Search

Data preparation is the first and a very important step towards any Large Language Model (LLM) development. This paper introduces an easy-to-use, extensible, and scale-flexible open-source data preparation toolkit called Data Prep Kit (DPK). DPK is architected and designed to enable users to scale their data preparation to their needs. With DPK they can prepare data on a local machine or effortlessly scale to run on a cluster with thousands of CPU Cores. DPK comes with a highly scalable, yet extensible set of modules that transform natural language and code data. If the user needs additional transforms, they can be easily developed using extensive DPK support for transform creation. These modules can be used independently or pipelined to perform a series of operations. In this paper, we describe DPK architecture and show its performance from a small scale to a very large number of CPUs. The modules from DPK have been used for the preparation of Granite Models [1] [2]. We believe DPK is a valuable contribution to the AI community to easily prepare data to enhance the performance of their LLM models or to fine-tune models with Retrieval-Augmented Generation (RAG)., Comment: 10 pages, 7 figures

Fake news poses a significant threat to the integrity of information ecosystems and public trust. The advent of Large Language Models (LLMs) holds considerable promise for transforming the battle against fake news. Generally, LLMs represent a double-edged sword in this struggle. One major concern is that LLMs can be readily used to craft and disseminate misleading information on a large scale. This raises the pressing questions: Can LLMs easily generate biased fake news? Do all LLMs have this capability? Conversely, LLMs offer valuable prospects for countering fake news, thanks to their extensive knowledge of the world and robust reasoning capabilities. This leads to other critical inquiries: Can we use LLMs to detect fake news, and do they outperform typical detection models? In this paper, we aim to address these pivotal questions by exploring the performance of various LLMs. Our objective is to explore the capability of various LLMs in effectively combating fake news, marking this as the first investigation to analyze seven such models. Our results reveal that while some models adhere strictly to safety protocols, refusing to generate biased or misleading content, other models can readily produce fake news across a spectrum of biases. Additionally, our results show that larger models generally exhibit superior detection abilities and that LLM-generated fake news are less likely to be detected than human-written ones. Finally, our findings demonstrate that users can benefit from LLM-generated explanations in identifying fake news.

To index the increasing volume of data, modern data indexes are typically stored on SSDs and cached in DRAM. However, searching such an index has resulted in significant I/O traffic due to limited access locality and inefficient cache utilization. At the heart of index searching is the operation of filtering through vast data spans to isolate a small, relevant subset, which involves basic equality tests rather than the complex arithmetic provided by modern CPUs. This paper introduces the Search-in-Memory (SiM) chip, which demonstrates the feasibility of performing data filtering directly within a NAND flash memory chip, transmitting only relevant search results rather than complete pages. Instead of adding complex circuits, we propose repurposing existing circuitry for efficient and accurate bitwise parallel matching. We demonstrate how different data structures can use our flexible SIMD command interface to offload index searches. This strategy not only frees up the CPU for more computationally demanding tasks, but it also optimizes DRAM usage for write buffering, significantly lowering energy consumption associated with I/O transmission between the CPU and DRAM. Extensive testing across a wide range of workloads reveals up to a 9X speedup in write-heavy workloads and up to 45% energy savings due to reduced read and write I/O. Furthermore, we achieve significant reductions in median and tail read latencies of up to 89% and 85% respectively., Comment: This paper has been accepted for presentation at the The International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS) in September, 2024. An extended abstract of this paper was presented in Design, Automation & Test in Europe Conference & Exhibition (DATE), 2024

In this paper, we address the topic of envelopes created by pseudo-circle families in the Minkowski plane, which exhibit some different properties when compared with the Euclidean case. We provide solutions to all four basic problems associated with these envelopes, namely the existence problem, representation problem, problem on the number of envelopes, and problem on the relationships of definitions.

