arXiv subjects

Dean Lee

Publications and source records attributed to Dean Lee.

At least 37 records · Page 2Linked to original sources

arXiv2025

MultiNRC: A Challenging and Native Multilingual Reasoning Evaluation Benchmark for LLMs

Although recent Large Language Models (LLMs) have shown rapid improvement on reasoning benchmarks in English, the evaluation of such LLMs' multilingual reasoning capability across diverse languages and cultural contexts remains limited. Existing multilingual reasoning benchmarks are typically constructed by translating existing English reasoning benchmarks, biasing these benchmarks towards reasoning problems with context in English language/cultures. In this work, we introduce the Multilingual Native Reasoning Challenge (MultiNRC), a benchmark designed to assess LLMs on more than 1,000 native, linguistic and culturally grounded reasoning questions written by native speakers in French, Spanish, and Chinese. MultiNRC covers four core reasoning categories: language-specific linguistic reasoning, wordplay & riddles, cultural/tradition reasoning, and math reasoning with cultural relevance. For cultural/tradition reasoning and math reasoning with cultural relevance, we also provide English equivalent translations of the multilingual questions by manual translation from native speakers fluent in English. This set of English equivalents can provide a direct comparison of LLM reasoning capacity in other languages vs. English on the same reasoning questions. We systematically evaluate current 14 leading LLMs covering most LLM families on MultiNRC and its English equivalent set. The results show that (1) current LLMs are still not good at native multilingual reasoning, with none scoring above 50% on MultiNRC; (2) LLMs exhibit distinct strengths and weaknesses in handling linguistic, cultural, and logical reasoning tasks; (3) Most models perform substantially better in math reasoning in English compared to in original languages (+10%), indicating persistent challenges with culturally grounded knowledge.

cs.CL↗

arXiv2025

Quantifying alpha clustering in the ground states of 16-O and 20-Ne

Understanding the role of multi-nucleon correlations in the structure of light nuclei is at the forefront of modern nuclear science. In this letter, we present a quantitative benchmark study of alpha-cluster correlations in the ground states of 16-O and 20-Ne. Experimental data provide direct evidence that the wave functions of the ground states of 16-O and 20-Ne are dominated by alpha-cluster correlations, in agreement with the predictions of sophisticated nuclear structure models. We also provide a new model-independent constraint for the alpha asymptotic normalization coefficient of the 16-O ground state and discuss the implications of these findings on the 12-C(alpha,gamma)16-O reaction, which is of critical importance for nuclear astrophysics.

nucl-ex↗

arXiv2025

Nuclear Physics Confronts Relativistic Collisions Of Isobars

High-energy collisions involving the $A=96$ isobars $^{96}$Zr and $^{96}$Ru have been performed in 2018 at Brookhaven National Laboratory's Relativistic Heavy Ion Collider (RHIC) as a means to search for the chiral magnetic effect in QCD. This would manifest itself as specific deviations from unity in the ratio of observables taken between $^{96}$Zr+$^{96}$Zr and $^{96}$Ru+$^{96}$Ru collisions. Measurements of such ratios (released at the end of 2021) indeed reveal deviations from unity, but these are primarily caused by the two collided isobars having different radial profiles and intrinsic deformations. To make progress in understanding RHIC data, nuclear physicists across the energy spectrum gathered in Heidelberg in 2022 as part of an EMMI Rapid Reaction Task Force (RRTF) to address the following question. Does the combined effort of low-energy nuclear structure physics and high-energy heavy-ion physics enable us to understand the observations made in isobar collisions at RHIC?

nucl-ex↗

arXiv2025

Ab initio study of the beryllium isotopes $^{7}$Be to $^{12}$Be

We present a systematic ab initio study of the low-lying states in beryllium isotopes from 7Be to 12Be using nuclear lattice effective field theory with the N3LO interaction. Our calculations achieve good agreement with experimental data for energies, radii, and electromagnetic properties. We introduce a novel, model-independent method to quantify nuclear shapes, uncovering a distinct pattern in the interplay between positive and negative parity states across the isotopic chain. By combining Monte Carlo sampling of the many-body density operator with a novel nucleon-grouping algorithm, the prominent two-center cluster structures, the emergence of one-neutron halo, complex nuclear molecular dynamics such as π orbital and σ orbital, emerge naturally.

nucl-th↗

arXiv2025

MultiChallenge: A Realistic Multi-Turn Conversation Evaluation Benchmark Challenging to Frontier LLMs

