LACE: Lattice Attention for Cross-thread Exploration

Computer Science > Artificial Intelligence [Submitted on 16 Apr 2026] LACE: Lattice Attention for Cross-thread Exploration Yang Li, Zirui Zhang, Yang Liu, Chengzhi Mao Current large language models reason in isolation. Although it is common to sample multiple reasoning paths in parallel, these trajectories do not interact, and often fail in the same redundant ways. We introduce LACE, a framework that transforms reasoning from a collection of independent trials into a coordinated, parallel process. By repurposing the model architecture to enable cross-thread attention, LACE allows concurrent reasoning paths to share intermediate insights and correct one another during inference. A central challenge is the absence of natural training data that exhibits such collaborative behavior. We address this gap with a synthetic data pipeline that explicitly teaches models to communicate and error-correct across threads. Experiments show that this unified exploration substantially outperforms standard parallel search, improving reasoning accuracy by over 7 points. Our results suggest that large language models can be more effective when parallel reasoning paths are allowed to interact. Comments: 22 pages, 15 figures Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2604.15529 [cs.AI]   (or arXiv:2604.15529v1 [cs.AI] for this version)   https://doi.org/10.48550/arXiv.2604.15529 Focus to learn more Submission history From: Yang Li [view email] [v1] Thu, 16 Apr 2026 21:19:35 UTC (1,559 KB) Access Paper: HTML (experimental) view license Current browse context: cs.AI < prev   |   next > new | recent | 2026-04 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)