Large Language Models (LLMs) have demonstrated impressive capabilities in creative tasks such as storytelling and E-mail generation. However, as LLMs are primarily trained on final text results rather than intermediate revisions, it might be challenging for them to perform text rewriting tasks. Most studies in the rewriting tasks focus on a particular transformation type within the boundaries of single sentences. In this work, we develop new strategies for instruction tuning and reinforcement learning to better align LLMs for cross-sentence rewriting tasks using diverse wording and structures expressed through natural languages including 1) generating rewriting instruction data from Wiki edits and public corpus through instruction generation and chain-of-thought prompting; 2) collecting comparison data for reward model training through a new ranking function. To facilitate this research, we introduce OpenRewriteEval, a novel benchmark covers a wide variety of rewriting types expressed through natural language instructions. Our results show significant improvements over a variety of baselines.

Yu-Chung Hsiao, Fedir Zubach, Gilles Baechler, Srinivas Sunkara, Victor Carbune, Jason Lin, Maria Wang, Yun Zhu, Jindong Chen. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Noncentrosymmetric (NCS) pnictides are indispensable for nonlinear optics, ferroelectrics, magnetic Weyl electronics, etc., areas, yet their structure design remains a substantial challenge. By using asymmetric ionic unit–driven covalent groups orienting and rigidity-flexibility coupling dual strategy, we successfully design and synthesize four NCS pnictides: [Sr ₄ Br] ₂ [M ^II ₃ Si ₂₅ P ₄₀ ] (M ^II = Mg, Cd) and [Ba ₃ Br][M ^III Si ₁₀ P ₁₆ ] (M ^III = Ga, In), which exhibit strong second harmonic generation effects (5.2 to 7.5 × AgGaS ₂ ), wide bandgaps (1.81 to 1.90 electron volts), and moderate birefringence (0.030 to 0.051). An unprecedented NCS structure-inducing mechanism analysis revealed that the (Sr ₄ Br) and (Ba ₄ Br) ionic units featuring the diamond-like electrostatic force field effectively break inversion symmetry and trigger uniform arrangement of the covalent tetrahedron groups. Furthermore, the nonlinear optical (NLO) properties and birefringence can be remarkably tuned by the secondary covalent building blocks (M ^II/III P ₄ tetrahedra) with distinct bond flexibility providing a broader space for regulating the key parameters. This work might expand chemical space for exploiting high-performance pnictide NLO materials.

The k-nearest neighbor (kNN) query is a cornerstone of similarity-based applications across various domains. While prior work has enhanced kNN search efficiency, it typically focuses on approximate methods for high-dimensional data or exact methods for low-dimensional data, often assuming static query and data distributions. This creates a significant gap in accelerating exact kNN search for low-to-medium dimensional data with dynamic query distributions. To fill this gap, we propose App2Exa, a cache-guided framework that integrates approximate and exact kNN search. App2Exa utilizes a dynamically maintained cache graph index to retrieve approximate results, which subsequently guide exact search using a VP-Tree with a best-first strategy. A benefit-driven caching mechanism further optimizes performance by prioritizing vectors based on frequency, recency, and computational cost. Experimental results demonstrate that App2Exa significantly boosts efficiency, providing a robust and scalable solution for evolving query patterns and enabling exact kNN search to support higher dimensionality more effectively.

Hg-based chalcogenides, Hg₃AsS₄Cl and Hg₃AsS₄Br, are synthesized, which show superior comprehensive linear and nonlinear optical properties for promising IR applications.

Hg₃AsSe₄X (X = Br and I) derived from honeycomb layered topology of GaSe, achieved superior optical properties including strong SHG effects, suitable band gaps, large birefringence and wide IR transparency range.