The search for baryon-number-violating (BNV) nucleon decay is an intriguing probe of new physics beyond the SM in future neutrino experiments with enhanced sensitivity. The dark sector states such as an axion or axion-like particle (ALP) can induce nucleon decays with distinct signature and kinematics from the conventional nucleon decays. In this work, we study the ALP effective field theories (EFTs) with baryon number violation and the impact of light ALP on BNV nucleon decays. We revisit the dimension-8 BNV operators in the extended EFTs with an ALP field $a$ respecting shift symmetry. The low-energy EFT operators with $|\Delta (B-L)|=2$ and $|\Delta (B-L)|=0$ are matched to the baryon chiral perturbation theory. We obtain the effective chiral Lagrangian and the BNV interactions between ALP and baryons/mesons. The ALP interactions lead to two-body baryon decays $B\to \ell~({\rm or}~\nu)~a$ and three-body nucleon decays $N\to M~\ell~({\rm or}~\nu)~a$. We obtain the constraints on the UV scale from the invisible $\Lambda^0$ decay search at BESIII, the invisible neutron decay search at KamLAND and proton decay search at Super-K. We also show the projections of some other baryon/nucleon decays and present the distinct distributions of kinematic observable., Comment: 31 pages, 4 figures, 7 tables. version accepted for publication in JHEP

Theorem 1 in Bikhchandani & Segal (2011; Theoretical Economics) suggests that a complete, transitive, monotonic, and continuous preference is regret based if and only if it is expected utility. Their Proposition 1 suggests that transitivity and continuity of a regret-based preference implies an equivalence condition: if random variables $X$ and $Y$ have the same distribution, then $X\sim Y$. We give counterexamples to Proposition 1.

Sixty years after the experimental discovery of CP violation in the quark sector, the existence of a similar CP violation in the lepton sector is still to be established. Actually, the structure of such a violation depends crucially on the origin of the neutrino masses. In an attempt at categorizing the leptonic sources of CP violation, we studied the $\nu$SM, the Standard Model extended with three generations of sterile neutrinos, that can interpolate continuously between the Dirac and Majorana scenarios of neutrino masses. In particular, we perform a classification of the Jarlskog-like flavor invariants entering CP-violating observables and we study their suppression with the heavy Majorana mass in the seesaw limit of the model. To simplify the construction of the invariants, we introduce a graph-based method. With the guidance of the Hilbert series and plethystic logarithm of the theory, we construct the \emph{generating} and \emph{primary} sets of invariants for the $\nu$SM for the first time. Unlike in the Standard Model and some other theories, we find that the numbers of generating invariants and the syzygies among them cannot immediately be read off from the plethystic logarithm, but require a more careful examination. Our analysis reveals that the \emph{generating} set contains 459 invariants, out of which 208 are CP-even and 251 are CP-odd. In the seesaw limit of the $\nu$SM, we show that all parameters of the UV theory can be captured in the effective theory with a certain suppression with the heavy Majorana mass, while these parameters can only appear in a \emph{flavor-invariant} way with a \emph{higher} mass suppression. Furthermore, we discuss how the necessary and sufficient conditions for CP violation can be captured by utilizing these invariants. Along the way, we present useful algorithms to enumerate and build the flavor invariants., Comment: 27 pages + appendices, 3 figures

The primary restrictions on 2D transition metal dichalcogenides (TMD) vdW heterostructures (vdWHs) are size limitation and alloying. Recently, a two-step vapor deposition method was reported to grow wafer-scale TMD vdWHs with little contamination [Nature 621, 499 (2023)]. In this letter, we developed a machine learning potential (MLP) which can accurately simulate the growth processes of bilayer MoS$_2$/WS$_2$ vdWHs under various conditions. Importantly, a SMMS (where M is Mo or W) structure is revealed as a highly stable intermediate easily introduces metal atom exchange and alloying. Eliminating the alloying contamination in TMD vdWHs is avoiding SMMS structure by preventing the landing of bare metal atoms. However, SMMS is revealed as an ideal electrode for MoS$_2$ FETs with low Schottky barrier.

