Search

We present BabyLlama-2, a 345 million parameter model distillation-pretrained from two teachers on a 10 million word corpus for the BabyLM competition. On BLiMP and SuperGLUE benchmarks, BabyLlama-2 outperforms baselines trained on both 10 and 100 million word datasets with the same data mix, as well as its teacher models. Through an extensive hyperparameter sweep, we demonstrate that the advantages of distillation cannot be attributed to suboptimal hyperparameter selection of the teachers. Our findings underscore the need for further investigation into distillation techniques, particularly in data-limited settings., Comment: 9 pages, 3 figures, 5 tables, submitted to the BabyLM Challenge (CoNLL 2024 Shared Task)

Axion-like particles (ALPs) and heavy neutral leptons (HNLs) are two well-motivated classes of particles beyond the Standard Model. It is intriguing to explore the new detection opportunities that may arise if both particle types coexist. Part of the authors already investigated this scenario in a previous publication, within a simplified model containing an ALP and a single HNL, identifying particularly promising processes that could be searched for at the LHC. In this paper, we first consider the same setup with a broader range of both production processes and final states, both at the High-Luminosity LHC and at a future muon collider. Subsequently, we expand it to the more realistic scenario with at least two HNLs, necessary to describe the active neutrino masses. Different phenomenological signals are expected and we examine the complexities that emerge in this setup. This study paves the way for dedicated analysis at (forthcoming) colliders, potentially pinpointing the dynamics of ALPs and HNLs., Comment: 43 pages, 12 figures

The increasing interest in Long-Lived Particles (LLPs) has led to numerous proposed experiments in order to search for them. However, the sensitivity estimates published by these experiments tend to rely on disparate assumptions. To ensure an accurate comparison of their potential to find LLPs, a unified estimation of their sensitivity is therefore required. In this contribution, we introduce \texttt{SensCalc}, a \texttt{Mathematica}-based code that uses a semi-analytic approach to calculate the event rate of GeV-scale LLPs, and we present several case studies., Comment: EPS-HEP 2023 proceedings

We present our submission to the BabyLM challenge, whose goal was to improve the sample efficiency of language models. We trained an ensemble consisting of a GPT-2 and small LLaMA models on the developmentally-plausible, 10M-word BabyLM dataset, then distilled it into a small, 58M-parameter LLaMA model, which exceeds in performance both of its teachers as well as a similar model trained without distillation. This suggests that distillation can not only retain the full performance of the teacher model when the latter is trained on a sufficiently small dataset; it can exceed it, and lead to significantly better performance than direct training., Comment: 11 pages, 4 figures, 4 tables, submitted to the BabyLM Challenge and accepted as archival full paper (CoNLL--CMCL 2023 Shared Task), checkpoint available at https://huggingface.co/timinar/baby-llama-58m, training code available at https://github.com/timinar/BabyLlama

Axion-like particles (ALPs) and heavy neutral leptons (HNLs) are both well-motivated extensions of the Standard Model. As ALPs couple to on-shell fermions proportionally to their masses, processes involving both types of particles may give rise, for TeV HNLs, to relevant phenomenology at colliders. In this work, we point out a particularly clean process, whose final state consists of four jets and two charged leptons, and we estimate its current and future sensitivity at the LHC. For on-shell HNLs, i) there is no dependence on the overall scale of the mixing between HNLs and the active neutrinos, making this process sensitive down to the type-I Seesaw line; ii) the signal strength of the process is sizable only as far as the HNL masses are below a few TeVs, contrary to what the proportionality to masses of the ALP couplings may suggest. Although ALPs and HNLs have been mainly studied independently in the literature, considering their interplay may lead to joint limits that are much stronger than those on the individual particles taken separately. This concise study paves the way for similar searches at colliders and other experiments., Comment: 7 pages, 5 figures

The existence of non-zero neutrino masses points to the likely existence of multiple SM neutral fermions. When such states are heavy enough that they cannot be produced in oscillations, they are referred to as Heavy Neutral Leptons (HNLs). In this white paper we discuss the present experimental status of HNLs including colliders, beta decay, accelerators, as well as astrophysical and cosmological impacts. We discuss the importance of continuing to search for HNLs, and its potential impact on our understanding on key fundamental questions, and additionally we outline the future prospects for next-generation future experiments or upcoming accelerator run scenarios., Comment: 82 pages, 34 figures. Contribution to Snowmass 2021

