{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:BOQ3GXOIQPOGFDRBM6WA62CJ67","short_pith_number":"pith:BOQ3GXOI","schema_version":"1.0","canonical_sha256":"0ba1b35dc883dc628e2167ac0f6849f7d130f898c086000e6b130646878f0617","source":{"kind":"arxiv","id":"2509.22504","version":3},"attestation_state":"computed","paper":{"title":"Estimating the Empowerment of Language Model Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Jeff Gore, Jinyeop Song, Max Kleiman-Weiner","submitted_at":"2025-09-26T15:46:14Z","abstract_excerpt":"As language model (LM) agents become increasingly capable and adopted in real-world applications, there is a growing need for scalable evaluation frameworks beyond costly, manually designed benchmarks. We propose information-theoretic evaluation based on empowerment, an information-theoretic measure of an agent's influence on future states through its actions. To handle the unique challenges of text-based environments, we introduce EELMA (Estimating Empowerment of Language Model Agents), an algorithm for approximating effective empowerment from multi-turn text interactions. We demonstrate EELM"},"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":"2509.22504","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-09-26T15:46:14Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"e16d4946b59f8089f0f32af25fffc4da457b6e3832a53f06640bd8657311a963","abstract_canon_sha256":"8c688d70bd627c408d334e2e941b58d47e2826d826b2833eb20c1734126c602d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T02:05:36.431421Z","signature_b64":"3J+ug37EtT4rxIB15XgekhIYZCc69/lff1Z/YHMoJiZk5XTRuskaDywE4G2kWgMpEJHU+fr9n0FjHNd1WHbqDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0ba1b35dc883dc628e2167ac0f6849f7d130f898c086000e6b130646878f0617","last_reissued_at":"2026-05-29T02:05:36.430982Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T02:05:36.430982Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Estimating the Empowerment of Language Model Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Jeff Gore, Jinyeop Song, Max Kleiman-Weiner","submitted_at":"2025-09-26T15:46:14Z","abstract_excerpt":"As language model (LM) agents become increasingly capable and adopted in real-world applications, there is a growing need for scalable evaluation frameworks beyond costly, manually designed benchmarks. We propose information-theoretic evaluation based on empowerment, an information-theoretic measure of an agent's influence on future states through its actions. To handle the unique challenges of text-based environments, we introduce EELMA (Estimating Empowerment of Language Model Agents), an algorithm for approximating effective empowerment from multi-turn text interactions. We demonstrate EELM"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.22504","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/2509.22504/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":"2509.22504","created_at":"2026-05-29T02:05:36.431040+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.22504v3","created_at":"2026-05-29T02:05:36.431040+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.22504","created_at":"2026-05-29T02:05:36.431040+00:00"},{"alias_kind":"pith_short_12","alias_value":"BOQ3GXOIQPOG","created_at":"2026-05-29T02:05:36.431040+00:00"},{"alias_kind":"pith_short_16","alias_value":"BOQ3GXOIQPOGFDRB","created_at":"2026-05-29T02:05:36.431040+00:00"},{"alias_kind":"pith_short_8","alias_value":"BOQ3GXOI","created_at":"2026-05-29T02:05:36.431040+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/BOQ3GXOIQPOGFDRBM6WA62CJ67","json":"https://pith.science/pith/BOQ3GXOIQPOGFDRBM6WA62CJ67.json","graph_json":"https://pith.science/api/pith-number/BOQ3GXOIQPOGFDRBM6WA62CJ67/graph.json","events_json":"https://pith.science/api/pith-number/BOQ3GXOIQPOGFDRBM6WA62CJ67/events.json","paper":"https://pith.science/paper/BOQ3GXOI"},"agent_actions":{"view_html":"https://pith.science/pith/BOQ3GXOIQPOGFDRBM6WA62CJ67","download_json":"https://pith.science/pith/BOQ3GXOIQPOGFDRBM6WA62CJ67.json","view_paper":"https://pith.science/paper/BOQ3GXOI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.22504&json=true","fetch_graph":"https://pith.science/api/pith-number/BOQ3GXOIQPOGFDRBM6WA62CJ67/graph.json","fetch_events":"https://pith.science/api/pith-number/BOQ3GXOIQPOGFDRBM6WA62CJ67/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BOQ3GXOIQPOGFDRBM6WA62CJ67/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BOQ3GXOIQPOGFDRBM6WA62CJ67/action/storage_attestation","attest_author":"https://pith.science/pith/BOQ3GXOIQPOGFDRBM6WA62CJ67/action/author_attestation","sign_citation":"https://pith.science/pith/BOQ3GXOIQPOGFDRBM6WA62CJ67/action/citation_signature","submit_replication":"https://pith.science/pith/BOQ3GXOIQPOGFDRBM6WA62CJ67/action/replication_record"}},"created_at":"2026-05-29T02:05:36.431040+00:00","updated_at":"2026-05-29T02:05:36.431040+00:00"}