We study a non-universal flavor scenario at the level of the Standard Model Effective Field Theory, according to which the matrix of Wilson coefficients $c_{uW}$ of an up-type electroweak quark dipole operator is aligned with the up-type Yukawa coupling. Such an alignment usually follows from the assumption of Minimal Flavor Violation (MFV), away from which we step by allowing the entries of $c_{uW}$ to be sizable along the first quark generations. A particular example, which we refer to as ``inverse hierarchy MFV", features Wilson coefficients inversely proportional to quark masses, and arises from BSM models respecting MFV and containing heavy fields that replicate the mass hierarchy of SM quarks. We then analyze the phenomenology driven by $c_{uW}$ at colliders and at lower-energy flavor experiments. We show that precision measurements of the process $pp\rightarrow W h\rightarrow \gamma\gamma\ell\nu$ at FCC-$hh$ could set an upper bound on $|c_{uW}|\lesssim\mathcal{O}(10^{-2})(\Lambda/{\rm TeV})^{2}$, with $\Lambda$ the cutoff of the effective field theory. This bound is an order of magnitude stronger than the existing LHC bounds. Moreover, we estimate that $W h\rightarrow b\bar b \ell\nu$ at HL-LHC could also give competitive bounds. In the low-energy regime, we consider bounds arising from rare kaon decays, which turn out to be loose, $|c_{uW}^{11}|<\mathcal{O}(1)(\Lambda/{\rm TeV})^{2}$. We finally demonstrate that our flavor and operator assumptions can be derived from a weakly-coupled UV model, which we choose to simultaneously illustrate the UV origin of inverse hierarchy MFV., Comment: 35 pages, 6 figures, 7 tables. Comments are welcomed

In this paper, we obtain the upper bound of the number of zeros of Abelian integral for a class of cubic Hamiltonian systems with nesting period annuli under perturbations of polynomials of degree n. Furthermore, we consider the Hopf and homoclinic bifurcation when a=-1,b=-2,c=1 and n=3, and obtain 18 distributions in which system has at least 3 limit cycles for each case.

The small atomic mass of boron indicates strong electron-phonon coupling, so it may have a brilliant performance in superconductivity. Recently, a new 2D boride sheet with ordered metal vacancies and surface terminals (Mo4/3B2-x) was realized in experiments (Science 2021, 373, 801). Here, the 2D monolayer freestanding Mo4/3B2is evidenced to be thermodynamically stable. Through electronic structure, phonon spectrum and electron-phonon coupling, monolayer Mo4/3B2 is found to be an intrinsic phonon-mediated superconductor. The superconducting transition temperature (Tc) is determined to be 4.06 K by the McMillian-Allen-Dynes formula. Remarkably, the Tc of monolayer Mo4/3B2 can be increased to 6.78 K with an appropriate biaxial tensile strain (+5%). Moreover, we predict that other transition metal replacing Mo atoms is also stable and retaining the superconductivity. Such as monolayer W4/3B2 is also a superconductor with the Tc of 2.37 K. Our research results enrich the database of 2D monolayer superconductors and boron-related formed materials science.

Deepfake videos, generated through AI faceswapping techniques, have garnered considerable attention due to their potential for powerful impersonation attacks. While existing research primarily focuses on binary classification to discern between real and fake videos, however determining the specific generation model for a fake video is crucial for forensic investigation. Addressing this gap, this paper investigates the model attribution problem of Deepfake videos from a recently proposed dataset, Deepfakes from Different Models (DFDM), derived from various Autoencoder models. The dataset comprises 6,450 Deepfake videos generated by five distinct models with variations in encoder, decoder, intermediate layer, input resolution, and compression ratio. This study formulates Deepfakes model attribution as a multiclass classification task, proposing a segment of VGG19 as a feature extraction backbone, known for its effectiveness in imagerelated tasks, while integrated a Capsule Network with a Spatio-Temporal attention mechanism. The Capsule module captures intricate hierarchies among features for robust identification of deepfake attributes. Additionally, the video-level fusion technique leverages temporal attention mechanisms to handle concatenated feature vectors, capitalizing on inherent temporal dependencies in deepfake videos. By aggregating insights across frames, our model gains a comprehensive understanding of video content, resulting in more precise predictions. Experimental results on the deepfake benchmark dataset (DFDM) demonstrate the efficacy of our proposed method, achieving up to a 4% improvement in accurately categorizing deepfake videos compared to baseline models while demanding fewer computational resources., Comment: Rejected from jounal and will have to conduct several more experiments

Facing with grave climate change and enormous energy demand, catalyzer gets more and more important due to its significant effect on reducing fossil fuels consumption. Hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) by water splitting are feasible ways to produce clean sustainable energy. Here we systematically explored atomic structures and related STM images of Se defects in PtSe2. The equilibrium fractions of vacancies under variable conditions were detailly predicted. Besides, we found the vacancies are highly kinetic stable, without recovering or aggregation. The Se vacancies in PtSe2 can dramatically enhance the HER performance, comparing with, even better than Pt(111). Beyond, we firstly revealed that PtSe2 monolayer with Se vacancies is also a good OER catalyst. The excellent bipolar catalysis of Se vacancies were further confirmed by experimental measurements. We produced defective PtSe2 by direct selenization of Pt foil at 773 K using a CVD process. Then we observed the HER and OER performance of defective PtSe2 is much highly efficient than Pt foils by a series of measurements. Our work with compelling theoretical and experimental studies indicates PtSe2 with Se defects is an ideal bipolar candidate for HER and OER.

