arXiv subjects

Keyu Pan

Publications and source records attributed to Keyu Pan.

5 recordsLinked to original sources

arXiv2026

How Foundational Skills Influence VLM-based Embodied Agents:A Native Perspective

Recent advances in vision-language models (VLMs) have shown promise for human-level embodied intelligence. However, existing benchmarks for VLM-driven embodied agents often rely on high-level commands or discretized action spaces, which are non-native settings that differ markedly from real-world control. In addition, current benchmarks focus primarily on high-level tasks and lack joint evaluation and analysis at both low and high levels. To address these limitations, we present NativeEmbodied, a challenging benchmark for VLM-driven embodied agents that uses a unified, native low-level action space. Built on diverse simulated scenes, NativeEmbodied includes three representative high-level tasks in complex scenarios to evaluate overall performance. For more detailed analysis, we further decouple the skills required by complex tasks and construct four types of low-level tasks, each targeting a fundamental embodied skill. This joint evaluation across task and skill granularities enables fine-grained assessment of embodied agents. Experiments with state-of-the-art VLMs reveal clear deficiencies in several fundamental embodied skills, and further analysis shows that these bottlenecks significantly limit performance on high-level tasks. NativeEmbodied highlights key challenges for current VLM-driven embodied agents and provides insights to guide future research.

cs.AI↗

arXiv2025

\(X\)-evolve: Solution space evolution powered by large language models

While combining large language models (LLMs) with evolutionary algorithms (EAs) shows promise for solving complex optimization problems, current approaches typically evolve individual solutions, often incurring high LLM call costs. We introduce \(X\)-evolve, a paradigm-shifting method that instead evolves solution spaces \(X\) (sets of individual solutions) - subsets of the overall search space \(S\). In \(X\)-evolve, LLMs generate tunable programs wherein certain code snippets, designated as parameters, define a tunable solution space. A score-based search algorithm then efficiently explores this parametrically defined space, guided by feedback from objective function scores. This strategy enables broader and more efficient exploration, which can potentially accelerate convergence at a much lower search cost, requiring up to two orders of magnitude fewer LLM calls than prior leading methods. We demonstrate \(X\)-evolve's efficacy across three distinct hard optimization problems. For the cap set problem, we discover a larger partial admissible set, establishing a new tighter asymptotic lower bound for the cap set constant (\(C \ge 2.2203\)). In information theory, we uncover a larger independent set for the 15-vertex cycle graph (\(\mathcal{C}_{15}^{\boxtimes 5}\), size 19,946), thereby raising the known lower bound on its Shannon capacity. Furthermore, for the NP-hard online bin packing problem, we generate heuristics that consistently outperform standard strategies across established benchmarks. By evolving solution spaces, our method considerably improves search effectiveness, making it possible to tackle high-dimensional problems that were previously computationally prohibitive.

cs.AI↗

arXiv2024

Observation of Klein bottle quadrupole topological insulators in electric circuits

The Klein bottle Benalcazar-Bernevig-Hughes (BBH) insulator phase plays a pivotal role in understanding higher-order topological phases. The insulator phase is characterized by a unique feature: a nonsymmorphic glide symmetry that exists within momentum space, rather than real space. This characteristic transforms the Brillouin zone's fundamental domain into a structure of Klein bottle. Here, we report an observation of a Klein bottle topoelectrical model under gauge fields. To provide a comprehensive understanding of the different corner distributions of odd and even unit cells, we present theoretical calculations and demonstrate that the symmetry properties significantly affect the topological nature. These theoretical predictions are confirmed by experimental results, which demonstrate the practical feasibility of such topological configurations in electronic circuits. Our work establishes a vital connection between the realms of condensed matter physics and circuit systems, thereby paving a pathway for investigating exotic condensed matter physics.

cond-mat.mes-hall↗

arXiv2024

Observation of Young's double-slit phenomenon in anti-PT-symmetric electrical circuits

In the last few decades, interference has been extensively studied in both the quantum and classical fields, which reveals light volatility and is widely used for high-precision measurements. We have put forward the phenomenon in which the discrete diffraction and interference phenomena, presented by the time-varying voltage of a Su-Schrieffer-Heeger (SSH) circuit model with an anti-PT (APT) symmetry. To demonstrate Young's double-slit phenomenon in an APT circuit, we initially explore the coupled mode theory (CMT) of voltage in the broken phase, observe discrete diffraction under single excitation and interference under double excitations. Furthermore, we design a phase-shifting circuit to observe the effects of phase difference and distance on discrete interference. Our work combines the effects in optics with condensed matter physics, show the Young's double-slit phenomenon in electrical circuits theoretically and experimentally.

cond-mat.other↗

arXiv2023

Do LLMs Possess a Personality? Making the MBTI Test an Amazing Evaluation for Large Language Models

The field of large language models (LLMs) has made significant progress, and their knowledge storage capacity is approaching that of human beings. Furthermore, advanced techniques, such as prompt learning and reinforcement learning, are being employed to address ethical concerns and hallucination problems associated with LLMs, bringing them closer to aligning with human values. This situation naturally raises the question of whether LLMs with human-like abilities possess a human-like personality? In this paper, we aim to investigate the feasibility of using the Myers-Briggs Type Indicator (MBTI), a widespread human personality assessment tool, as an evaluation metric for LLMs. Specifically, extensive experiments will be conducted to explore: 1) the personality types of different LLMs, 2) the possibility of changing the personality types by prompt engineering, and 3) How does the training dataset affect the model's personality. Although the MBTI is not a rigorous assessment, it can still reflect the similarity between LLMs and human personality. In practice, the MBTI has the potential to serve as a rough indicator. Our codes are available at https://github.com/HarderThenHarder/transformers_tasks/tree/main/LLM/llms_mbti.

cs.CL↗