{"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/graph-diffusion-policy-optimization","title":"Graph Diffusion Policy Optimization","arxiv_id":"2402.16302","date":"2024-02-26","proceeding":null,"authors":["Yijing Liu","Chao Du","Tianyu Pang","Chongxuan Li","Min Lin","Wei Chen"],"abstract":"Recent research has made significant progress in optimizing diffusion models for downstream objectives, which is an important pursuit in fields such as graph generation for drug design. However, directly applying these models to graph presents challenges, resulting in suboptimal performance. This paper introduces graph diffusion policy optimization (GDPO), a novel approach to optimize graph diffusion models for arbitrary (e.g., non-differentiable) objectives using reinforcement learning. GDPO is based on an eager policy gradient tailored for graph diffusion models, developed through meticulous analysis and promising improved performance. Experimental results show that GDPO achieves state-of-the-art performance in various graph generation tasks with complex and diverse objectives. Code is available at https://github.com/sail-sg/GDPO.","url_abs":"https://arxiv.org/abs/2402.16302v2","url_pdf":"https://arxiv.org/pdf/2402.16302v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"graph-diffusion-policy-optimization","repo_url":"https://github.com/sail-sg/gdpo","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"drug-design","task_name":"Drug Design"},{"task_slug":"graph-generation","task_name":"Graph Generation"}],"methods":[{"method_slug":"diffusion","method_name":"Diffusion"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2402.16302","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2402.16302"}},"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":[{"provenance":"deterministic:regex_extraction","url":"https://github.com/sail-sg/GDPO","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/sail-sg/gdpo","reach":{"status":"ok"}}],"summary":{"ran":4,"ran_draft_wrong":3,"unverified":6},"by_repo_kind":{"official":{"samples":10,"ran":4,"repositories":1},"community":{"samples":3,"ran":3,"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":10,"samples":[{"code_sha256_prefix":"a17d699ffe7e149f","entry":"get_positive_expectation","repo":"sail-sg/gdpo","repo_kind":"official","path":"model/infomax.py","file_url":"https://github.com/sail-sg/gdpo/blob/HEAD/model/infomax.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a17d699ffe7e149f"}},{"code_sha256_prefix":"a6cece2ffebe0f9c","entry":"nanstd","repo":"NVlabs/GDPO","repo_kind":"community","path":"trl-GDPO/trl-0.18.0-gdpo/trl/trainer/grpo_trainer.py","file_url":"https://github.com/NVlabs/GDPO/blob/HEAD/trl-GDPO/trl-0.18.0-gdpo/trl/trainer/grpo_trainer.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"a6cece2ffebe0f9c"}},{"code_sha256_prefix":"b58843c8f0e6cce0","entry":"random_permute","repo":"sail-sg/gdpo","repo_kind":"official","path":"model/infomax.py","file_url":"https://github.com/sail-sg/gdpo/blob/HEAD/model/infomax.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"b58843c8f0e6cce0"}},{"code_sha256_prefix":"cb06d97097718c4f","entry":"setup_wandb","repo":"sail-sg/gdpo","repo_kind":"official","path":"main_ppo.py","file_url":"https://github.com/sail-sg/gdpo/blob/HEAD/main_ppo.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"cb06d97097718c4f"}},{"code_sha256_prefix":"4f7f0f20f97f300e","entry":"shuffle_tensor_dict","repo":"NVlabs/GDPO","repo_kind":"community","path":"trl-GDPO/trl-0.18.0-gdpo/trl/trainer/grpo_trainer.py","file_url":"https://github.com/NVlabs/GDPO/blob/HEAD/trl-GDPO/trl-0.18.0-gdpo/trl/trainer/grpo_trainer.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4f7f0f20f97f300e"}},{"code_sha256_prefix":"efd67e2adf6b1ba5","entry":"split_tensor_dict","repo":"NVlabs/GDPO","repo_kind":"community","path":"trl-GDPO/trl-0.18.0-gdpo/trl/trainer/grpo_trainer.py","file_url":"https://github.com/NVlabs/GDPO/blob/HEAD/trl-GDPO/trl-0.18.0-gdpo/trl/trainer/grpo_trainer.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"efd67e2adf6b1ba5"}},{"code_sha256_prefix":"44c3cf76becae1f4","entry":"to_dense","repo":"sail-sg/gdpo","repo_kind":"official","path":"process.py","file_url":"https://github.com/sail-sg/gdpo/blob/HEAD/process.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"44c3cf76becae1f4"}},{"code_sha256_prefix":"4d13d6ce3ccd0cc1","entry":"compress","repo":"sail-sg/gdpo","repo_kind":"official","path":"model/diffusion_discrete.py","file_url":"https://github.com/sail-sg/gdpo/blob/HEAD/model/diffusion_discrete.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4d13d6ce3ccd0cc1"}},{"code_sha256_prefix":"86b15e5a9c92ef51","entry":"encode_no_edge","repo":"sail-sg/gdpo","repo_kind":"official","path":"process.py","file_url":"https://github.com/sail-sg/gdpo/blob/HEAD/process.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"86b15e5a9c92ef51"}},{"code_sha256_prefix":"fc28afeec6883d4e","entry":"generate_random_letter","repo":"sail-sg/GDPO","repo_kind":"official","path":"main_generate.py","file_url":"https://github.com/sail-sg/GDPO/blob/HEAD/main_generate.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"fc28afeec6883d4e"}},{"code_sha256_prefix":"c03feba6bc4986f6","entry":"log_sum_exp","repo":"sail-sg/gdpo","repo_kind":"official","path":"model/infomax.py","file_url":"https://github.com/sail-sg/gdpo/blob/HEAD/model/infomax.py","link_basis":"plan_row","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c03feba6bc4986f6"}},{"code_sha256_prefix":"1717b7476aced94e","entry":"masked_softmax","repo":"sail-sg/gdpo","repo_kind":"official","path":"model/layers.py","file_url":"https://github.com/sail-sg/gdpo/blob/HEAD/model/layers.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1717b7476aced94e"}},{"code_sha256_prefix":"7c6f8ff8273c13ec","entry":"to_sparse_batch","repo":"sail-sg/gdpo","repo_kind":"official","path":"model/diffusion_discrete.py","file_url":"https://github.com/sail-sg/gdpo/blob/HEAD/model/diffusion_discrete.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"7c6f8ff8273c13ec"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}