{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4NAAIVRQR37XUDSVH6SVVAJDIE","short_pith_number":"pith:4NAAIVRQ","schema_version":"1.0","canonical_sha256":"e3400456308eff7a0e553fa55a8123411aa3c26e08ae0895dda394a15dae3589","source":{"kind":"arxiv","id":"2605.24539","version":1},"attestation_state":"computed","paper":{"title":"DemoEvolve: Overcoming Sparse Feedback in Agentic Harness Evolution with Demonstrations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chuang Wang, Jian Su, Lirong Che, Peiwen Lin, Xueqian Wang, Yuzhe Yang","submitted_at":"2026-05-23T12:10:52Z","abstract_excerpt":"Agent harness evolution improves frozen language-model agents by modifying the executable structures around them. We study this paradigm as a form of sample-efficient fast adaptation: instead of updating model weights, an agent can acquire task-specific competence by changing its external harness, while leaving the base model's general capabilities intact. Prior work shows that self-generated rollouts can support harness search, suggesting that agents may acquire new task competence through practice. Yet in long-horizon stochastic environments, self-practice becomes fragile: rewards are sparse"},"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":"2605.24539","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-23T12:10:52Z","cross_cats_sorted":[],"title_canon_sha256":"774f16cbff730297077718d88ec9c91729e0c19049568502c8d4ca0599fb9162","abstract_canon_sha256":"886220316e378775f333694e2f548ceaf38e38615248837896cf604cc4fc010b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:03:45.332466Z","signature_b64":"UcNyvoBdCpwSTI5teSbp80uKqx8yqCDW+UanCklzdgSKUIovzHCd1g1CPFrCJwRUB3yFaE77HbLZTYQdq8n4BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e3400456308eff7a0e553fa55a8123411aa3c26e08ae0895dda394a15dae3589","last_reissued_at":"2026-05-26T01:03:45.331691Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:03:45.331691Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DemoEvolve: Overcoming Sparse Feedback in Agentic Harness Evolution with Demonstrations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chuang Wang, Jian Su, Lirong Che, Peiwen Lin, Xueqian Wang, Yuzhe Yang","submitted_at":"2026-05-23T12:10:52Z","abstract_excerpt":"Agent harness evolution improves frozen language-model agents by modifying the executable structures around them. We study this paradigm as a form of sample-efficient fast adaptation: instead of updating model weights, an agent can acquire task-specific competence by changing its external harness, while leaving the base model's general capabilities intact. Prior work shows that self-generated rollouts can support harness search, suggesting that agents may acquire new task competence through practice. Yet in long-horizon stochastic environments, self-practice becomes fragile: rewards are sparse"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24539","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/2605.24539/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":"2605.24539","created_at":"2026-05-26T01:03:45.331811+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.24539v1","created_at":"2026-05-26T01:03:45.331811+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24539","created_at":"2026-05-26T01:03:45.331811+00:00"},{"alias_kind":"pith_short_12","alias_value":"4NAAIVRQR37X","created_at":"2026-05-26T01:03:45.331811+00:00"},{"alias_kind":"pith_short_16","alias_value":"4NAAIVRQR37XUDSV","created_at":"2026-05-26T01:03:45.331811+00:00"},{"alias_kind":"pith_short_8","alias_value":"4NAAIVRQ","created_at":"2026-05-26T01:03:45.331811+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/4NAAIVRQR37XUDSVH6SVVAJDIE","json":"https://pith.science/pith/4NAAIVRQR37XUDSVH6SVVAJDIE.json","graph_json":"https://pith.science/api/pith-number/4NAAIVRQR37XUDSVH6SVVAJDIE/graph.json","events_json":"https://pith.science/api/pith-number/4NAAIVRQR37XUDSVH6SVVAJDIE/events.json","paper":"https://pith.science/paper/4NAAIVRQ"},"agent_actions":{"view_html":"https://pith.science/pith/4NAAIVRQR37XUDSVH6SVVAJDIE","download_json":"https://pith.science/pith/4NAAIVRQR37XUDSVH6SVVAJDIE.json","view_paper":"https://pith.science/paper/4NAAIVRQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.24539&json=true","fetch_graph":"https://pith.science/api/pith-number/4NAAIVRQR37XUDSVH6SVVAJDIE/graph.json","fetch_events":"https://pith.science/api/pith-number/4NAAIVRQR37XUDSVH6SVVAJDIE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4NAAIVRQR37XUDSVH6SVVAJDIE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4NAAIVRQR37XUDSVH6SVVAJDIE/action/storage_attestation","attest_author":"https://pith.science/pith/4NAAIVRQR37XUDSVH6SVVAJDIE/action/author_attestation","sign_citation":"https://pith.science/pith/4NAAIVRQR37XUDSVH6SVVAJDIE/action/citation_signature","submit_replication":"https://pith.science/pith/4NAAIVRQR37XUDSVH6SVVAJDIE/action/replication_record"}},"created_at":"2026-05-26T01:03:45.331811+00:00","updated_at":"2026-05-26T01:03:45.331811+00:00"}