{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:NRPU7DSDS5LDD2XNY6RATMYRSG","short_pith_number":"pith:NRPU7DSD","schema_version":"1.0","canonical_sha256":"6c5f4f8e43975631eaedc7a209b31191a4a20d9fdea7dbc351de662a7ab81745","source":{"kind":"arxiv","id":"2602.08964","version":2},"attestation_state":"computed","paper":{"title":"A Behavioural and Representational Evaluation of Goal-Directedness in Language Model Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CY"],"primary_cat":"cs.LG","authors_text":"Angelos Nalmpantis, Calum McNamara, Evgenii Kortukov, Fade Chen, Gabriele Sarti, Mario Giulianelli, Moksh Nirvaan, Niall Dalton, Raghu Arghal","submitted_at":"2026-02-09T18:00:28Z","abstract_excerpt":"Understanding an agent's goals helps explain and predict its behaviour, yet there is no established methodology for reliably attributing goals to agentic systems. We propose a framework for evaluating goal-directedness that integrates behavioural evaluation with interpretability-based analyses of models' internal representations. As a case study, we examine an LLM agent navigating a 2D grid world towards a goal state. Behaviourally, we evaluate the agent against optimal policies across varying grid sizes, obstacle densities, and goal structures, finding that performance scales with task diffic"},"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":"2602.08964","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-09T18:00:28Z","cross_cats_sorted":["cs.AI","cs.CL","cs.CY"],"title_canon_sha256":"2dccb0ecdf73d37338bfce391c6ff9c3958d528aa95da399ec8188f7263301e6","abstract_canon_sha256":"a10af04004cbaabdd004c3933055d72bd49c84bddb88a423c0ff831a63b9d49f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T02:03:34.695691Z","signature_b64":"7b9mM910VA4Pp/MHAXqvREqCi01yGwd6L3ooO0QHaXfK/jWeDVmzYWttziWNqxrVDrRpph0DX38PwZgIjuPiDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6c5f4f8e43975631eaedc7a209b31191a4a20d9fdea7dbc351de662a7ab81745","last_reissued_at":"2026-06-01T02:03:34.694555Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T02:03:34.694555Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Behavioural and Representational Evaluation of Goal-Directedness in Language Model Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CY"],"primary_cat":"cs.LG","authors_text":"Angelos Nalmpantis, Calum McNamara, Evgenii Kortukov, Fade Chen, Gabriele Sarti, Mario Giulianelli, Moksh Nirvaan, Niall Dalton, Raghu Arghal","submitted_at":"2026-02-09T18:00:28Z","abstract_excerpt":"Understanding an agent's goals helps explain and predict its behaviour, yet there is no established methodology for reliably attributing goals to agentic systems. We propose a framework for evaluating goal-directedness that integrates behavioural evaluation with interpretability-based analyses of models' internal representations. As a case study, we examine an LLM agent navigating a 2D grid world towards a goal state. Behaviourally, we evaluate the agent against optimal policies across varying grid sizes, obstacle densities, and goal structures, finding that performance scales with task diffic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.08964","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/2602.08964/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":"2602.08964","created_at":"2026-06-01T02:03:34.694722+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.08964v2","created_at":"2026-06-01T02:03:34.694722+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.08964","created_at":"2026-06-01T02:03:34.694722+00:00"},{"alias_kind":"pith_short_12","alias_value":"NRPU7DSDS5LD","created_at":"2026-06-01T02:03:34.694722+00:00"},{"alias_kind":"pith_short_16","alias_value":"NRPU7DSDS5LDD2XN","created_at":"2026-06-01T02:03:34.694722+00:00"},{"alias_kind":"pith_short_8","alias_value":"NRPU7DSD","created_at":"2026-06-01T02:03:34.694722+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/NRPU7DSDS5LDD2XNY6RATMYRSG","json":"https://pith.science/pith/NRPU7DSDS5LDD2XNY6RATMYRSG.json","graph_json":"https://pith.science/api/pith-number/NRPU7DSDS5LDD2XNY6RATMYRSG/graph.json","events_json":"https://pith.science/api/pith-number/NRPU7DSDS5LDD2XNY6RATMYRSG/events.json","paper":"https://pith.science/paper/NRPU7DSD"},"agent_actions":{"view_html":"https://pith.science/pith/NRPU7DSDS5LDD2XNY6RATMYRSG","download_json":"https://pith.science/pith/NRPU7DSDS5LDD2XNY6RATMYRSG.json","view_paper":"https://pith.science/paper/NRPU7DSD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.08964&json=true","fetch_graph":"https://pith.science/api/pith-number/NRPU7DSDS5LDD2XNY6RATMYRSG/graph.json","fetch_events":"https://pith.science/api/pith-number/NRPU7DSDS5LDD2XNY6RATMYRSG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NRPU7DSDS5LDD2XNY6RATMYRSG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NRPU7DSDS5LDD2XNY6RATMYRSG/action/storage_attestation","attest_author":"https://pith.science/pith/NRPU7DSDS5LDD2XNY6RATMYRSG/action/author_attestation","sign_citation":"https://pith.science/pith/NRPU7DSDS5LDD2XNY6RATMYRSG/action/citation_signature","submit_replication":"https://pith.science/pith/NRPU7DSDS5LDD2XNY6RATMYRSG/action/replication_record"}},"created_at":"2026-06-01T02:03:34.694722+00:00","updated_at":"2026-06-01T02:03:34.694722+00:00"}