1095089 results (page 64 of 43804)
-
RDP LoRA: Geometry-Driven Identification for Parameter-Efficient Adaptation in Large Language Models
Fine-tuning Large Language Models (LLMs) remains structurally uncertain despite parameter-efficient methods such as Low-Rank Adaptation (LoRA), as the layer-specific roles of internal representations are poorly understood, leading to heuristic decisions about where adaptation should be applied. We model the evolution of hidden states as a high-dimensional geometric trajectory and propose using the…
-
An Oracle-Free Quantum Algorithm for Nonadiabatic Quantum Molecular Dynamics
Quantum computation is an attractive front for many problems that are intractable for computers today. One such problem is nonadiabatic quantum molecular dynamics, where quantized internal states coupling to parameterized modes result in a Hamiltonian resistant to oracle-based models and spectral decomposition. This dissertation applies diabatic Hamiltonian operators directly to the computational …
-
Multi-view Crowd Tracking Transformer with View-Ground Interactions Under Large Real-World Scenes
Multi-view crowd tracking estimates each person's tracking trajectories on the ground of the scene. Recent research works mainly rely on CNNs-based multi-view crowd tracking architectures, and most of them are evaluated and compared on relatively small datasets, such as Wildtrack and MultiviewX. Since these two datasets are collected in small scenes and only contain tens of frames in the evaluatio…
-
Merger rate of initially clustered primordial black holes for the two-body channel
Primordial black holes (PBHs) may form an initially clustered population depending on their production mechanism. Motivated by binary black-hole merger events observed by gravitational-wave interferometers, we revisit the evaluation of the merger rate of PBH binaries and extend the formalism to include the effects of clustering. We show that, in the presence of relatively weak PBH clustering, the …
-
Improving LLM-Driven Test Generation by Learning from Mocking Information
Large Language Models (LLMs) have recently shown strong potential for automated unit test generation. This has motivated us to investigate whether developer-defined test doubles (commonly referred to as mocks) available in existing test suites can be leveraged to improve LLM-driven test generation. To this end, we propose MOCKMILL, an LLM-based technique and tool that generates test cases by explo…
-
Framelet-Based Blind Image Restoration with Minimax Concave Regularization
Recovering corrupted images is one of the most challenging problems in image processing. Among various restoration tasks, blind image deblurring has been extensively studied due to its practical importance and inherent difficulty. In this problem, both the point spread function (PSF) and the underlying latent sharp image must be estimated simultaneously. This problem cannot be solved directly due …
-
On the Conditioning Consistency Gap in Conditional Neural Processes
Neural processes are meta-learning models that map context sets to predictive distributions. While inspired by stochastic processes, NPs do not generally satisfy the Kolmogorov consistency conditions required to define a valid stochastic process. This inconsistency is widely acknowledged but poorly understood. Practitioners note that NPs work well despite the violation, without quantifying what th…
-
Extrinsic geometry and Hamiltonian analysis of symmetric teleparallel gravity
We analyze the properties of foliations in presence of non-metricity, deriving the generalized Gauss-Codazzi relations in full generality. These results are employed to study the teleparallel framework of non-metric geometry, obtaining constraints on the extrinsic and intrinsic tensors. In particular, an extrinsic symmetric two-tensor plays the role of the extrinsic curvature in Riemannian geometr…
-
Co-Refine: AI-Powered Tool Supporting Qualitative Analysis
Qualitative coding relies on a researcher's application of codes to textual data. As coding proceeds across large datasets, interpretations of codes often shift (temporal drift), reducing the credibility of the analysis. Existing Computer-Assisted Qualitative Data Analysis (CAQDAS) tools provide support for data management but offer no workflow for real-time detection of these drifts. We present C…
-
Mapping-based Hard-constrained Physics-Informed Neural Networks for unbounded wave problems
The aim of this paper is to introduce a Mapping-based Hard-constrained Physics-Informed Neural Network (MH-PINN) for efficiently and accurately solving unbounded wave problems. First, we propose a coordinate mapping technique that compactifies the infinite physical domain into a finite computational space. This effectively resolves the sampling difficulties inherent to standard PINNs in unbounded …
-
DebugRepair: Enhancing LLM-Based Automated Program Repair via Self-Directed Debugging
Automated Program Repair (APR) has benefited from the code understanding and generation capabilities of Large Language Models (LLMs). Existing feedback-based APR methods iteratively refine candidate patches using test execution feedback and have shown promising results. However, most rely on outcome-level failure symptoms, such as stack traces, which show how failures are observed but fail to expo…
-
Large Language Models Exhibit Normative Conformity
The conformity bias exhibited by large language models (LLMs) can pose a significant challenge to decision-making in LLM-based multi-agent systems (LLM-MAS). While many prior studies have treated "conformity" simply as a matter of opinion change, this study introduces the social psychological distinction between informational conformity and normative conformity in order to understand LLM conformit…
-