We present MultiChallenge, a pioneering benchmark evaluating large language models (LLMs) on conducting multi-turn conversations with human users, a crucial yet underexamined capability for their applications. MultiChallenge identifies four categories of challenges in multi-turn conversations that are not only common and realistic among current human-LLM interactions, but are also challenging to all current frontier LLMs. All 4 challenges require accurate instruction-following, context allocation, and in-context reasoning at the same time. We also develop LLM as judge with instance-level rubrics to facilitate an automatic evaluation method with fair agreement with experienced human raters. Despite achieving near-perfect scores on existing multi-turn evaluation benchmarks, all frontier models have less than 50% accuracy on MultiChallenge, with the top-performing Claude 3.5 Sonnet (June 2024) achieving just a 41.4% average accuracy.

cs.CL↗

arXiv2025

Anisotropic flow in fixed-target $^{208}$Pb+$^{20}$Ne collisions as a probe of quark-gluon plasma

The System for Measuring Overlap with Gas (SMOG2) at the LHCb detector enables the study of fixed-target ion-ion collisions at relativistic energies ($\sqrt{s_{\rm NN}}\sim100$ GeV in the centre-of-mass). With input from \textit{ab initio} calculations of the structure of $^{16}$O and $^{20}$Ne, we compute 3+1D hydrodynamic predictions for the anisotropic flow of Pb+Ne and Pb+O collisions, to be tested with upcoming LHCb data. This will allow the detailed study of quark-gluon plasma (QGP) formation as well as experimental tests of the predicted nuclear shapes. Elliptic flow ($v_2$) in Pb+Ne collisions is greatly enhanced compared to the Pb+O baseline due to the shape of $^{20}$Ne, which is deformed in a bowling-pin geometry. Owing to the large $^{208}$Pb radius, this effect is seen in a broad centrality range, a unique feature of this collision configuration. Larger elliptic flow further enhances the quadrangular flow ($v_4$) of Pb+Ne collisions via non-linear coupling, and impacts the sign of the kurtosis of the elliptic flow vector distribution ($c_2\{4\}$). Exploiting the shape of $^{20}$Ne proves thus an ideal method to investigate the formation of QGP in fixed-target experiments at LHCb, and demonstrates the power of SMOG2 as a tool to image nuclear ground states.

nucl-th↗

arXiv2025

EnigmaEval: A Benchmark of Long Multimodal Reasoning Challenges

As language models master existing reasoning benchmarks, we need new challenges to evaluate their cognitive frontiers. Puzzle-solving events are rich repositories of challenging multimodal problems that test a wide range of advanced reasoning and knowledge capabilities, making them a unique testbed for evaluating frontier language models. We introduce EnigmaEval, a dataset of problems and solutions derived from puzzle competitions and events that probes models' ability to perform implicit knowledge synthesis and multi-step deductive reasoning. Unlike existing reasoning and knowledge benchmarks, puzzle solving challenges models to discover hidden connections between seemingly unrelated pieces of information to uncover solution paths. The benchmark comprises 1184 puzzles of varying complexity -- each typically requiring teams of skilled solvers hours to days to complete -- with unambiguous, verifiable solutions that enable efficient evaluation. State-of-the-art language models achieve extremely low accuracy on these puzzles, even lower than other difficult benchmarks such as Humanity's Last Exam, unveiling models' shortcomings when challenged with problems requiring unstructured and lateral reasoning.

cs.AI↗

arXiv2025

Lattice Effective Field Theory Simulations of Nuclei

Lattice effective field theory applies the principles of effective field theory in a lattice framework where space and time are discretized. Nucleons are placed on the lattice sites, and the interactions are tuned to replicate the observed features of the nuclear force. Monte Carlo simulations are then employed to predict the properties of nuclear few- and many-body systems. We review the basic methods and several theoretical and algorithmic advances that have been used to further our understanding of atomic nuclei.

nucl-th↗

arXiv2025

Parametric Matrix Models

We present a general class of machine learning algorithms called parametric matrix models. In contrast with most existing machine learning models that imitate the biology of neurons, parametric matrix models use matrix equations that emulate physical systems. Similar to how physics problems are usually solved, parametric matrix models learn the governing equations that lead to the desired outputs. Parametric matrix models can be efficiently trained from empirical data, and the equations may use algebraic, differential, or integral relations. While originally designed for scientific computing, we prove that parametric matrix models are universal function approximators that can be applied to general machine learning problems. After introducing the underlying theory, we apply parametric matrix models to a series of different challenges that show their performance for a wide range of problems. For all the challenges tested here, parametric matrix models produce accurate results within an efficient and interpretable computational framework that allows for input feature extrapolation.

cs.LG↗

arXiv2024

Imaging the initial condition of heavy-ion collisions and nuclear structure across the nuclide chart

