Publications

Manipulating properties of matter at the nanoscale is the essence of nanotechnology, which has enabled the realization of quantum dots, nanotubes, metamaterials, and two-dimensional materials with tailored electronic and optical properties. Two-dimensional semiconductors have revealed promising perspectives in nanotechnology. However, the tunability of their physical properties is challenging for semiconductors studied until now. Here we show the ability of morphological manipulation strategies, such as nanotexturing or, at the limit, important surface roughness, to enhance light absorption and the luminescent response of atomically thin indium selenide nanosheets. Besides, quantum-size confinement effects make this two-dimensional semiconductor to exhibit one of the largest band gap tunability ranges observed in a two-dimensional semiconductor: from infrared, in bulk material, to visible wavelengths, at the single layer. These results are relevant for the design of new optoelectronic devices, including heterostructures of two-dimensional materials with optimized band gap functionalities and in-plane heterojunctions with minimal junction defect density.

A hot phonon bottleneck may be responsible for slow hot carrier cooling in methylammonium lead iodide hybrid perovskite, creating the potential for more efficient hot carrier photovoltaics. In room-temperature 2D electronic spectra near the band edge, we observe amplitude oscillations due to a remarkably long lived 0.9 THz coherent phonon population at room temperature. This phonon (or set of phonons) is assigned to angular distortions of the Pb-I lattice, not coupled to cation rotations. The strong coupling between the electronic transition and the 0.9 THz mode(s), together with relative isolation from other phonon modes, makes it likely to cause a phonon bottleneck. The pump frequency resolution of the 2D spectra also enables independent observation of photoinduced absorptions and bleaches independently and confirms that features due to band gap renormalization are longer-lived than in transient absorption spectra.

Acinetobacter baumannii is a Gram-negative opportunistic pathogen that causes diverse infections, including pneumonia, bacteremia, and wound infections. Due to multiple intrinsic and acquired antimicrobial-resistance mechanisms, A. baumannii isolates are commonly multidrug resistant, and infections are notoriously difficult to treat.

Large reasoning models (LRMs) tackle complex reasoning problems by following long chain-of-thoughts (Long CoT) that incorporate reflection, backtracking, and self-validation. However, the training techniques and data requirements to elicit Long CoT remain poorly understood. In this work, we find that a Large Language model (LLM) can effectively learn Long CoT reasoning through data-efficient supervised fine-tuning (SFT) and parameter-efficient low-rank adaptation (LoRA). With just 17k long CoT training samples, the Qwen2.5-32B-Instruct model achieves significant improvements on a wide range of math and coding benchmarks, including 56.7% (+40.0%) on AIME 2024 and 57.0% (+8.1%) on LiveCodeBench, competitive to the proprietary o1-preview model's score of 44.6% and 59.1%. More importantly, we find that the structure of Long CoT is critical to the learning process, whereas the content of individual reasoning steps has minimal impact. Perturbations affecting content, such as training on incorrect samples or removing reasoning keywords, have little impact on performance. In contrast, structural modifications that disrupt logical consistency in the Long CoT, such as shuffling or deleting reasoning steps, significantly degrade accuracy. For example, a model trained on Long CoT samples with incorrect answers still achieves only 3.2% lower accuracy compared to training with fully correct samples. These insights deepen our understanding of how to elicit reasoning capabilities in LLMs and highlight key considerations for efficiently training the next generation of reasoning models. This is the academic paper of our previous released Sky-T1-32B-Preview model. Codes are available at https://github.com/NovaSky-AI/SkyThought.

