arXiv subjects

Daniel King

Publications and source records attributed to Daniel King.

At least 19 recordsLinked to original sources

arXiv2026

Spectroscopy With Intelligent Feature Tracking for automated characterization of excited-state structure in semiconductor quantum dot devices

The characterization of excited-state structure in semiconductor quantum dot (QD) devices is an important component of tuning them for spin-qubit operation. We present Spectroscopy With Intelligent Feature Tracking (SWIFT), a framework that combines machine-learning (ML)-assisted feature identification with physics-informed geometric processing to extract energy-level splittings from pulsed-gate spectroscopy data. SWIFT isolates the relevant spectral features and exploits their characteristic geometry to reduce the two-dimensional spectroscopy analysis to a one-dimensional peak-detection problem. It further combines an ensemble-based confidence metric with sequential accumulation of rapid, low-SNR scans, allowing the inferred spectrum to be reevaluated as experimental evidence accumulates. Using Si/SiGe QD devices, we demonstrate SWIFT both offline and in real time, including automated tracking of QD excited states and lead resonances. Benchmarking on 255 manually labeled scans shows that SWIFT reduces the median splitting error to 0.05 mV from 0.11 mV for a classical baseline, with the largest improvement on lower-quality measurements. These results provide a path toward incorporating excited-state spectroscopy into autonomous QD characterization, tuning, and optimization, which will be essential in large-scale quantum dot devices.

cond-mat.mes-hall↗

arXiv2026

Proceedings of the 2nd International Workshop on Low Carbon Computing (LOCO 2026)

This volume contains the proceedings of the 2nd International Workshop on Low Carbon Computing (LOCO 2026), held at Lancaster University, United Kingdom, on 10-11 September 2026. LOCO provides an interdisciplinary forum for research, practical tools, early-stage work, radical ideas, and critical perspectives addressing the reduction of greenhouse gas emissions associated with computing. The proceedings cover topics including carbon measurement and reporting, sustainable software engineering, energy-efficient computing, carbon-aware systems, hardware longevity, embodied carbon, circular computing, resource management, frugal and sufficiency-oriented computing, sustainable artificial intelligence, scientific computing, and the wider environmental effects of digital technologies. Full workshop papers were evaluated through a non-blind peer-review process by members of the LOCO 2026 Programme Committee. Submissions were assessed for originality, relevance to the workshop, quality, and potential to stimulate constructive discussion. The volume also includes accepted lightning-talk abstracts where authors elected to include them.

cs.CY↗

arXiv2024

LoRA Learns Less and Forgets Less

Low-Rank Adaptation (LoRA) is a widely-used parameter-efficient finetuning method for large language models. LoRA saves memory by training only low rank perturbations to selected weight matrices. In this work, we compare the performance of LoRA and full finetuning on two target domains, programming and mathematics. We consider both the instruction finetuning (approximately 100K prompt-response pairs) and continued pretraining (20B unstructured tokens) data regimes. Our results show that, in the standard low-rank settings, LoRA substantially underperforms full finetuning. Nevertheless, LoRA better maintains the base model's performance on tasks outside the target domain. We show that LoRA mitigates forgetting more than common regularization techniques such as weight decay and dropout; it also helps maintain more diverse generations. Finally, we show that full finetuning learns perturbations with a rank that is 10-100X greater than typical LoRA configurations, possibly explaining some of the reported gaps. We conclude by proposing best practices for finetuning with LoRA.

cs.LG↗

arXiv2024

Automatic State Interaction with Large Localized Active Spaces for Multimetallic Systems

The localized active space self consistent field (LASSCF) method factorizes a complete active space (CAS) wave function into an antisymmetrized product of localized active space wave function fragments. Correlation between fragments is then reintroduced through LAS state interaction (LASSI), in which the Hamiltonian is diagonalized in a model space of LAS states. However, the optimal procedure for defining the LAS fragments and LASSI model space is unknown. We here present an automated framework to explore systematically convergent sets of model spaces, which we call LASSI[$r$,$q$]. This method requires the user to select only $r$, the number of electron hops from one fragment to another and $q$, the number of fragment basis functions per Hilbert space, which converges to CASCI in the limit of $r,q\to\infty$. Numerical tests of this method on the tri-metal complexes [Fe(III)Al(III)Fe(II)($\mu_3$-O)]$^{6+}$ and [Fe(III)$_2$Fe(II)($\mu_3$-O)]$^{6+}$ show efficient convergence to the CASCI limit with 4-10 orders of magnitude fewer states.

