{"id":"W4417086410","doi":"10.48550/arxiv.2512.04105","title":"LegalWebAgent: Empowering Access to Justice via LLM-Based Web Agents","year":2025,"lang":"","type":"preprint","venue":"ArXiv.org","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Natural language; Economic Justice; Identification (biology); Autonomy; The Internet; Process (computing); Legal aspects of computing; Bridge (graph theory)","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","sts","scholarly_communication","open_science","research_integrity","insufficient_payload"],"consensus_categories":["metaepi_narrow","insufficient_payload"],"category_scores_codex":[0.002463502,0.001337659,0.001384079,0.0007136711,0.002951728,0.001454934,0.006902157,0.001351184,0.007085287],"category_scores_gemma":[0.00217834,0.001598889,0.0008692757,0.002908431,0.001181953,0.001177741,0.004595147,0.002057087,0.005860518],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001369398,"about_ca_system_score_gemma":0.002898798,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009477316,"about_ca_topic_score_gemma":0.008141975,"domain_scores_codex":[0.9894452,0.0009930524,0.002247764,0.003004027,0.001965697,0.002344239],"domain_scores_gemma":[0.9933584,0.00103394,0.0009765586,0.002345621,0.001042024,0.001243442],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001300548,0.005022654,0.6664927,0.004709106,0.001715601,0.0006548991,0.08693652,0.1220341,0.007308803,0.007442476,0.03813866,0.05824398],"study_design_scores_gemma":[0.0006669317,0.0004757702,0.03607776,0.004567132,0.002444794,0.000002527451,0.01368403,0.0328249,0.05425355,0.00248468,0.8466069,0.005910997],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8369164,0.0002997787,0.02399732,0.008318917,0.02578391,0.003563078,0.0001639336,0.000554911,0.1004017],"genre_scores_gemma":[0.9768921,0.0003538396,0.0005349104,0.006780294,0.002171574,0.000389018,0.00003847207,0.0001085738,0.01273119],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8084683,"threshold_uncertainty_score":0.9999453,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1917931480550452,"score_gpt":0.449536000138815,"score_spread":0.2577428520837698,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}