{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:P62JGUG3BAIIBTGCLDBAIGSRLJ","short_pith_number":"pith:P62JGUG3","schema_version":"1.0","canonical_sha256":"7fb49350db081080ccc258c2041a515a48ae77344500613a0d0da2c14da0d8a1","source":{"kind":"arxiv","id":"2508.05614","version":2},"attestation_state":"computed","paper":{"title":"GroundAct: Can LLM Agents Ground Actions in Environmental States?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dingming Li, Hongxing Li, Jun Xiao, Shuo Chen, Weiming Lu, Wenqi Zhang, Yanrui Miao, Yongliang Shen, Yuchen Yan, Yueting Zhuang, Zixuan Wang","submitted_at":"2025-08-07T17:54:15Z","abstract_excerpt":"LLM agents achieve 85-96% success on tasks where instructions fully specify the action, but drop to 29-53% when action feasibility depends on environmental state that the instruction does not mention. We argue that this gap reflects a missing capability: action grounding, the ability to infer from structured environmental state whether an action is feasible, what prerequisites it lacks, and whether it exceeds individual capacity. We introduce GroundAct, a benchmark of 1,500 scenarios and 16,592 task instances in text-based interactive environments spanning 11 domains, with tasks organized into"},"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":"2508.05614","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-07T17:54:15Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4085dfd07fd7068b520dd368c246daeb935ac58dfcceb30aa7a3c74e31e4a747","abstract_canon_sha256":"ece689838301ae5f703a3ed1b44c82709398d9ad477fdb744a81fcd950bd1374"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:04:54.844438Z","signature_b64":"/PvBedmMQlENtxdPsPpijcboKm+t5Svb6LrvKd9S7ltPGW9tT4yGmGbYuzcu52rfZrczqQc9dPMEZm4c7o8tDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7fb49350db081080ccc258c2041a515a48ae77344500613a0d0da2c14da0d8a1","last_reissued_at":"2026-05-29T01:04:54.843938Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:04:54.843938Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GroundAct: Can LLM Agents Ground Actions in Environmental States?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dingming Li, Hongxing Li, Jun Xiao, Shuo Chen, Weiming Lu, Wenqi Zhang, Yanrui Miao, Yongliang Shen, Yuchen Yan, Yueting Zhuang, Zixuan Wang","submitted_at":"2025-08-07T17:54:15Z","abstract_excerpt":"LLM agents achieve 85-96% success on tasks where instructions fully specify the action, but drop to 29-53% when action feasibility depends on environmental state that the instruction does not mention. We argue that this gap reflects a missing capability: action grounding, the ability to infer from structured environmental state whether an action is feasible, what prerequisites it lacks, and whether it exceeds individual capacity. We introduce GroundAct, a benchmark of 1,500 scenarios and 16,592 task instances in text-based interactive environments spanning 11 domains, with tasks organized into"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.05614","kind":"arxiv","version":2},"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/2508.05614/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":"2508.05614","created_at":"2026-05-29T01:04:54.844000+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.05614v2","created_at":"2026-05-29T01:04:54.844000+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.05614","created_at":"2026-05-29T01:04:54.844000+00:00"},{"alias_kind":"pith_short_12","alias_value":"P62JGUG3BAII","created_at":"2026-05-29T01:04:54.844000+00:00"},{"alias_kind":"pith_short_16","alias_value":"P62JGUG3BAIIBTGC","created_at":"2026-05-29T01:04:54.844000+00:00"},{"alias_kind":"pith_short_8","alias_value":"P62JGUG3","created_at":"2026-05-29T01:04:54.844000+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.06078","citing_title":"Milestone-Guided Policy Learning for Long-Horizon Language Agents","ref_index":48,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P62JGUG3BAIIBTGCLDBAIGSRLJ","json":"https://pith.science/pith/P62JGUG3BAIIBTGCLDBAIGSRLJ.json","graph_json":"https://pith.science/api/pith-number/P62JGUG3BAIIBTGCLDBAIGSRLJ/graph.json","events_json":"https://pith.science/api/pith-number/P62JGUG3BAIIBTGCLDBAIGSRLJ/events.json","paper":"https://pith.science/paper/P62JGUG3"},"agent_actions":{"view_html":"https://pith.science/pith/P62JGUG3BAIIBTGCLDBAIGSRLJ","download_json":"https://pith.science/pith/P62JGUG3BAIIBTGCLDBAIGSRLJ.json","view_paper":"https://pith.science/paper/P62JGUG3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.05614&json=true","fetch_graph":"https://pith.science/api/pith-number/P62JGUG3BAIIBTGCLDBAIGSRLJ/graph.json","fetch_events":"https://pith.science/api/pith-number/P62JGUG3BAIIBTGCLDBAIGSRLJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P62JGUG3BAIIBTGCLDBAIGSRLJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P62JGUG3BAIIBTGCLDBAIGSRLJ/action/storage_attestation","attest_author":"https://pith.science/pith/P62JGUG3BAIIBTGCLDBAIGSRLJ/action/author_attestation","sign_citation":"https://pith.science/pith/P62JGUG3BAIIBTGCLDBAIGSRLJ/action/citation_signature","submit_replication":"https://pith.science/pith/P62JGUG3BAIIBTGCLDBAIGSRLJ/action/replication_record"}},"created_at":"2026-05-29T01:04:54.844000+00:00","updated_at":"2026-05-29T01:04:54.844000+00:00"}