physics.chem-ph↗

arXiv2023

MosaicBERT: A Bidirectional Encoder Optimized for Fast Pretraining

Although BERT-style encoder models are heavily used in NLP research, many researchers do not pretrain their own BERTs from scratch due to the high cost of training. In the past half-decade since BERT first rose to prominence, many advances have been made with other transformer architectures and training configurations that have yet to be systematically incorporated into BERT. Here, we introduce MosaicBERT, a BERT-style encoder architecture and training recipe that is empirically optimized for fast pretraining. This efficient architecture incorporates FlashAttention, Attention with Linear Biases (ALiBi), Gated Linear Units (GLU), a module to dynamically remove padded tokens, and low precision LayerNorm into the classic transformer encoder block. The training recipe includes a 30% masking ratio for the Masked Language Modeling (MLM) objective, bfloat16 precision, and vocabulary size optimized for GPU throughput, in addition to best-practices from RoBERTa and other encoder models. When pretrained from scratch on the C4 dataset, this base model achieves a downstream average GLUE (dev) score of 79.6 in 1.13 hours on 8 A100 80 GB GPUs at a cost of roughly $20. We plot extensive accuracy vs. pretraining speed Pareto curves and show that MosaicBERT base and large are consistently Pareto optimal when compared to a competitive BERT base and large. This empirical speed up in pretraining enables researchers and engineers to pretrain custom BERT-style models at low cost instead of finetune on existing generic models. We open source our model weights and code.

cs.CL↗

arXiv2023

The Semantic Scholar Open Data Platform

The volume of scientific output is creating an urgent need for automated tools to help scientists keep up with developments in their field. Semantic Scholar (S2) is an open data platform and website aimed at accelerating science by helping scholars discover and understand scientific literature. We combine public and proprietary data sources using state-of-the-art techniques for scholarly PDF content extraction and automatic knowledge graph construction to build the Semantic Scholar Academic Graph, the largest open scientific literature graph to-date, with 200M+ papers, 80M+ authors, 550M+ paper-authorship edges, and 2.4B+ citation edges. The graph includes advanced semantic features such as structurally parsed text, natural language summaries, and vector embeddings. In this paper, we describe the components of the S2 data processing pipeline and the associated APIs offered by the platform. We will update this living document to reflect changes as we add new data offerings and improve existing services.

cs.DL↗

arXiv2022

ACCoRD: A Multi-Document Approach to Generating Diverse Descriptions of Scientific Concepts

Systems that can automatically define unfamiliar terms hold the promise of improving the accessibility of scientific texts, especially for readers who may lack prerequisite background knowledge. However, current systems assume a single "best" description per concept, which fails to account for the many potentially useful ways a concept can be described. We present ACCoRD, an end-to-end system tackling the novel task of generating sets of descriptions of scientific concepts. Our system takes advantage of the myriad ways a concept is mentioned across the scientific literature to produce distinct, diverse descriptions of target scientific concepts in terms of different reference concepts. To support research on the task, we release an expert-annotated resource, the ACCoRD corpus, which includes 1,275 labeled contexts and 1,787 hand-authored concept descriptions. We conduct a user study demonstrating that (1) users prefer descriptions produced by our end-to-end system, and (2) users prefer multiple descriptions to a single "best" description.

cs.CL↗

arXiv2022

S2AMP: A High-Coverage Dataset of Scholarly Mentorship Inferred from Publications

Mentorship is a critical component of academia, but is not as visible as publications, citations, grants, and awards. Despite the importance of studying the quality and impact of mentorship, there are few large representative mentorship datasets available. We contribute two datasets to the study of mentorship. The first has over 300,000 ground truth academic mentor-mentee pairs obtained from multiple diverse, manually-curated sources, and linked to the Semantic Scholar (S2) knowledge graph. We use this dataset to train an accurate classifier for predicting mentorship relations from bibliographic features, achieving a held-out area under the ROC curve of 0.96. Our second dataset is formed by applying the classifier to the complete co-authorship graph of S2. The result is an inferred graph with 137 million weighted mentorship edges among 24 million nodes. We release this first-of-its-kind dataset to the community to help accelerate the study of scholarly mentorship: \url{https://github.com/allenai/S2AMP-data}

cs.DL↗

arXiv2022

Don't Say What You Don't Know: Improving the Consistency of Abstractive Summarization by Constraining Beam Search

Abstractive summarization systems today produce fluent and relevant output, but often "hallucinate" statements not supported by the source text. We analyze the connection between hallucinations and training data, and find evidence that models hallucinate because they train on target summaries that are unsupported by the source. Based on our findings, we present PINOCCHIO, a new decoding method that improves the consistency of a transformer-based abstractive summarizer by constraining beam search to avoid hallucinations. Given the model states and outputs at a given step, PINOCCHIO detects likely model hallucinations based on various measures of attribution to the source text. PINOCCHIO backtracks to find more consistent output, and can opt to produce no summary at all when no consistent generation can be found. In experiments, we find that PINOCCHIO improves the consistency of generation (in terms of F1) by an average of~67% on two abstractive summarization datasets.

