{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:XBEJZENZKFYYM2QOCNI5V63AOS","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"9b03b1fad8d7d1605a205e230fd31d2c7a7d56543fe7ca23a8fe4d1f6ac0e536","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-16T02:44:32Z","title_canon_sha256":"dea59d1b818dfa16b43ada94cc95b2c81b22d22c39a484b5484fd26120dd7d21"},"schema_version":"1.0","source":{"id":"2605.16770","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.16770","created_at":"2026-05-20T00:03:21Z"},{"alias_kind":"arxiv_version","alias_value":"2605.16770v1","created_at":"2026-05-20T00:03:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16770","created_at":"2026-05-20T00:03:21Z"},{"alias_kind":"pith_short_12","alias_value":"XBEJZENZKFYY","created_at":"2026-05-20T00:03:21Z"},{"alias_kind":"pith_short_16","alias_value":"XBEJZENZKFYYM2QO","created_at":"2026-05-20T00:03:21Z"},{"alias_kind":"pith_short_8","alias_value":"XBEJZENZ","created_at":"2026-05-20T00:03:21Z"}],"graph_snapshots":[{"event_id":"sha256:39a7083dc0f9d570a0cf6156a8c0df5e688f9ceed1ffd37233df5e2e94dce8c9","target":"graph","created_at":"2026-05-20T00:03:21Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Experimental results provide new insights into the trade-offs between model architecture, model size, and the influence between different tasks, highlighting the potential of lightweight LLMs in judicial AI applications."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That training on a mixed dataset from multiple sources and evaluating on individual test sets produces fair, generalizable comparisons without significant domain shift or distribution mismatch biasing the architecture and size effects."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Lightweight LLMs are benchmarked for court view generation and charge prediction across architectures, sizes, DNN comparisons, and task ordering on three datasets using the new CVGEvalKit framework."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Lightweight LLMs under 2 billion parameters generate court views from case facts and support charge prediction with competitive results against DNNs."}],"snapshot_sha256":"3cff6fa2ef0558f1a5be941cdc974a04121181f6f53f66d274f6b76b6865322b"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T22:01:19.738818Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T21:40:53.480853Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T19:01:56.311188Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.445190Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.16770/integrity.json","findings":[],"snapshot_sha256":"f7295949c536f5092a5695c2545afe311474ed3aa7993dfcce8009358e205d04","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Criminal Court View Generation (CVG) is a critical task in Legal Artificial Intelligence (Legal AI), involving the generation of court view based on case facts. In this work, we systematically explore the capabilities of lightweight (smaller than 2B) large language models (LLMs) in CVG and their impact on charge prediction. Our study addresses four key questions: (1) how does different architecture of LLMs affect the CVG quality and charge prediction. (2) how does LLMs size contribute to the performance, (3) how do lightweight LLMs compare with Deep Neural Networks (DNNs) in these tasks, and (","authors_text":"Kun Zeng, Nanli Zeng, Tianyong Hao, Zhitian Hou, Zhixiong Chao","cross_cats":["cs.AI"],"headline":"Lightweight LLMs under 2 billion parameters generate court views from case facts and support charge prediction with competitive results against DNNs.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-16T02:44:32Z","title":"Exploring Lightweight Large Language Models for Court View Generation"},"references":{"count":22,"internal_anchors":5,"resolved_work":22,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"arXiv preprint arXiv:2509.09969 , year=","work_id":"5314ada5-57ee-4f0a-a11b-3f794ebcc781","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"2024 International Joint Conference on Neural Networks (IJCNN) , pages=","work_id":"ded7237c-d7a2-4fae-b82b-f0a143bd94bd","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"2024 International Joint Conference on Neural Networks (IJCNN) , pages=","work_id":"f4ff00ab-c1ab-4a49-93ca-e0a0f4a997b7","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"2024 International Joint Conference on Neural Networks (IJCNN) , pages=","work_id":"b9e6f164-de14-4383-8388-e69de8ce3025","year":2024},{"cited_arxiv_id":"1802.08504","doi":"","is_internal_anchor":true,"ref_index":5,"title":"Interpretable Charge Predictions for Criminal Cases: Learning to Generate Court Views from Fact Descriptions","work_id":"44f98285-a981-4c01-8ef9-91512ae25120","year":null}],"snapshot_sha256":"0b21901548b1ca10b17061f3fcd60e516625a38610e85804c2d7f59b28785728"},"source":{"id":"2605.16770","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T21:30:30.632644Z","id":"bdf10edc-f645-4aad-99cd-7a9d4e491700","model_set":{"reader":"grok-4.3"},"one_line_summary":"Lightweight LLMs are benchmarked for court view generation and charge prediction across architectures, sizes, DNN comparisons, and task ordering on three datasets using the new CVGEvalKit framework.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Lightweight LLMs under 2 billion parameters generate court views from case facts and support charge prediction with competitive results against DNNs.","strongest_claim":"Experimental results provide new insights into the trade-offs between model architecture, model size, and the influence between different tasks, highlighting the potential of lightweight LLMs in judicial AI applications.","weakest_assumption":"That training on a mixed dataset from multiple sources and evaluating on individual test sets produces fair, generalizable comparisons without significant domain shift or distribution mismatch biasing the architecture and size effects."}},"verdict_id":"bdf10edc-f645-4aad-99cd-7a9d4e491700"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ef706d453c767cfa632d0c0d69d8124dcacb6c9ffe9ea16c3e8a5ae9e2c2d16b","target":"record","created_at":"2026-05-20T00:03:21Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"9b03b1fad8d7d1605a205e230fd31d2c7a7d56543fe7ca23a8fe4d1f6ac0e536","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-16T02:44:32Z","title_canon_sha256":"dea59d1b818dfa16b43ada94cc95b2c81b22d22c39a484b5484fd26120dd7d21"},"schema_version":"1.0","source":{"id":"2605.16770","kind":"arxiv","version":1}},"canonical_sha256":"b8489c91b95171866a0e1351dafb6074875adb91ac9c042577a5778cc8ebb068","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b8489c91b95171866a0e1351dafb6074875adb91ac9c042577a5778cc8ebb068","first_computed_at":"2026-05-20T00:03:21.050047Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:03:21.050047Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rZ+0jbF7HNWg44e5K4TOGgM6TJijI9QkST/fX5mYeQxQ8kggTS5Fj0XLaXinFxf55EOXneCxe/uaSY+hGhJUAw==","signature_status":"signed_v1","signed_at":"2026-05-20T00:03:21.051035Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.16770","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ef706d453c767cfa632d0c0d69d8124dcacb6c9ffe9ea16c3e8a5ae9e2c2d16b","sha256:39a7083dc0f9d570a0cf6156a8c0df5e688f9ceed1ffd37233df5e2e94dce8c9"],"state_sha256":"44588c240e5ceeb064120ed711191b97c0e59942ded8e4daf80bfe2bce19a1b2"}