{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:V3FYH6XJNUHF5ZVCMJBDSRL5N4","short_pith_number":"pith:V3FYH6XJ","schema_version":"1.0","canonical_sha256":"aecb83fae96d0e5ee6a2624239457d6f0733c61b2b84ca52ea41943f3370e0bc","source":{"kind":"arxiv","id":"2605.21740","version":1},"attestation_state":"computed","paper":{"title":"SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Amir Farimani, Hamed Mahdavi, Kathy Wei, Kevin Han, Niloofar Mireshghallah, Renfei Zhang","submitted_at":"2026-05-20T21:08:15Z","abstract_excerpt":"LLM agents have incredible potential for scientific discovery applications. However, the performance of LLM agents on real-world, small molecule drug design (SMDD) tasks across diverse chemistries and targets is unclear. Current evaluation methods are either ad hoc, too simple for real-world discovery, limited in scale, or restricted to single-turn question answering. In effort to standardize the evaluation of LLM agents on small molecule design, we introduce SMDD-Bench, a challenging, multi-turn, long-horizon agentic benchmark consisting of 502 guaranteed-solvable task instances spanning 5 ta"},"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":"2605.21740","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-20T21:08:15Z","cross_cats_sorted":[],"title_canon_sha256":"6650e4c52ae7b20bc402f949d8941b16dc557fffc0af1dddfe09335e3a7f7ac5","abstract_canon_sha256":"db684c1ee6c492af250d9a4bb7d3763942bd9f51317dff611a024afe2364e3ec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:03:30.136361Z","signature_b64":"icSxeIqegtJOBsgz5s6OU2dEITpyTbYy/Vv9mt9nFMx2Dfz2RUXS5ooVCk+HBZqSlPa551TRV8uQEk8GQBgnDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aecb83fae96d0e5ee6a2624239457d6f0733c61b2b84ca52ea41943f3370e0bc","last_reissued_at":"2026-05-22T01:03:30.135661Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:03:30.135661Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Amir Farimani, Hamed Mahdavi, Kathy Wei, Kevin Han, Niloofar Mireshghallah, Renfei Zhang","submitted_at":"2026-05-20T21:08:15Z","abstract_excerpt":"LLM agents have incredible potential for scientific discovery applications. However, the performance of LLM agents on real-world, small molecule drug design (SMDD) tasks across diverse chemistries and targets is unclear. Current evaluation methods are either ad hoc, too simple for real-world discovery, limited in scale, or restricted to single-turn question answering. In effort to standardize the evaluation of LLM agents on small molecule design, we introduce SMDD-Bench, a challenging, multi-turn, long-horizon agentic benchmark consisting of 502 guaranteed-solvable task instances spanning 5 ta"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21740","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/2605.21740/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":"2605.21740","created_at":"2026-05-22T01:03:30.135771+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.21740v1","created_at":"2026-05-22T01:03:30.135771+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21740","created_at":"2026-05-22T01:03:30.135771+00:00"},{"alias_kind":"pith_short_12","alias_value":"V3FYH6XJNUHF","created_at":"2026-05-22T01:03:30.135771+00:00"},{"alias_kind":"pith_short_16","alias_value":"V3FYH6XJNUHF5ZVC","created_at":"2026-05-22T01:03:30.135771+00:00"},{"alias_kind":"pith_short_8","alias_value":"V3FYH6XJ","created_at":"2026-05-22T01:03:30.135771+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/V3FYH6XJNUHF5ZVCMJBDSRL5N4","json":"https://pith.science/pith/V3FYH6XJNUHF5ZVCMJBDSRL5N4.json","graph_json":"https://pith.science/api/pith-number/V3FYH6XJNUHF5ZVCMJBDSRL5N4/graph.json","events_json":"https://pith.science/api/pith-number/V3FYH6XJNUHF5ZVCMJBDSRL5N4/events.json","paper":"https://pith.science/paper/V3FYH6XJ"},"agent_actions":{"view_html":"https://pith.science/pith/V3FYH6XJNUHF5ZVCMJBDSRL5N4","download_json":"https://pith.science/pith/V3FYH6XJNUHF5ZVCMJBDSRL5N4.json","view_paper":"https://pith.science/paper/V3FYH6XJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.21740&json=true","fetch_graph":"https://pith.science/api/pith-number/V3FYH6XJNUHF5ZVCMJBDSRL5N4/graph.json","fetch_events":"https://pith.science/api/pith-number/V3FYH6XJNUHF5ZVCMJBDSRL5N4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V3FYH6XJNUHF5ZVCMJBDSRL5N4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V3FYH6XJNUHF5ZVCMJBDSRL5N4/action/storage_attestation","attest_author":"https://pith.science/pith/V3FYH6XJNUHF5ZVCMJBDSRL5N4/action/author_attestation","sign_citation":"https://pith.science/pith/V3FYH6XJNUHF5ZVCMJBDSRL5N4/action/citation_signature","submit_replication":"https://pith.science/pith/V3FYH6XJNUHF5ZVCMJBDSRL5N4/action/replication_record"}},"created_at":"2026-05-22T01:03:30.135771+00:00","updated_at":"2026-05-22T01:03:30.135771+00:00"}