High-energy nuclear collisions encompass three key stages: the structure of the colliding nuclei informed by low-energy nuclear physics, the initial condition (IC) leading to the formation of quark-gluon plasma (QGP), and the hydrodynamic expansion and hadronization of the QGP leading to final-state hadrons observed experimentally. Recent advances in experimental and theoretical methods have ushered in a precision era, enabling an increasingly accurate understanding of these stages. However, most approaches involve simultaneously determining both QGP properties and initial conditions from a single collision system, creating complexity due to the coupled contributions of various stages to the final-state observables. To avoid this, we propose leveraging known knowledge of low-energy nuclear structure and hydrodynamic observables to constrain the IC independently. By conducting comparative studies of collisions involving isobar-like nuclei - species with similar mass numbers but different structures - we disentangle the initial condition's impacts from the QGP properties. This approach not only refines our understanding of the IC but also turns high-energy experiments into a precision tool for imaging nuclear structures, offering insights that complement traditional low-energy approaches. Opportunities for carrying out such comparative experiments at the LHC and other facilities could significantly advance both high-energy and low-energy nuclear physics. Additionally, this approach has implications for the future EIC. While the possibilities are extensive, we focus on selected proposals that could benefit both the high-energy and low-energy nuclear physics communities. Originally prepared as input for the long-range plan of U.S. nuclear physics, this white paper reflects the status as of September 2022, with a brief update on developments since then.

nucl-ex↗

arXiv2024

Non-perturbative three-nucleon simulation using chiral lattice EFT

We study the three-nucleon system at next-to-next-to-next-to-leading order ($\mathrm{N^3LO}$) in the framework of chiral effective field theory (EFT) on the lattice. Our calculations do not rely on a perturbative treatment of subleading contributions to the nuclear forces. For the two-nucleon potential, we apply the previously developed $\mathrm{N^3LO}$ lattice interaction. For the leading contribution to the three-nucleon force, we determine the two low-energy constants (LECs) in the contact interactions by adjusting the ground state energy and half-life of triton, where the latter employs the nuclear axial current at $\mathrm{N^2LO}$ in chiral EFT. Additionally, the ground state energy of helion and the charge radii of the two considered nuclei are computed. No effect of the smearing regularization in the three-nucleon contact interaction is observed here. We compare our results with recent lattice-EFT calculations that are based on a potential tuned to light and medium-mass nuclei using the wave-function-matching technique to circumvent the Monte-Carlo sign problem.

nucl-th↗

arXiv2024

A Careful Examination of Large Language Model Performance on Grade School Arithmetic

Large language models (LLMs) have achieved impressive success on many benchmarks for mathematical reasoning. However, there is growing concern that some of this performance actually reflects dataset contamination, where data closely resembling benchmark questions leaks into the training data, instead of true reasoning ability. To investigate this claim rigorously, we commission Grade School Math 1000 (GSM1k). GSM1k is designed to mirror the style and complexity of the established GSM8k benchmark, the gold standard for measuring elementary mathematical reasoning. We ensure that the two benchmarks are comparable across important metrics such as human solve rates, number of steps in solution, answer magnitude, and more. When evaluating leading open- and closed-source LLMs on GSM1k, we observe accuracy drops of up to 8%, with several families of models showing evidence of systematic overfitting across almost all model sizes. Further analysis suggests a positive relationship (Spearman's r^2 = 0.36) between a model's probability of generating an example from GSM8k and its performance gap between GSM8k and GSM1k, suggesting that some models may have partially memorized GSM8k. Nevertheless, many models, especially those on the frontier, show minimal signs of overfitting, and all models broadly demonstrate generalization to novel math problems guaranteed to not be in their training data.

cs.CL↗

arXiv2024

Motivations for Early High-Profile FRIB Experiments

This white paper is the result of a collaboration by those that attended a workshop at the Facility for Rare Isotope Beams (FRIB), organized by the FRIB Theory Alliance (FRIB-TA), on Theoretical Justifications and Motivations for Early High-Profile FRIB Experiments. It covers a wide range of topics related to the science that will be explored at FRIB. After a brief introduction, the sections address: (II) Overview of theoretical methods, (III) Experimental capabilities, (IV) Structure, (V) Near-threshold Physics, (VI) Reaction mechanisms, (VII) Nuclear equations of state, (VIII) Nuclear astrophysics, (IX) Fundamental symmetries, and (X) Experimental design and uncertainty quantification.

nucl-th↗

arXiv2024

Efficient State Preparation for the Schwinger Model with a Theta Term

We present a comparison of different quantum state preparation algorithms and their overall efficiency for the Schwinger model with a theta term. While adiabatic state preparation (ASP) is proved to be effective, in practice it leads to large CNOT gate counts to prepare the ground state. The quantum approximate optimization algorithm (QAOA) provides excellent results while keeping the CNOT counts small by design, at the cost of an expensive classical minimization process. We introduce a ``blocked'' modification of the Schwinger Hamiltonian to be used in the QAOA that further decreases the length of the algorithms as the size of the problem is increased. The rodeo algorithm (RA) provides a powerful tool to efficiently prepare any eigenstate of the Hamiltonian, as long as its overlap with the initial guess is large enough. We obtain the best results when combining the blocked QAOA ansatz and the RA, as this provides an excellent initial state with a relatively short algorithm without the need to perform any classical steps for large problem sizes.