Three new germanate solids, ASU-19, ASU-20-DAPe, and ASU-20-DACH, have been synthesized under hydrothermal conditions using respectively 1,4-diaminobutane (DAB), 1,5-diaminopentane (DAPe), and 1,4-diaminocyclohexane (DACH) as bases. The structures of ASU-19 and ASU-20-DACH have been characterized by single-crystal X-ray diffraction: ASU-19, Ge14O29X4·[GeOX2]·[H2DAB]3·3.8H2O (X = F or OH), space group P1̄, a = 11.4191(5), b = 12.0525(6), c = 18.1847(8) Å, α = 90.704(1)°, β = 92.635(1)°, γ = 91.389(1)°, V = 2499.1(2) Å3; ASU-20-DACH: Ge7O14X3·[H2DACH]1.5·2H2O, space group C2/c, a = 15.9525(11), b = 17.5476(12), c = 19.0027(13) Å, β = 109.446(1)°, and V = 5015.9(6) Å3. The structure of ASU-20-DAPe, Ge7O14X3·[H2DAPe]1.5·H2O, has been determined from X-ray powder diffraction data: space group C2/c, a = 16.3180(5), b = 16.6125(4), c = 17.8898(6) Å, β = 99.684(2)°, V = 4780.5(3) Å3. All three structures are based on the assembly of the same cluster Ge7(O,OH,F)19. The two ASU-20 structures consist of a slab of four-connected clusters. In ASU-19, the same layers are connected pairwise through a GeO2X2 spacer, generating a slab structure, with slab thickness ca. 20 Å. The existence of the same layer in the presence of three different organic bases shows the adaptability of the structures to molecules differing in size, shape, and symmetry.

Journal Article Unraveling Li Growth Kinetics in Solid Electrolytes Due to Charging Effect under Electron Microscopy Get access Tofunmi Ogunfunmi, Tofunmi Ogunfunmi National Center for Electron Microscopy, Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, California, United StatesDepartment of Materials Science and Engineering, University of California, Berkeley, California, United States Search for other works by this author on: Oxford Academic Google Scholar Xinxing Peng, Xinxing Peng National Center for Electron Microscopy, Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, California, United StatesDepartment of Materials Science and Engineering, University of California, Berkeley, California, United States Search for other works by this author on: Oxford Academic Google Scholar Qingsong Tu, Qingsong Tu Materials Science Division, Lawrence Berkeley National Laboratory, Berkeley, California, United StatesDepartment of Mechanical Engineering, Rochester Institute of Technology, Rochester, New York, United States Corresponding authors: mary.scott@berkeley.edu; howard.tu@rit.edu Search for other works by this author on: Oxford Academic Google Scholar Yaqian Zhang, Yaqian Zhang National Center for Electron Microscopy, Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, California, United StatesDepartment of Materials Science and Engineering, University of California, Berkeley, California, United States Search for other works by this author on: Oxford Academic Google Scholar KyuJung Jun, KyuJung Jun Department of Materials Science and Engineering, University of California, Berkeley, California, United StatesMaterials Science Division, Lawrence Berkeley National Laboratory, Berkeley, California, United States Search for other works by this author on: Oxford Academic Google Scholar Fengyu Shen, Fengyu Shen Energy Storage and Distributed Resources Division, Lawrence Berkeley National Laboratory, Berkeley, California, United States Search for other works by this author on: Oxford Academic Google Scholar Yingzhi Sun, Yingzhi Sun Department of Materials Science and Engineering, University of California, Berkeley, California, United StatesMaterials Science Division, Lawrence Berkeley National Laboratory, Berkeley, California, United States Search for other works by this author on: Oxford Academic Google Scholar Michael C Tucker, Michael C Tucker Energy Storage and Distributed Resources Division, Lawrence Berkeley National Laboratory, Berkeley, California, United States Search for other works by this author on: Oxford Academic Google Scholar Gerbrand Ceder, Gerbrand Ceder Department of Materials Science and Engineering, University of California, Berkeley, California, United StatesMaterials Science Division, Lawrence Berkeley National Laboratory, Berkeley, California, United States Search for other works by this author on: Oxford Academic Google Scholar M C Scott M C Scott National Center for Electron Microscopy, Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, California, United StatesDepartment of Materials Science and Engineering, University of California, Berkeley, California, United States Corresponding authors: mary.scott@berkeley.edu; howard.tu@rit.edu Search for other works by this author on: Oxford Academic Google Scholar Microscopy and Microanalysis, Volume 29, Issue Supplement_1, 1 August 2023, Pages 1335–1336, https://doi.org/10.1093/micmic/ozad067.684 Published: 22 July 2023

An entry from the Cambridge Structural Database, the world’s repository for small molecule crystal structures. The entry contains experimental data from a crystal diffraction study. The deposited dataset for this entry is freely available from the CCDC and typically includes 3D coordinates, cell parameters, space group, experimental conditions and quality measures.