{"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-2603-01047","title":"Evaluating GFlowNet from partial episodes for stable and flexible policy-based training","arxiv_id":"2603.01047","date":"2026-03-01","proceeding":null,"authors":["Puhua Niu","Shili Wu","Xiaoning Qian"],"abstract":"Generative Flow Networks (GFlowNets) were developed to learn policies for efficiently sampling combinatorial candidates by interpreting their generative processes as trajectories in directed acyclic graphs. In the value-based training workflow, the objective is to enforce the balance over partial episodes between the flows of the learned policy and the estimated flows of the desired policy, implicitly encouraging policy divergence minimization. The policy-based strategy alternates between estimating the policy divergence and updating the policy, but reliable estimation of the divergence under directed acyclic graphs remains a major challenge. This work bridges the two perspectives by showing that flow balance also yields a principled policy evaluator that measures the divergence, and an evaluation balance objective over partial episodes is proposed for learning the evaluator. As demonstrated on both synthetic and real-world tasks, evaluation balance not only strengthens the reliability of policy-based training but also broadens its flexibility by seamlessly supporting parameterized backward policies and enabling the integration of offline data-collection techniques.","url_abs":"https://arxiv.org/abs/2603.01047","url_pdf":"https://arxiv.org/pdf/2603.01047","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":"https://app.syntology.ai/?focus=2603.01047","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2603.01047"}},"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/niupuhua1234/Sub-EB","reach":null}],"summary":{"unverified":7},"by_repo_kind":{"found_in_text":{"samples":7,"ran":0,"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":"874cb43146e2c04f","entry":"download","repo":"niupuhua1234/Sub-EB","repo_kind":"found_in_text","path":"gfn-subeb/data_dag/factories.py","file_url":"https://github.com/niupuhua1234/Sub-EB/blob/HEAD/gfn-subeb/data_dag/factories.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":"874cb43146e2c04f"}},{"code_sha256_prefix":"2d3fc4a1e259fd6d","entry":"expected_edges","repo":"niupuhua1234/Sub-EB","repo_kind":"found_in_text","path":"gfn-subeb/data_dag/metrics.py","file_url":"https://github.com/niupuhua1234/Sub-EB/blob/HEAD/gfn-subeb/data_dag/metrics.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":"2d3fc4a1e259fd6d"}},{"code_sha256_prefix":"89736f5ba822ec0a","entry":"expected_shd","repo":"niupuhua1234/Sub-EB","repo_kind":"found_in_text","path":"gfn-subeb/data_dag/metrics.py","file_url":"https://github.com/niupuhua1234/Sub-EB/blob/HEAD/gfn-subeb/data_dag/metrics.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":"89736f5ba822ec0a"}},{"code_sha256_prefix":"a2b19bef0331f930","entry":"ix_","repo":"niupuhua1234/Sub-EB","repo_kind":"found_in_text","path":"gfn-subeb/data_dag/scores/base.py","file_url":"https://github.com/niupuhua1234/Sub-EB/blob/HEAD/gfn-subeb/data_dag/scores/base.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":"a2b19bef0331f930"}},{"code_sha256_prefix":"0cb1419ec492cafd","entry":"logdet","repo":"niupuhua1234/Sub-EB","repo_kind":"found_in_text","path":"gfn-subeb/data_dag/scores/base.py","file_url":"https://github.com/niupuhua1234/Sub-EB/blob/HEAD/gfn-subeb/data_dag/scores/base.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":"0cb1419ec492cafd"}},{"code_sha256_prefix":"52ac2620e1dc4258","entry":"logdet","repo":"niupuhua1234/Sub-EB","repo_kind":"found_in_text","path":"gfn-subeb/data_dag/scores/base_torch.py","file_url":"https://github.com/niupuhua1234/Sub-EB/blob/HEAD/gfn-subeb/data_dag/scores/base_torch.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":"52ac2620e1dc4258"}},{"code_sha256_prefix":"0fca2cf84bdf5862","entry":"threshold_metrics","repo":"niupuhua1234/Sub-EB","repo_kind":"found_in_text","path":"gfn-subeb/data_dag/metrics.py","file_url":"https://github.com/niupuhua1234/Sub-EB/blob/HEAD/gfn-subeb/data_dag/metrics.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":"0fca2cf84bdf5862"}}]},"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}