cs.CL↗

arXiv2021

$SU(3)$ breaking effects in $B$ and $D$ meson lifetimes

In the heavy quark expansion (HQE) of the total decay rates of $B_s$ and $D_s^+$ mesons non-perturbative matrix elements of four quark operators are arising as phase space enhanced contributions. We present the first determination of $m_s$ effects to the dimension six matrix elements of these four quark operators via a heavy quark effective theory (HQET) sum rule analysis. In addition we calculate for the first time eye contractions of the four quark operators as well as matrix elements of penguin operators. For the perturbative part we solve the 3-loop contribution to the sum rule and we evaluate condensate contributions. In this study we work in the strict HQET limit and our results can also be used to estimate the size of the matrix element of the Darwin operator via equations of motion.

hep-ph↗

arXiv2021

Revisiting Inclusive Decay Widths of Charmed Mesons

Determining for the first time the Darwin operator contribution for the non-leptonic charm-quark decays and using new non-perturbative results for the matrix elements of $\Delta C=0$ four-quark operators, including eye-contractions, we present a comprehensive study of the lifetimes of charmed mesons and inclusive semileptonic decay rates as well as the ratios, within the framework of the Heavy Quark Expansion (HQE). We find good agreement with experiment for the ratio $\tau(D^+)/\tau(D^0)$, for the total $D_s^+$-meson decay rate, for the semileptonic rates of all three mesons $D^0$, $D^+$ and $D_s^+$, and for the semileptonic ratio $\Gamma_{sl}^{D^+}/\Gamma_{sl}^{D^0}$. The total decay rates of the $D^0$ and $D^+$ mesons are underestimated in our HQE approach and we suspect that this is due to missing higher-order QCD corrections to the free charm quark decay and the Pauli interference contribution. For the $SU(3)_F$ breaking ratios $\tau (D_s^+) / \tau (D^0) $ and $\Gamma_{sl}^{D_s^+}/\Gamma_{sl}^{D^0} $ our predictions lie closer to one than experiment. This might originate from the poor knowledge of the non-perturbative parameters $\mu_G^2$, $\mu_\pi^2$ and $\rho_D^3$ in the $D^0$ and $D_s^+$ systems. These parameters could be determined by experimental studies of the moments of inclusive semileptonic $D$ meson decays.

hep-ph↗

arXiv2021

Reducing Annotating Load: Active Learning with Synthetic Images in Surgical Instrument Segmentation

Accurate instrument segmentation in endoscopic vision of robot-assisted surgery is challenging due to reflection on the instruments and frequent contacts with tissue. Deep neural networks (DNN) show competitive performance and are in favor in recent years. However, the hunger of DNN for labeled data poses a huge workload of annotation. Motivated by alleviating this workload, we propose a general embeddable method to decrease the usage of labeled real images, using active generated synthetic images. In each active learning iteration, the most informative unlabeled images are first queried by active learning and then labeled. Next, synthetic images are generated based on these selected images. The instruments and backgrounds are cropped out and randomly combined with each other with blending and fusion near the boundary. The effectiveness of the proposed method is validated on 2 sinus surgery datasets and 1 intraabdominal surgery dataset. The results indicate a considerable improvement in performance, especially when the budget for annotation is small. The effectiveness of different types of synthetic images, blending methods, and external background are also studied. All the code is open-sourced at: https://github.com/HaonanPeng/active_syn_generator.

cs.CV↗

arXiv2021

S2AND: A Benchmark and Evaluation System for Author Name Disambiguation

Author Name Disambiguation (AND) is the task of resolving which author mentions in a bibliographic database refer to the same real-world person, and is a critical ingredient of digital library applications such as search and citation analysis. While many AND algorithms have been proposed, comparing them is difficult because they often employ distinct features and are evaluated on different datasets. In response to this challenge, we present S2AND, a unified benchmark dataset for AND on scholarly papers, as well as an open-source reference model implementation. Our dataset harmonizes eight disparate AND datasets into a uniform format, with a single rich feature set drawn from the Semantic Scholar (S2) database. Our evaluation suite for S2AND reports performance split by facets like publication year and number of papers, allowing researchers to track both global performance and measures of fairness across facet values. Our experiments show that because previous datasets tend to cover idiosyncratic and biased slices of the literature, algorithms trained to perform well on one on them may generalize poorly to others. By contrast, we show how training on a union of datasets in S2AND results in more robust models that perform well even on datasets unseen in training. The resulting AND model also substantially improves over the production algorithm in S2, reducing error by over 50% in terms of $B^3$ F1. We release our unified dataset, model code, trained models, and evaluation suite to the research community. https://github.com/allenai/S2AND/

