{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:U5ZII53PUDBI3ZPHQWNR4376YF","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":"f672d682a77307538cd790975e1ceeb9b292fdfdf0b6e93ac9090019acd7d1a0","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-24T23:11:44Z","title_canon_sha256":"26fe96969ff62fbc8062cc7d547f2bc67ec548d8cb116d88c4db89722698a604"},"schema_version":"1.0","source":{"id":"2604.23061","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.23061","created_at":"2026-05-28T01:04:08Z"},{"alias_kind":"arxiv_version","alias_value":"2604.23061v2","created_at":"2026-05-28T01:04:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.23061","created_at":"2026-05-28T01:04:08Z"},{"alias_kind":"pith_short_12","alias_value":"U5ZII53PUDBI","created_at":"2026-05-28T01:04:08Z"},{"alias_kind":"pith_short_16","alias_value":"U5ZII53PUDBI3ZPH","created_at":"2026-05-28T01:04:08Z"},{"alias_kind":"pith_short_8","alias_value":"U5ZII53P","created_at":"2026-05-28T01:04:08Z"}],"graph_snapshots":[{"event_id":"sha256:9a69d40d0673c6e64b806a1c5a4717071796ecb004c13f43a30bd102bbe21ce1","target":"graph","created_at":"2026-05-28T01:04:08Z","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":"Experiments on the C-MuMOInstruct benchmark show that C-Moral consistently outperforms state-of-the-art models across both in-domain and out-of-domain settings, achieving the best Success Optimized Rate (SOR) of 48.9% on IND tasks and 39.5% on OOD tasks, while largely preserving scaffold similarity."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the reported performance gains on the C-MuMOInstruct benchmark are attributable to the proposed components (group-based relative optimization, property score alignment, and continuous non-linear reward aggregation) rather than implementation details or benchmark-specific artifacts, and that these gains generalize to practical drug design."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"C-MORAL applies reinforcement learning post-training with group-based optimization and non-linear reward aggregation to align LLMs for controllable multi-objective molecular optimization, achieving 48.9% SOR on in-domain and 39.5% on out-of-domain benchmark tasks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Reinforcement learning post-training with group-based optimization and non-linear rewards aligns LLMs to optimize molecules across multiple competing properties."}],"snapshot_sha256":"657467683ba83042f4dcf01cc6cc105b90e1b61a083cfc5c3bc2a97de78d7259"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-21T09:40:14.227532Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T23:32:21.434242Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2604.23061/integrity.json","findings":[],"snapshot_sha256":"25f72fb6a56f92bbb4ebade731bfc90899b4b01f3d120cf23222560e699cec42","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) show promise for molecular optimization, but aligning them with selective and competing drug-design constraints remains challenging. We propose C-Moral, a reinforcement learning post-training framework for controllable multi-objective molecular optimization. C-Moral combines group-based relative optimization, property score alignment for heterogeneous objectives, and bottleneck-sensitive non-linear reward aggregation to improve stability across competing molecular properties. Experiments on C-MuMOInstruct and S$^2$-Bench MolOpt show that C-Moral achieves the best p","authors_text":"Morteza Ziyadi, Rui Gao, Swastik Roy, Xiang 'Anthony' Chen, Youngseung Jeon","cross_cats":["cs.AI"],"headline":"Reinforcement learning post-training with group-based optimization and non-linear rewards aligns LLMs to optimize molecules across multiple competing properties.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-24T23:11:44Z","title":"C-MORAL: Controllable Multi-Objective Molecular Optimization with Reinforcement Alignment for LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.23061","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-08T11:55:55.843945Z","id":"d9398012-17c5-4c2a-b2bf-48766a493a30","model_set":{"reader":"grok-4.3"},"one_line_summary":"C-MORAL applies reinforcement learning post-training with group-based optimization and non-linear reward aggregation to align LLMs for controllable multi-objective molecular optimization, achieving 48.9% SOR on in-domain and 39.5% on out-of-domain benchmark tasks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Reinforcement learning post-training with group-based optimization and non-linear rewards aligns LLMs to optimize molecules across multiple competing properties.","strongest_claim":"Experiments on the C-MuMOInstruct benchmark show that C-Moral consistently outperforms state-of-the-art models across both in-domain and out-of-domain settings, achieving the best Success Optimized Rate (SOR) of 48.9% on IND tasks and 39.5% on OOD tasks, while largely preserving scaffold similarity.","weakest_assumption":"That the reported performance gains on the C-MuMOInstruct benchmark are attributable to the proposed components (group-based relative optimization, property score alignment, and continuous non-linear reward aggregation) rather than implementation details or benchmark-specific artifacts, and that these gains generalize to practical drug design."}},"verdict_id":"d9398012-17c5-4c2a-b2bf-48766a493a30"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:6bf42c6e43f13b392ccd721b09569c3bab5a4f83f7c97c49eb84bf3ba5177006","target":"record","created_at":"2026-05-28T01:04:08Z","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":"f672d682a77307538cd790975e1ceeb9b292fdfdf0b6e93ac9090019acd7d1a0","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-24T23:11:44Z","title_canon_sha256":"26fe96969ff62fbc8062cc7d547f2bc67ec548d8cb116d88c4db89722698a604"},"schema_version":"1.0","source":{"id":"2604.23061","kind":"arxiv","version":2}},"canonical_sha256":"a77284776fa0c28de5e7859b1e6ffec142b4c702f1634c29a92b4b09d8da9e09","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a77284776fa0c28de5e7859b1e6ffec142b4c702f1634c29a92b4b09d8da9e09","first_computed_at":"2026-05-28T01:04:08.586605Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T01:04:08.586605Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"S781j/cMVMUtBWhDpspbmmAmzEZWO4Zhk6LNt9x19ahWh4vdQRFa3WohvOqH7+8TX/idC70Vd11K2fEWniDjAg==","signature_status":"signed_v1","signed_at":"2026-05-28T01:04:08.587206Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.23061","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6bf42c6e43f13b392ccd721b09569c3bab5a4f83f7c97c49eb84bf3ba5177006","sha256:9a69d40d0673c6e64b806a1c5a4717071796ecb004c13f43a30bd102bbe21ce1"],"state_sha256":"3446598f8621e830dae8770e96e506d803fdf5d620d61c41025633dd1d26b496"}