{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KDIVL6T445NZ25YUDAJ6PCYGQV","short_pith_number":"pith:KDIVL6T4","schema_version":"1.0","canonical_sha256":"50d155fa7ce75b9d77141813e78b068569982c121aafd81f62cf5bdbecb576a5","source":{"kind":"arxiv","id":"2502.07942","version":2},"attestation_state":"computed","paper":{"title":"Symbiotic Cooperation for Web Agents: Harnessing Complementary Strengths of Large and Small LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.MA","authors_text":"Jie Peng, Kaidi Xu, Leandro Z. Agudelo, Mohan Zhang, Mufan Qiu, Peter Qian, Ruichen Zhang, Tianlong Chen, Vincent Lu, Zhen Tan","submitted_at":"2025-02-11T20:41:49Z","abstract_excerpt":"Web browsing agents powered by large language models (LLMs) have shown tremendous potential in automating complex web-based tasks. Existing approaches typically rely on large LLMs (e.g., GPT-4o) to explore web environments and generate trajectory data, which is then used either for demonstration retrieval (for large LLMs) or to distill small LLMs (e.g., Llama3) in a process that remains decoupled from the exploration. In this paper, we propose AgentSymbiotic, an iterative framework that couples data synthesis with task-performance, yielding a \"symbiotic improvement\" for both large and small LL"},"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":"2502.07942","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MA","submitted_at":"2025-02-11T20:41:49Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1dd1df966126e4d66ef55f40e106c7c5c3639342aede397614c7688ead5582a5","abstract_canon_sha256":"a01dc87f6637dad4260df70216f367a2e213ea749691d2be0b688191bd3175fa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:25:47.243989Z","signature_b64":"bT3gjmX6BCmWhdv61kwSgwI6bOpO+HAukv4BhFqR0ygVSCZFZqb38LMha2xGo+5IIasZZY70W+XqMwA7a2IXBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"50d155fa7ce75b9d77141813e78b068569982c121aafd81f62cf5bdbecb576a5","last_reissued_at":"2026-07-05T10:25:47.243386Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:25:47.243386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Symbiotic Cooperation for Web Agents: Harnessing Complementary Strengths of Large and Small LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.MA","authors_text":"Jie Peng, Kaidi Xu, Leandro Z. Agudelo, Mohan Zhang, Mufan Qiu, Peter Qian, Ruichen Zhang, Tianlong Chen, Vincent Lu, Zhen Tan","submitted_at":"2025-02-11T20:41:49Z","abstract_excerpt":"Web browsing agents powered by large language models (LLMs) have shown tremendous potential in automating complex web-based tasks. Existing approaches typically rely on large LLMs (e.g., GPT-4o) to explore web environments and generate trajectory data, which is then used either for demonstration retrieval (for large LLMs) or to distill small LLMs (e.g., Llama3) in a process that remains decoupled from the exploration. In this paper, we propose AgentSymbiotic, an iterative framework that couples data synthesis with task-performance, yielding a \"symbiotic improvement\" for both large and small LL"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.07942","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/2502.07942/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":"2502.07942","created_at":"2026-07-05T10:25:47.243461+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.07942v2","created_at":"2026-07-05T10:25:47.243461+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.07942","created_at":"2026-07-05T10:25:47.243461+00:00"},{"alias_kind":"pith_short_12","alias_value":"KDIVL6T445NZ","created_at":"2026-07-05T10:25:47.243461+00:00"},{"alias_kind":"pith_short_16","alias_value":"KDIVL6T445NZ25YU","created_at":"2026-07-05T10:25:47.243461+00:00"},{"alias_kind":"pith_short_8","alias_value":"KDIVL6T4","created_at":"2026-07-05T10:25:47.243461+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20683","citing_title":"From Question Answering to Task Completion: A Survey on Agent System and Harness Design","ref_index":255,"is_internal_anchor":false},{"citing_arxiv_id":"2411.18279","citing_title":"Large Language Model-Brained GUI Agents: A Survey","ref_index":300,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KDIVL6T445NZ25YUDAJ6PCYGQV","json":"https://pith.science/pith/KDIVL6T445NZ25YUDAJ6PCYGQV.json","graph_json":"https://pith.science/api/pith-number/KDIVL6T445NZ25YUDAJ6PCYGQV/graph.json","events_json":"https://pith.science/api/pith-number/KDIVL6T445NZ25YUDAJ6PCYGQV/events.json","paper":"https://pith.science/paper/KDIVL6T4"},"agent_actions":{"view_html":"https://pith.science/pith/KDIVL6T445NZ25YUDAJ6PCYGQV","download_json":"https://pith.science/pith/KDIVL6T445NZ25YUDAJ6PCYGQV.json","view_paper":"https://pith.science/paper/KDIVL6T4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.07942&json=true","fetch_graph":"https://pith.science/api/pith-number/KDIVL6T445NZ25YUDAJ6PCYGQV/graph.json","fetch_events":"https://pith.science/api/pith-number/KDIVL6T445NZ25YUDAJ6PCYGQV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KDIVL6T445NZ25YUDAJ6PCYGQV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KDIVL6T445NZ25YUDAJ6PCYGQV/action/storage_attestation","attest_author":"https://pith.science/pith/KDIVL6T445NZ25YUDAJ6PCYGQV/action/author_attestation","sign_citation":"https://pith.science/pith/KDIVL6T445NZ25YUDAJ6PCYGQV/action/citation_signature","submit_replication":"https://pith.science/pith/KDIVL6T445NZ25YUDAJ6PCYGQV/action/replication_record"}},"created_at":"2026-07-05T10:25:47.243461+00:00","updated_at":"2026-07-05T10:25:47.243461+00:00"}