cs.DL↗

arXiv2020

High-Precision Extraction of Emerging Concepts from Scientific Literature

Identification of new concepts in scientific literature can help power faceted search, scientific trend analysis, knowledge-base construction, and more, but current methods are lacking. Manual identification cannot keep up with the torrent of new publications, while the precision of existing automatic techniques is too low for many applications. We present an unsupervised concept extraction method for scientific literature that achieves much higher precision than previous work. Our approach relies on a simple but novel intuition: each scientific concept is likely to be introduced or popularized by a single paper that is disproportionately cited by subsequent papers mentioning the concept. From a corpus of computer science papers on arXiv, we find that our method achieves a Precision@1000 of 99%, compared to 86% for prior work, and a substantially better precision-yield trade-off across the top 15,000 extractions. To stimulate research in this area, we release our code and data (https://github.com/allenai/ForeCite).

cs.IR↗

arXiv2019

$|V_{cb}|$ and $\gamma$ from $B$-mixing -- Addendum to "$B_s$ mixing observables and $|V_{td}/V_{ts}|$ from sum rules"

In this addendum to "$B_s$ mixing observables and $|V_{td}/V_{ts}|$ from sum rules" \cite{King:2019lal} we study the impact of the recent improvements in the theoretical precision of $B$ meson mixing onto CKM unitarity fits. Our key results are the most precise determination of the angle $\gamma = \left(63.4\pm0.9\right)^\circ$ in the unitarity triangle and a new value for the CKM element $|V_{cb}|=(41.6\pm0.7)\cdot10^{-3}$.

hep-ph↗

arXiv2019

Pretrained Language Models for Sequential Sentence Classification

As a step toward better document-level understanding, we explore classification of a sequence of sentences into their corresponding categories, a task that requires understanding sentences in context of the document. Recent successful models for this task have used hierarchical models to contextualize sentence representations, and Conditional Random Fields (CRFs) to incorporate dependencies between subsequent labels. In this work, we show that pretrained language models, BERT (Devlin et al., 2018) in particular, can be used for this task to capture contextual dependencies without the need for hierarchical encoding nor a CRF. Specifically, we construct a joint sentence representation that allows BERT Transformer layers to directly utilize contextual information from all words in all sentences. Our approach achieves state-of-the-art results on four datasets, including a new dataset of structured scientific abstracts.

cs.CL↗

arXiv2019

Strong Baselines for Complex Word Identification across Multiple Languages

Complex Word Identification (CWI) is the task of identifying which words or phrases in a sentence are difficult to understand by a target audience. The latest CWI Shared Task released data for two settings: monolingual (i.e. train and test in the same language) and cross-lingual (i.e. test in a language not seen during training). The best monolingual models relied on language-dependent features, which do not generalise in the cross-lingual setting, while the best cross-lingual model used neural networks with multi-task learning. In this paper, we present monolingual and cross-lingual CWI models that perform as well as (or better than) most models submitted to the latest CWI Shared Task. We show that carefully selected features and simple learning models can achieve state-of-the-art performance, and result in strong baselines for future development in this area. Finally, we discuss how inconsistencies in the annotation of the data can explain some of the results obtained.

cs.CL↗

arXiv2019

Bs mixing observables and Vtd/Vts from sum rules

We consider the effects of a non-vanishing strange-quark mass in the determination of the full basis of dimension six matrix elements for $B_{s}$ mixing, in particular we get for the ratio of the $V-A$ Bag parameter in the $B_s$ and $B_d$ system: $\overline{B}^s_{Q_1} / \overline{B}^d_{Q_1} = 0.987^{+0.007}_{-0.009}$. Combining these results with the most recent lattice values for the ratio of decay constants $f_{B_s} / f_{B_d}$ we obtain the most precise determination of the ratio $\xi = f_{B_s} \sqrt{\overline{B}^s_{Q_1}}/ f_{B_d} \sqrt{\overline{B}^d_{Q_1}} = 1.2014^{+0.0065}_{-0.0072}$ in agreement with recent lattice determinations. We find $\Delta M_s=(18.5_{-1.5}^{+1.2})\text{ps}^{-1}$ and $\Delta M_d=(0.547_{-0.046}^{+0.035})\text{ps}^{-1}$ to be consistent with experiments at below one sigma. Assuming the validity of the SM, our calculation can be used to directly determine the ratio of CKM elements $|V_{td} / V_{ts} | = 0.2045^{+0.0012}_{-0.0013}$, which is compatible with the results from the CKM fitting groups, but again more precise.

hep-ph↗