{"id":"fdc42b4e-97bc-4114-889f-943455ad5785","arxiv_id":"2504.12037","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Tensor-network-based variational autoregressive networks (TNVAN) fix a small 'width set' of spins, contract the rest of the network exactly, and fit a neural variational distribution on the reduced system to estimate free energy.","lead":"This paper combines tensor-network contraction with autoregressive neural networks to estimate free energy upper bounds for spin systems. The hybrid method is tested on 2D Ising models, random graph spin glasses, and fully connected spin glasses, where it reports better accuracy than pure neural or tensor network methods.","discovery_kind":"new_method","skeptic_critique":null,"referee_report":null,"author_rebuttal":null,"desk_editor":null,"rs_alignment":null,"lean_confirmation":null,"pith_extraction":null,"created_at":"2026-08-16T12:39:38.936621+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":null,"supporting_citations":[],"review_version":1}