Y6R5YAJGUGKOXMHNNROYLEOI3Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y6R5YAJGUGKOXMHNNROYLEOI3Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y6R5YAJGUGKOXMHNNROYLEOI3Q/action/storage_attestation","attest_author":"https://pith.science/pith/Y6R5YAJGUGKOXMHNNROYLEOI3Q/action/author_attestation","sign_citation":"https://pith.science/pith/Y6R5YAJGUGKOXMHNNROYLEOI3Q/action/citation_signature","submit_replication":"https://pith.science/pith/Y6R5YAJGUGKOXMHNNROYLEOI3Q/action/replication_record"}},"created_at":"2026-06-11T01:10:01.090577+00:00","updated_at":"2026-06-11T01:10:01.090577+00:00"}