{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6C4NR4PJTVLM2K56QR3UPDPU3Z","short_pith_number":"pith:6C4NR4PJ","schema_version":"1.0","canonical_sha256":"f0b8d8f1e99d56cd2bbe8477478df4de7e9604cfa5ceae18a89cdd896c96ea86","source":{"kind":"arxiv","id":"2606.02386","version":1},"attestation_state":"computed","paper":{"title":"AgentPLM: Agentic Protein Language Models with Reasoning-Augmented Decoding for Protein Sequence Design","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["q-bio.QM"],"primary_cat":"cs.AI","authors_text":"Maxx Richard Rahman, Sahil Rahman","submitted_at":"2026-06-01T15:35:02Z","abstract_excerpt":"Protein language models (PLMs) are passive oracles: they generate sequences in a single forward pass with no mechanism to consult external biophysical feedback or redirect generation when a candidate violates thermodynamic or structural constraints. We introduce AgentPLM, which addresses this by equipping a pre-trained PLM with i) Reasoning-Augmented Decoding (RAD), which interleaves autoregressive generation with tool calls (ESMFold, FoldX, AutoDock Vina), and ii) Contrastive Agent Policy Optimisation (CAPO), a trajectory-level extension of direct preference optimisation that trains the polic"},"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":"2606.02386","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-01T15:35:02Z","cross_cats_sorted":["q-bio.QM"],"title_canon_sha256":"9afb59fba6da1298bd53d942ae79e13fd9643678a1e633151916410c19da1d43","abstract_canon_sha256":"e74f109c69d7c0afd87c1bea9850e9dcba892b029183dfd152c9ed9b9de21e80"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T03:04:57.783961Z","signature_b64":"vSLgqSvNSamEs1YoPqCML0uTNHijHG+5OF5h6jISlKuQF3hDpK7z68ecKZDA+EbJus/UTmaXvvAKWh8/hv2KCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f0b8d8f1e99d56cd2bbe8477478df4de7e9604cfa5ceae18a89cdd896c96ea86","last_reissued_at":"2026-06-02T03:04:57.783584Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T03:04:57.783584Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AgentPLM: Agentic Protein Language Models with Reasoning-Augmented Decoding for Protein Sequence Design","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["q-bio.QM"],"primary_cat":"cs.AI","authors_text":"Maxx Richard Rahman, Sahil Rahman","submitted_at":"2026-06-01T15:35:02Z","abstract_excerpt":"Protein language models (PLMs) are passive oracles: they generate sequences in a single forward pass with no mechanism to consult external biophysical feedback or redirect generation when a candidate violates thermodynamic or structural constraints. We introduce AgentPLM, which addresses this by equipping a pre-trained PLM with i) Reasoning-Augmented Decoding (RAD), which interleaves autoregressive generation with tool calls (ESMFold, FoldX, AutoDock Vina), and ii) Contrastive Agent Policy Optimisation (CAPO), a trajectory-level extension of direct preference optimisation that trains the polic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.02386","kind":"arxiv","version":1},"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/2606.02386/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":"2606.02386","created_at":"2026-06-02T03:04:57.783640+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.02386v1","created_at":"2026-06-02T03:04:57.783640+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.02386","created_at":"2026-06-02T03:04:57.783640+00:00"},{"alias_kind":"pith_short_12","alias_value":"6C4NR4PJTVLM","created_at":"2026-06-02T03:04:57.783640+00:00"},{"alias_kind":"pith_short_16","alias_value":"6C4NR4PJTVLM2K56","created_at":"2026-06-02T03:04:57.783640+00:00"},{"alias_kind":"pith_short_8","alias_value":"6C4NR4PJ","created_at":"2026-06-02T03:04:57.783640+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6C4NR4PJTVLM2K56QR3UPDPU3Z","json":"https://pith.science/pith/6C4NR4PJTVLM2K56QR3UPDPU3Z.json","graph_json":"https://pith.science/api/pith-number/6C4NR4PJTVLM2K56QR3UPDPU3Z/graph.json","events_json":"https://pith.science/api/pith-number/6C4NR4PJTVLM2K56QR3UPDPU3Z/events.json","paper":"https://pith.science/paper/6C4NR4PJ"},"agent_actions":{"view_html":"https://pith.science/pith/6C4NR4PJTVLM2K56QR3UPDPU3Z","download_json":"https://pith.science/pith/6C4NR4PJTVLM2K56QR3UPDPU3Z.json","view_paper":"https://pith.science/paper/6C4NR4PJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.02386&json=true","fetch_graph":"https://pith.science/api/pith-number/6C4NR4PJTVLM2K56QR3UPDPU3Z/graph.json","fetch_events":"https://pith.science/api/pith-number/6C4NR4PJTVLM2K56QR3UPDPU3Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6C4NR4PJTVLM2K56QR3UPDPU3Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6C4NR4PJTVLM2K56QR3UPDPU3Z/action/storage_attestation","attest_author":"https://pith.science/pith/6C4NR4PJTVLM2K56QR3UPDPU3Z/action/author_attestation","sign_citation":"https://pith.science/pith/6C4NR4PJTVLM2K56QR3UPDPU3Z/action/citation_signature","submit_replication":"https://pith.science/pith/6C4NR4PJTVLM2K56QR3UPDPU3Z/action/replication_record"}},"created_at":"2026-06-02T03:04:57.783640+00:00","updated_at":"2026-06-02T03:04:57.783640+00:00"}