Experiments searching for Heavy Neutral Leptons (HNLs) typically interpret their results within simplified models consisting of a single HNL coupled to a single lepton flavor. However, any model which aims to describe neutrino oscillations necessarily features more than one HNL, coupled to several flavors. As we show in this work, the reinterpretation of the results of experimental searches in terms of realistic models is a non-trivial task. We perform a detailed reinterpretation of the latest ATLAS search for prompt HNLs in W decays within a minimal low-scale seesaw with two HNLs. We show that the exclusion limits obtained using the detailed reinterpretation can differ by several orders of magnitude from the limits quoted for the simplified models. Hence naively comparing the mixing angles from a realistic model to the reported limits could lead to wrongly excluding entire regions of parameter space! To overcome this issue without requiring experiments to report constraints on all possible HNL models, we propose a simple framework that allows one to easily and accurately reinterpret exclusion limits within closely-related models. We outline a number of concrete steps that can be taken by experiments to implement this method with minimal effort, and we discuss its applicability to other models of feebly interacting particles., Comment: 6 pages, 4 figures, contribution to the proceedings of the European Physical Society conference on high energy physics 2021 (EPS-HEP2021), July 26-30 2021, online

Heavy neutral leptons (HNLs) are hypothetical particles, motivated in the first place by their ability to explain neutrino oscillations. Experimental searches for HNLs are typically conducted under the assumption of a single HNL mixing with a single neutrino flavor. However, the resulting exclusion limits may not directly constrain the corresponding mixing angles in realistic HNL models -- those which can explain neutrino oscillations. The reinterpretation of the results of these experimental searches turns out to be a non-trivial task, that requires significant knowledge of the details of the experiment. In this work, we perform a reinterpretation of the latest ATLAS search for HNLs decaying promptly to a tri-lepton final state. We show that in a realistic model with two HNLs, the actual limits can vary by several orders of magnitude depending on the free parameters of the model. Marginalizing over the unknown model parameters leads to an exclusion limit on the total mixing angle which can be up to 3 orders of magnitude weaker than the limits reported in arXiv:1905.09787. This demonstrates that the reinterpretation of results from experimental searches is a necessary step to obtain meaningful limits on realistic models. We detail a few steps that can be taken by experimental collaborations in order to simplify the reuse of their results., Comment: 40 pages, 12 figures, 7 tables, matches the published version

Heavy neutral leptons (HNLs) appear in many extensions of the Standard Model of particle physics. In this study, we investigate to which extent the NA62 experiment at CERN could improve the existing bounds on the HNL mixing angle $|U_e|^2$ by performing a missing mass search in $K^+ \to \pi^0 e^+ N$ decays in flight. We show that the limit $|U_e|^2 \simeq 2\times 10^{-6}$ can be reached with the currently available data in the mass range 125 -- 144 MeV, which is currently not well covered by production searches. Future data, together with a dedicated trigger and/or improvements in rejection of out-of-acceptance photons, can improve this limit by another order of magnitude., Comment: 7 pages, 4 figures, matches published version

SHiP is a proposed high-intensity beam dump experiment set to operate at the CERN SPS. It is expected to have an unprecedented sensitivity to a variety of models containing feebly interacting particles, such as Heavy Neutral Leptons (HNLs). Two HNLs or more could successfully explain the observed neutrino masses through the seesaw mechanism. If, in addition, they are quasi-degenerate, they could be responsible for the baryon asymmetry of the Universe. Depending on their mass splitting, HNLs can have very different phenomenologies: they can behave as Majorana fermions -- with lepton number violating (LNV) signatures, such as same-sign dilepton decays -- or as Dirac fermions with only lepton number conserving (LNC) signatures. In this work, we quantitatively demonstrate that LNV processes can be distinguished from LNC ones at SHiP, using only the angular distribution of the HNL decay products. Accounting for spin correlations in the simulation and using boosted decision trees for discrimination, we show that SHiP will be able to distinguish Majorana-like and Dirac-like HNLs in a significant fraction of the currently unconstrained parameter space. If the mass splitting is of order $10^{-6}$ eV, SHiP could even be capable of resolving HNL oscillations, thus providing a direct measurement of the mass splitting. This analysis highlights the potential of SHiP to not only search for feebly interacting particles, but also perform model selection., Comment: 37 pages, 13 figures