{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:H56A4BZIF2DD5OMSS437L5Q6TV","short_pith_number":"pith:H56A4BZI","schema_version":"1.0","canonical_sha256":"3f7c0e07282e863eb9929737f5f61e9d4955708efc77e58f5e35263f990a7b68","source":{"kind":"arxiv","id":"2505.03733","version":2},"attestation_state":"computed","paper":{"title":"WebGen-Bench: Evaluating LLMs on Generating Interactive and Functional Websites from Scratch","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aojun Zhou, Han Xiao, Haotian Hou, Hongsheng Li, Houxing Ren, Ke Wang, Mingjie Zhan, Weikang Shi, Yunqiao Yang, Zimu Lu","submitted_at":"2025-05-06T17:59:15Z","abstract_excerpt":"LLM-based agents have demonstrated great potential in generating and managing code within complex codebases. In this paper, we introduce WebGen-Bench, a novel benchmark designed to measure an LLM-based agent's ability to create multi-file website codebases from scratch. It contains diverse instructions for website generation, created through the combined efforts of human annotators and GPT-4o. These instructions span three major categories and thirteen minor categories, encompassing nearly all important types of web applications. To assess the quality of the generated websites, we use GPT-4o t"},"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":"2505.03733","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-06T17:59:15Z","cross_cats_sorted":[],"title_canon_sha256":"9386d3627a0af3eedd2a189bc301e998e282d9a6fc051fef05771093b44f7e13","abstract_canon_sha256":"d11d7d5d61a3a78aadded7afa2b8cf9f27f1fd5473be8736302f8e33ffea34aa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:51:26.682739Z","signature_b64":"tnWNaA7RJ/g8f5mLatHSWuqyWOCMuP1Lm6A6NzvtK4v5NJluECCrm1jV9NN3CGZ90FimUj3R+5/Bu6DgHOohCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3f7c0e07282e863eb9929737f5f61e9d4955708efc77e58f5e35263f990a7b68","last_reissued_at":"2026-07-05T11:51:26.682234Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:51:26.682234Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"WebGen-Bench: Evaluating LLMs on Generating Interactive and Functional Websites from Scratch","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aojun Zhou, Han Xiao, Haotian Hou, Hongsheng Li, Houxing Ren, Ke Wang, Mingjie Zhan, Weikang Shi, Yunqiao Yang, Zimu Lu","submitted_at":"2025-05-06T17:59:15Z","abstract_excerpt":"LLM-based agents have demonstrated great potential in generating and managing code within complex codebases. In this paper, we introduce WebGen-Bench, a novel benchmark designed to measure an LLM-based agent's ability to create multi-file website codebases from scratch. It contains diverse instructions for website generation, created through the combined efforts of human annotators and GPT-4o. These instructions span three major categories and thirteen minor categories, encompassing nearly all important types of web applications. To assess the quality of the generated websites, we use GPT-4o t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.03733","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/2505.03733/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":"2505.03733","created_at":"2026-07-05T11:51:26.682293+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.03733v2","created_at":"2026-07-05T11:51:26.682293+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.03733","created_at":"2026-07-05T11:51:26.682293+00:00"},{"alias_kind":"pith_short_12","alias_value":"H56A4BZIF2DD","created_at":"2026-07-05T11:51:26.682293+00:00"},{"alias_kind":"pith_short_16","alias_value":"H56A4BZIF2DD5OMS","created_at":"2026-07-05T11:51:26.682293+00:00"},{"alias_kind":"pith_short_8","alias_value":"H56A4BZI","created_at":"2026-07-05T11:51:26.682293+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":15,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.05920","citing_title":"Asuka-Bench: Benchmarking Code Agents on Underspecified User Intent and Multi-Round Refinement","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30358","citing_title":"QASM-Eval: A Dataset to Train and Evaluate LLMs on OpenQASM-3 Beyond Quantum Circuits","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26807","citing_title":"HTMLCure: Turning Browser Experience into State Guided Repair for Interactive HTML","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30000","citing_title":"Cookie-Bench: Continuous On-screen Key Interaction Evaluation for Web Generation","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00750","citing_title":"I-WebGenBench : Evaluating Interactivity in LLM-Generated Scientific Web Applications","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17637","citing_title":"WebGameBench: Requirement-to-Application Evaluation for Coding Agents via Browser-Native Games","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17439","citing_title":"DiagEval: Trajectory-Conditioned Diagnosis for Reliable Software Evaluation with GUI Agents","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17637","citing_title":"WebGameBench: Requirement-to-Application Evaluation for Coding Agents via Browser-Native Games","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17242","citing_title":"From Runnable to Shippable: Multi-Agent Test-Driven Development for Generating Full-Stack Web Applications from Requirements","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17439","citing_title":"DiagEval: Trajectory-Conditioned Diagnosis for Reliable Software Evaluation with GUI Agents","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17526","citing_title":"SaaSBench: Exploring the Boundaries of Coding Agents in Long-Horizon Enterprise SaaS Engineering","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04637","citing_title":"SWE-WebDevBench: Evaluating Coding Agent Application Platforms as Virtual Software Agencies","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07442","citing_title":"GameGen-Verifier: Parallel Keypoint-Based Verification for LLM-Generated Games via Runtime State Injection","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15309","citing_title":"MM-WebAgent: A Hierarchical Multimodal Web Agent for Webpage Generation","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20398","citing_title":"WebGen-R1: Incentivizing Large Language Models to Generate Functional and Aesthetic Websites with Reinforcement Learning","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/H56A4BZIF2DD5OMSS437L5Q6TV","json":"https://pith.science/pith/H56A4BZIF2DD5OMSS437L5Q6TV.json","graph_json":"https://pith.science/api/pith-number/H56A4BZIF2DD5OMSS437L5Q6TV/graph.json","events_json":"https://pith.science/api/pith-number/H56A4BZIF2DD5OMSS437L5Q6TV/events.json","paper":"https://pith.science/paper/H56A4BZI"},"agent_actions":{"view_html":"https://pith.science/pith/H56A4BZIF2DD5OMSS437L5Q6TV","download_json":"https://pith.science/pith/H56A4BZIF2DD5OMSS437L5Q6TV.json","view_paper":"https://pith.science/paper/H56A4BZI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.03733&json=true","fetch_graph":"https://pith.science/api/pith-number/H56A4BZIF2DD5OMSS437L5Q6TV/graph.json","fetch_events":"https://pith.science/api/pith-number/H56A4BZIF2DD5OMSS437L5Q6TV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/H56A4BZIF2DD5OMSS437L5Q6TV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/H56A4BZIF2DD5OMSS437L5Q6TV/action/storage_attestation","attest_author":"https://pith.science/pith/H56A4BZIF2DD5OMSS437L5Q6TV/action/author_attestation","sign_citation":"https://pith.science/pith/H56A4BZIF2DD5OMSS437L5Q6TV/action/citation_signature","submit_replication":"https://pith.science/pith/H56A4BZIF2DD5OMSS437L5Q6TV/action/replication_record"}},"created_at":"2026-07-05T11:51:26.682293+00:00","updated_at":"2026-07-05T11:51:26.682293+00:00"}