Bilayer borophene, very recently synthesized on Ag and Cu, possesses extremely flat large surface and excellent conductivity. Besides, the van der Waals gap of bilayer borophene can be intercalated by metal atoms, thereby tailoring the properties of bilayer borophene. Herein, we propose that sandwiched B/TM/B (TM=Co, Ni, Cu, Pd) could be a new 2D formation by transiton metal atoms intercalated into bilayer borophene network, it is quiet robust with both energetic, structural and thermal stability, and exhibits heat resistance of at least 1300 K. Besides, it is novel platform for electrocatalytic hydrogen evolution reaction (HER). The interecalation metal atom serves as single-atomic catalyst, which acting the nonmetal boron layers. Beyond that, the transtion metal is protected by outside boron layers from being corroded by acidic/alkaline solution. B/Cux/B, B/Pdx/B and B/Alx/B with different metal coverage exhibit defect-independent extremely low HER free energy in the range of -0.162 ~ 0.179 eV, -0.134 ~ 0.183 eV and -0.082 ~ 0.086 eV which are comparable to noble metal Pt. Combining excellent conduction, high structural and thermal stability, low resistance to intercalated behaviour, effortless water splitting process, excellent defect-independent catalytic performance, cheapness and abundance of raw materials, free of corrodation, 2D sandwiched B/TM/B (TM=Co, Ni, Cu, Pd) is believed to promising for electrocatalytic HER applications.

Axions and axion-like particles (ALPs) are ubiquitous in popular attempts to solve supercalifragilisticexpialidocious puzzles of Nature. A widespread and vivid experimental programme spanning a vast range of mass scales and decades of couplings strives to find evidence for these elusive but theoretically well-motivated particles. In the absence of clear guiding principle, effective field theories (EFTs) prove to be an efficient tool in this experimental quest. Hilbert series technologies are a privileged instrument of the EFT toolbox to enumerate and classify operators. In this work, we compute explicitly the Hilbert series capturing the interactions of a generic ALP to the Standard Model particles above and below the electroweak symmetry scale, which allow us to build bases of operators up to dimension 8. In particular, we revealed a remarkable structure of the Hilbert series that isolates the shift-symmetry breaking and preserving interactions. In addition, with the Hilbert series method, we enumerate the sources of CP violation in terms of CP-even, CP-odd and CP-violating operators. Furthermore, we provide an ancillary file of the Hilbert series up to dimension 15 to supplement our findings, which can be used for further analysis and exploration., Comment: 33 pages + appendices, 2 figures, 13 tables, added discussion about CP, updated the ancillary file

High-energy muon collider can play as an emitter of electroweak gauge bosons and thus leads to substantial vector boson scattering (VBS) processes. In this work, we investigate the production of heavy neutral lepton (HNL) $N$ and lepton number violation (LNV) signature through VBS at high-energy muon colliders. VBS induces LNV processes $W^\pm Z/\gamma\to \ell^\pm N \to \ell^\pm \ell^\pm W^\mp\to \ell^\pm \ell^\pm q\bar{q}'$ with an on-shell HNL $N$ at $\mu^+\mu^-$ colliders. In analogy to neutrinoless double-beta decay with the HNL in t-channel, the LNV signature $W^+W^+\to \ell^+\ell^+$ can also happen via VBS at same-sign muon collider. They provide clean and robust LNV signatures to tell the nature of Majorana HNLs and thus have more advantageous benefits than direct $\mu\mu$ annihilation. We analyze the potential of searching for Majorana HNL and obtain the exclusion limits on mixing $V_{\ell N}$. Based on this same-sign lepton signature, we also obtain the sensitivity of muon collider to the Weinberg operator., Comment: 24 pages, 8 figures, 2 tables. Accepted for publication in JHEP