{"id":2220,"date":"2026-11-04T15:17:55","date_gmt":"2026-11-04T07:17:55","guid":{"rendered":"https:\/\/www.translab.top\/?p=2220"},"modified":"2026-11-04T15:17:59","modified_gmt":"2026-11-04T07:17:59","slug":"%e3%80%90ep54-%c2%b7-research-highlights%e3%80%91-neurips-2026-a-large-language-model-based-temporal-causal-neuro-symbolic-framework-for-energy-system-decision-making","status":"publish","type":"post","link":"https:\/\/www.translab.top\/index.php\/en\/2026\/11\/04\/%e3%80%90ep54-%c2%b7-research-highlights%e3%80%91-neurips-2026-a-large-language-model-based-temporal-causal-neuro-symbolic-framework-for-energy-system-decision-making\/","title":{"rendered":"\u3010EP54 \u00b7 Research Highlights\u3011\u2014NeurIPS 2026: A Large Language Model-Based Temporal Causal Neuro-Symbolic Framework for Energy System Decision-Making"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\">In September 2026, the TRANS Research Group published a study on temporal causal reasoning in large language models. Inspired by document layout reordering, this work proposes a paradigm shift from discrete narratives to causal topologies and introduces two architectures: the training-free neuro-symbolic TCR-Agent and the end-to-end TCR-Former, which internalizes causal-topological attention masks and temporal span biases into Transformer attention. On the TempoCausal benchmark, TCR-Former achieves 16.5% higher accuracy than SOTA methods and, with only about 8.6B parameters, outperforms GPT-5.4 and DeepSeek-V3.2, while revealing and bridging the structure-reasoning disconnection in existing methods.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">CITATION: Yang, Z., Luo, W., Chen, Y., &amp; Ou, S. (2026). From document layout to causal topology: An end-to-end architecture for temporal causal reasoning. arXiv preprint arXiv:2610.34094. https:\/\/doi.org\/10.48550\/arXiv.2610.34094.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>In September 2026, the TRANS Research Group published a [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":2212,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[70],"tags":[],"class_list":["post-2220","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-publications-en"],"blocksy_meta":[],"jetpack_sharing_enabled":true,"jetpack_featured_media_url":"https:\/\/www.translab.top\/wp-content\/uploads\/2026\/11\/1.jpg","_links":{"self":[{"href":"https:\/\/www.translab.top\/index.php\/wp-json\/wp\/v2\/posts\/2220","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.translab.top\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.translab.top\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.translab.top\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.translab.top\/index.php\/wp-json\/wp\/v2\/comments?post=2220"}],"version-history":[{"count":1,"href":"https:\/\/www.translab.top\/index.php\/wp-json\/wp\/v2\/posts\/2220\/revisions"}],"predecessor-version":[{"id":2221,"href":"https:\/\/www.translab.top\/index.php\/wp-json\/wp\/v2\/posts\/2220\/revisions\/2221"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.translab.top\/index.php\/wp-json\/wp\/v2\/media\/2212"}],"wp:attachment":[{"href":"https:\/\/www.translab.top\/index.php\/wp-json\/wp\/v2\/media?parent=2220"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.translab.top\/index.php\/wp-json\/wp\/v2\/categories?post=2220"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.translab.top\/index.php\/wp-json\/wp\/v2\/tags?post=2220"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}