{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/arxiv-2606-01595","title":"Uncertainty-Calibrated Diffusion for Reliable 3D Molecular Graph Generation","arxiv_id":"2606.01595","date":"2026-06-01","proceeding":null,"authors":["Fang Wan","Jingxiang Qu","Yi Liu"],"abstract":"Bayesian inference provides a principled framework for modeling epistemic uncertainty in neural networks by treating predictions as distributions rather than deterministic values. Meanwhile, diffusion-based models for 3D molecular graph generation operate on fragile geometric structures governed by strict chemical constraints, making inference highly sensitive to uncertainty miscalibration. A largely overlooked issue is that epistemic uncertainty arising from the learned denoiser interacts with the aleatoric uncertainty intentionally injected during reverse diffusion, leading to systematic variance inflation and a mismatch between the true distribution and the simulated distribution. This effect is particularly detrimental for high-precision molecular generation, where even small deviations can violate chemical validity. In this work, we provide a theoretical and empirical analysis of how epistemic uncertainty propagates through diffusion inference and degrades sampling quality. Building on this investigation, we propose UCD (Uncertainty-Calibrated Diffusion), a simple yet effective method that calibrates the reverse diffusion process to account for epistemic uncertainty. Extensive experiments on standard 3D molecular benchmarks demonstrate that UCD consistently improves sampling quality across diverse baseline methods, establishing new state-of-the-art performance for 3D molecular diffusion. The code is available at https://github.com/jiuguaiwf/UCD.","url_abs":"https://arxiv.org/abs/2606.01595","url_pdf":"https://arxiv.org/pdf/2606.01595","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2606.01595"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/jiuguaiwf/UCD","reach":null}],"summary":{"ran_draft_wrong":1,"ran":5,"unverified":3},"by_repo_kind":{"found_in_text":{"samples":9,"ran":6,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"0f786c407fb1ee4c","entry":"swish","repo":"jiuguaiwf/UCD","repo_kind":"found_in_text","path":"edm/egnn/dropout_layer.py","file_url":"https://github.com/jiuguaiwf/UCD/blob/HEAD/edm/egnn/dropout_layer.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":2,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0f786c407fb1ee4c"}},{"code_sha256_prefix":"1b4121b61259e9d2","entry":"build_diagonal_laplace_state","repo":"jiuguaiwf/UCD","repo_kind":"found_in_text","path":"edm/egnn/last_layer_laplace.py","file_url":"https://github.com/jiuguaiwf/UCD/blob/HEAD/edm/egnn/last_layer_laplace.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1b4121b61259e9d2"}},{"code_sha256_prefix":"4ed06d5114046b7e","entry":"compute_loss_and_nll","repo":"jiuguaiwf/UCD","repo_kind":"found_in_text","path":"edm/qm9/losses.py","file_url":"https://github.com/jiuguaiwf/UCD/blob/HEAD/edm/qm9/losses.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4ed06d5114046b7e"}},{"code_sha256_prefix":"dba4c0c168802618","entry":"compute_mc_dropout_scalar_uncertainty","repo":"jiuguaiwf/UCD","repo_kind":"found_in_text","path":"edm/egnn/models.py","file_url":"https://github.com/jiuguaiwf/UCD/blob/HEAD/edm/egnn/models.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"dba4c0c168802618"}},{"code_sha256_prefix":"24d665e94f616fb4","entry":"load_split_data","repo":"jiuguaiwf/UCD","repo_kind":"found_in_text","path":"edm/build_geom_dataset.py","file_url":"https://github.com/jiuguaiwf/UCD/blob/HEAD/edm/build_geom_dataset.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"24d665e94f616fb4"}},{"code_sha256_prefix":"90c529ca8e85d3db","entry":"sum_except_batch","repo":"jiuguaiwf/UCD","repo_kind":"found_in_text","path":"edm/qm9/losses.py","file_url":"https://github.com/jiuguaiwf/UCD/blob/HEAD/edm/qm9/losses.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"90c529ca8e85d3db"}},{"code_sha256_prefix":"f576c0a0e0ae29c2","entry":"get_args_gen","repo":"jiuguaiwf/UCD","repo_kind":"found_in_text","path":"edm/eval_conditional_qm9.py","file_url":"https://github.com/jiuguaiwf/UCD/blob/HEAD/edm/eval_conditional_qm9.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f576c0a0e0ae29c2"}},{"code_sha256_prefix":"d3bd9393f95d2c44","entry":"maybe_load_last_layer_laplace","repo":"jiuguaiwf/UCD","repo_kind":"found_in_text","path":"edm/eval_analyze-uncertainty.py","file_url":"https://github.com/jiuguaiwf/UCD/blob/HEAD/edm/eval_analyze-uncertainty.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d3bd9393f95d2c44"}},{"code_sha256_prefix":"8e8422c6144b4f89","entry":"sample_diagonal_last_layer","repo":"jiuguaiwf/UCD","repo_kind":"found_in_text","path":"edm/egnn/last_layer_laplace.py","file_url":"https://github.com/jiuguaiwf/UCD/blob/HEAD/edm/egnn/last_layer_laplace.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8e8422c6144b4f89"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.LG","source":"arxiv_2026.jsonl"},"syntology_extracted_results":null}