arXiv subjects

Xiaoliang Qi

Publications and source records attributed to Xiaoliang Qi.

6 recordsLinked to original sources

arXiv2026

CMT-Benchmark: A Benchmark for Condensed Matter Theory Built by Expert Researchers

Large language models (LLMs) have shown remarkable progress in coding and math problem-solving, but evaluation on advanced research-level problems in hard sciences remains scarce. To fill this gap, we present CMT-Benchmark, a dataset of 50 problems covering condensed matter theory (CMT) at the level of an expert researcher. Topics span analytical and computational approaches in quantum many-body, and classical statistical mechanics. The dataset was designed and verified by a panel of expert researchers from around the world. We built the dataset through a collaborative environment that challenges the panel to write and refine problems they would want a research assistant to solve, including Hartree-Fock, exact diagonalization, quantum/variational Monte Carlo, density matrix renormalization group (DMRG), quantum/classical statistical mechanics, and model building. We evaluate LLMs by programmatically checking solutions against expert-supplied ground truth. We developed machine-grading, including symbolic handling of non-commuting operators via normal ordering. They generalize across tasks too. Our evaluations show that frontier models struggle with all of the problems in the dataset, highlighting a gap in the physical reasoning skills of current LLMs. Notably, experts identified strategies for creating increasingly difficult problems by interacting with the LLMs and exploiting common failure modes. The best model, GPT5, solves 30\% of the problems; average across 17 models (GPT, Gemini, Claude, DeepSeek, Llama) is 11.4\pm2.1\%. Moreover, 18 problems are solved by none of the 17 models, and 26 by at most one. These unsolved problems span Quantum Monte Carlo, Variational Monte Carlo, and DMRG. Answers sometimes violate fundamental symmetries or have unphysical scaling dimensions. We believe this benchmark will guide development toward capable AI research assistants and tutors.

cs.LG↗

arXiv2025

Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems

The advent of large language models (LLMs) has catalyzed a transformative shift in artificial intelligence, paving the way for advanced intelligent agents capable of sophisticated reasoning, robust perception, and versatile action across diverse domains. As these agents increasingly drive AI research and practical applications, their design, evaluation, and continuous improvement present intricate, multifaceted challenges. This book provides a comprehensive overview, framing intelligent agents within modular, brain-inspired architectures that integrate principles from cognitive science, neuroscience, and computational research. We structure our exploration into four interconnected parts. First, we systematically investigate the modular foundation of intelligent agents, systematically mapping their cognitive, perceptual, and operational modules onto analogous human brain functionalities and elucidating core components such as memory, world modeling, reward processing, goal, and emotion. Second, we discuss self-enhancement and adaptive evolution mechanisms, exploring how agents autonomously refine their capabilities, adapt to dynamic environments, and achieve continual learning through automated optimization paradigms. Third, we examine multi-agent systems, investigating the collective intelligence emerging from agent interactions, cooperation, and societal structures. Finally, we address the critical imperative of building safe and beneficial AI systems, emphasizing intrinsic and extrinsic security threats, ethical alignment, robustness, and practical mitigation strategies necessary for trustworthy real-world deployment. By synthesizing modular AI architectures with insights from different disciplines, this survey identifies key research challenges and opportunities, encouraging innovations that harmonize technological advancement with meaningful societal benefit.

cs.AI↗

arXiv2013

Recent developments in transport phenomena in Weyl semimetals

The last decade has witnessed great advancements in the science and engineering of systems with unconventional band structures, seeded by studies of graphene and topological insulators. While the band structure of graphene simulates massless relativistic electrons in two dimensions, topological insulators have bands that wind non-trivially over momentum space in a certain abstract sense. Over the last couple of years, enthusiasm has been burgeoning in another unconventional and topological (although, not quite in the same sense as topological insulators) phase -- the Weyl Semimetal. In this phase, electrons mimic Weyl fermions that are well-known in high-energy physics, and inherit many of their properties, including an apparent violation of charge conservation known as the Chiral Anomaly. In this review, we recap some of the unusual transport properties of Weyl semimetals discussed in the literature so far, focusing on signatures whose roots lie in the anomaly. We also mention several proposed realizations of this phase in condensed matter systems, since they were what arguably precipitated activity on Weyl semimetals in the first place.

cond-mat.str-el↗

arXiv2011

Electron fractionalization and unconventional order parameters of the t-J model

In the $t-J$ model, the electron fractionalization is unique due to the non-perturbative phase string effect. We formulated a lattice field theory taking this effect into full account. Basing on this field theory, we introduced a pair of Wilson loops which constitute a complete set of order parameters determining the phase diagram in the underdoped regime. We also established a general composition rule for electric transport expressing the electric conductivity in terms of the spinon and the holon conductivities. The general theory is applied to studies of the quantum phase diagram. We found that the antiferromagnetic and the superconducting phases are dual: in the former, holons are confined while spinons are deconfined, and {\it vice versa} in the latter. These two phases are separated by a novel phase, the so-called Bose-insulating phase, where both holons and spinons are deconfined and the system is electrically insulating.

cond-mat.str-el↗

arXiv2010

Crossover of Three-Dimensional Topological Insulator of Bi2Se3 to the Two-Dimensional Limit

Bi2Se3 is theoretically predicted1 2and experimentally observed2,3 to be a three dimensional topological insulator. For possible applications, it is important to understand the electronic structure of the planar device. In this work, thickness dependent band structure of molecular beam epitaxy grown ultrathin films of Bi2Se3 is investigated by in situ angle-resolved photoemission spectroscopy. An energy gap is observed for the first time in the topologically protected metallic surface states of bulk Bi2Se3 below the thickness of six quintuple layers, due to the coupling between the surface states from two opposite surfaces of the Bi2Se3 film. The gapped surface states exhibit sizable Rashba-type spin-orbit splitting, due to breaking of structural inversion symmetry induced by SiC substrate. The spin-splitting can be controlled by tuning the potential difference between the two surfaces.

cond-mat.mtrl-sci↗

arXiv2009

Dynamical Axion Field in Topological Magnetic Insulators

Axions are very light, very weakly interacting particles postulated more than 30 years ago in the context of the Standard Model of particle physics. Their existence could explain the missing dark matter of the universe. However, despite intensive searches, they have yet to be detected. In this work, we show that magnetic fluctuations of topological insulators couple to the electromagnetic fields exactly like the axions, and propose several experiments to detect this dynamical axion field. In particular, we show that the axion coupling enables a nonlinear modulation of the electromagnetic field, leading to attenuated total reflection. We propose a novel optical modulators device based on this principle.

cond-mat.other↗