Search for: All records

Large language models (LLMs) are increasingly adopted for a variety of tasks with implicit graphical structures, such as planning in robotics, multi-hop question answering or knowledge probing, structured commonsense reasoning, and more. While LLMs have advanced the state-of-the-art on these tasks with structure implications, whether LLMs could explicitly process textual descriptions of graphs and structures, map them to grounded conceptual spaces, and perform structured operations remains underexplored. To this end, we propose NLGraph (Natural Language Graph), a comprehensive benchmark of graph-based problem solving designed in natural language. NLGraph contains 29,370 problems, covering eight graph reasoning tasks with varying complexity from simple tasks such as connectivity and shortest path up to complex problems such as maximum flow and simulating graph neural networks. We evaluate LLMs (GPT-3/4) with various prompting approaches on the NLGraph benchmark and find that 1) language models do demonstrate preliminary graph reasoning abilities, 2) the benefit of advanced prompting and in-context learning diminishes on more complex graph problems, while 3) LLMs are also (un)surprisingly brittle in the face of spurious correlations in graph and problem settings. We then propose Build-a-Graph Prompting and Algorithmic Prompting, two instruction-based approaches to enhance LLMs in solving natural language graph problems. Build-a-Graph and Algorithmic prompting improve the performance of LLMs on NLGraph by 3.07% to 16.85% across multiple tasks and settings, while how to solve the most complicated graph reasoning tasks in our setup with language models remains an open research question. 

In this perspective, we discuss the optimized performance of thermoelectric cooling devices and how it is affected by materials properties. The discussion is based on simulations using a numerical method with one dimensional transport equations and the concept of relative current density. The coefficient of performance (COP), representing the efficiency of a device, is of key importance such that when designing a new type of device, it is the parameter to be maximized, whereas others such as the cooling power, can be set by adjusting the dimensions of the design. The COP of a single stage device under a given temperature difference, is only determined by the materials’ figure of merit zT (or z) and the Seebeck coefficient . While it is the higher the better for the former, the influence of  is complicated. While higher zTs are always preferred, materials with comparably high zT and very different  could be valuable in constructing graded legs that outperform uniform ones. Lastly, proper pairing of legs is important to ensure the materials properties are used to their full potential. 

Low‐dimensional thermoelectric materials systems are proven to possess improved thermoelectric performance, either by enhancing the power factorS²σthrough quantum confinement, or decreasing thermal conductivity with numerous interfaces. The 2D tellurium, also called tellurene, is a newly discovered 2D material which showed great potential for thermoelectric applications. In this article, high‐quality tellurene‐like nanosheets are synthesized by the hydrothermal method and assembled into nanostructured bulk materials by low‐temperature hot press, and their thermoelectric performance is investigated. Ultraviolet–ozone treatment is used to remove organic surface ligands. Doping is realized with surface doping with chalcogenidometalates. It is found that the Seebeck coefficient and the thermal conductivity of the nanosheet‐assembled bulk samples increased by ≈20% and decreased by 43% compared to bulk tellurium, respectively. Meanwhile, the carrier mobility is approaching, yet still lower than bulk tellurium. Overall, the best bulk sample possesses azTof 0.1 at room temperature which is comparable to bulk Te. By further improving the mobility, this solution processable material can provide useful thermoelectric performance for room‐temperature applications.