hep-lat↗

arXiv2024

Accelerating Eigenvalue Computation for Nuclear Structure Calculations via Perturbative Corrections

We present a new method for computing the lowest few eigenvalues and the corresponding eigenvectors of a nuclear many-body Hamiltonian represented in a truncated configuration interaction subspace, i.e., the no-core shell model (NCSM). The method uses the hierarchical structure of the NCSM Hamiltonian to partition the Hamiltonian as the sum of two matrices. The first matrix corresponds to the Hamiltonian represented in a small configuration space, whereas the second is viewed as the perturbation to the first matrix. Eigenvalues and eigenvectors of the first matrix can be computed efficiently. Perturbative corrections to the eigenvectors of the first matrix can be obtained from the solutions of a sequence of linear systems of equations defined in the small configuration space. These correction vectors can be combined with the approximate eigenvectors of the first matrix to construct a subspace from which more accurate approximations of the desired eigenpairs can be obtained. We call this method a Subspace Projection with Perturbative Corrections (SPPC) method. We show by numerical examples that the SPPC method can be more efficient than conventional iterative methods for solving large-scale eigenvalue problems such as the Lanczos, block Lanczos and the locally optimal block preconditioned conjugate gradient (LOBPCG) method. The method can also be combined with other methods to avoid convergence stagnation.

physics.comp-ph↗

arXiv2024

Structure Factors for Hot Neutron Matter from Ab Initio Lattice Simulations with High-Fidelity Chiral Interactions

We present the first ab initio lattice calculations of spin and density correlations in hot neutron matter using high-fidelity interactions at next-to-next-to-next-to-leading order (N3LO) in chiral effective field theory. These correlations have a large impact on neutrino heating and shock revival in core-collapse supernovae and are encapsulated in functions called structure factors. Unfortunately, calculations of structure factors using high-fidelity chiral interactions were well out of reach using existing computational methods. In this work, we solve the problem using a computational approach called the rank-one operator (RO) method. The RO method is a general technique with broad applications to simulations of fermionic many-body systems. It solves the problem of exponential scaling of computational effort when using perturbation theory for higher-body operators and higher-order corrections. Using the RO method, we compute the vector and axial static structure factors for hot neutron matter as a function of temperature and density. The ab initio lattice results are in good agreement with virial expansion calculations at low densities but are more reliable at higher densities. Random phase approximation codes used to estimate neutrino opacity in core-collapse supernovae simulations can now be calibrated with ab initio lattice calculations.

nucl-th↗

arXiv2024

Conformal duality of the nonlinear Schrödinger equation: Theory and applications to parameter estimation

The nonlinear Schrödinger equation (NLSE) is a rich and versatile model, which in one spatial dimension has stationary solutions similar to those of the linear Schrödinger equation as well as more exotic solutions such as solitary waves and quantum droplets. Here we present the unified theory of the NLSE, showing that all stationary solutions of the local one-dimensional cubic-quintic NLSE can be classified according to a single number called the cross-ratio. Any two solutions with the same cross-ratio can be converted into one another using a conformal transformation, and the same also holds true for traveling wave solutions. Further, we introduce an optimization afterburner that relies on this conformal symmetry to substantially improve NLSE parameter estimation from noisy empirical data. The new method therefore should have far reaching practical applications for nonlinear physical systems.

cond-mat.quant-gas↗

arXiv2024

Demonstration of the Rodeo Algorithm on a Quantum Computer

The rodeo algorithm is an efficient algorithm for eigenstate preparation and eigenvalue estimation for any observable on a quantum computer. This makes it a promising tool for studying the spectrum and structure of atomic nuclei as well as other fields of quantum many-body physics. The only requirement is that the initial state has sufficient overlap probability with the desired eigenstate. While it is exponentially faster than well-known algorithms such as phase estimation and adiabatic evolution for eigenstate preparation, it has yet to be implemented on an actual quantum device. In this work, we apply the rodeo algorithm to determine the energy levels of a random one-qubit Hamiltonian, resulting in a relative error of $0.08\%$ using mid-circuit measurements on the IBM Q device Casablanca. This surpasses the accuracy of directly-prepared eigenvector expectation values using the same quantum device. We take advantage of the high-accuracy energy determination and use the Hellmann-Feynman theorem to compute eigenvector expectation values for a different random one-qubit observable. For the Hellmann-Feynman calculations, we find a relative error of $0.7\%$. We conclude by discussing possible future applications of the rodeo algorithm for multi-qubit Hamiltonians.

quant-ph↗