HalluAudio: A Comprehensive Benchmark for Hallucination Detection in Large Audio-Language Models
Large Audio-Language Models (LALMs) have recently achieved strong performance across various audio-centric tasks. However, hallucination, where models generate responses that are semantically incorrect or acoustically unsupported, remains largely underexplored in the audio domain. Existing hallucination benchmarks mainly focus on text or vision, while the few audio-oriented studies are limited in …
-
Rethinking Scale: Deployment Trade-offs of Small Language Models under Agent Paradigms
Despite the impressive capabilities of large language models, their substantial computational costs, latency, and privacy risks hinder their widespread deployment in real-world applications. Small Language Models (SLMs) with fewer than 10 billion parameters present a promising alternative; however, their inherent limitations in knowledge and reasoning curtail their effectiveness. Existing research…
-
IndiaFinBench: An Evaluation Benchmark for Large Language Model Performance on Indian Financial Regulatory Text
We introduce IndiaFinBench, to our knowledge the first publicly available evaluation benchmark for assessing large language model (LLM) performance on Indian financial regulatory text. Existing financial NLP benchmarks draw exclusively from Western financial corpora (SEC filings, US earnings reports, and English-language financial news), leaving a significant gap in coverage of non-Western regulat…
-
Debiased neural operators for estimating functionals
Neural operators are widely used to approximate solution maps of complex physical systems. In many applications, however, the goal is not to recover the full solution trajectory, but to summarize the solution trajectory via a scalar target quantity (e.g., a functional such as time spent in a target range, time above a threshold, accumulated cost, or total energy). In this paper, we introduce DOPE …
-
TEMPO: Scaling Test-time Training for Large Reasoning Models
Test-time training (TTT) adapts model parameters on unlabeled test instances during inference time, which continuously extends capabilities beyond the reach of offline training. Despite initial gains, existing TTT methods for LRMs plateau quickly and do not benefit from additional test-time compute. Without external calibration, the self-generated reward signal increasingly drifts as the policy mo…
-
Location Not Found: Exposing Implicit Local and Global Biases in Multilingual LLMs
Multilingual large language models (LLMs) have minimized the fluency gap between languages. This advancement, however, exposes models to the risk of biased behavior, as knowledge and norms may propagate across languages. In this work, we aim to quantify models' inter- and intra-lingual biases, via their ability to answer locale-ambiguous questions. To this end, we present LocQA, a test set contain…
-
Community Detection with the Canonical Ensemble
Network community detection is usually considered as an unsupervised learning problem. Given a network, the aim is to partition it using some general purpose algorithm. In this paper we instead treat community detection as a hypothesis testing problem. Given a network, we examine the evidence for specific community structure in the observed network compared to a null model. To do this we define an…
-
Orthogonal reparametrization of the Nelson-Siegel-Svensson interest rate curve model: conditioning, diagnostics, and identifiability
The Nelson-Siegel-Svensson (NSS) interest rate curve model yields a separable nonlinear least-squares problem whose inner linear block is often ill-conditioned because the basis functions become nearly collinear. We analyze this instability via an exact orthogonal reparametrization of the design matrix. A thin QR decomposition produces orthogonal linear parameters for which, conditional on the non…
-
Mass Matrix Assembly on Tensor Cores for Implicit Particle-In-Cell Methods
Matrix-multiply-accumulate (MMA) units, or tensor cores, are now widespread across modern computing architectures. Yet, their use for particle-grid operators remains limited. In implicit particle methods, mass-matrix assembly is a reduction-dominated kernel in which weighted outer products of interpolation weights are accumulated over particle support. We show that this operation can be reformulat…
-
When Transparency Falls Short: Auditing Platform Moderation During a High-Stakes Election
During major political events, social media platforms encounter increased systemic risks. However, it is still unclear if and how they adjust their moderation practices in response. The Digital Services Act Transparency Database provides-for the first time-an opportunity to systematically examine content moderation at scale, allowing researchers and policymakers to evaluate platforms' compliance a…
-
Spatio-temporal modelling of electric vehicle charging demand
Accurate forecasting of electric vehicle (EV) charging demand is critical for grid management and infrastructure planning. Yet the field continues to rely on legacy benchmarks; such as the Palo Alto (2020) dataset; that fail to reflect the scale and behavioral diversity of modern charging networks. To address this, we introduce a novel large-scale longitudinal dataset collected across Scotland (20…
-
Beyond Semantic Similarity: A Component-Wise Evaluation Framework for Medical Question Answering Systems with Health Equity Implications
The use of Large Language Models (LLMs) to support patients in addressing medical questions is becoming increasingly prevalent. However, most of the measures currently used to evaluate the performance of these models in this context only measure how closely a model's answers match semantically, and therefore do not provide a true indication of the model's medical accuracy or of the health equity r…
-
A Lagrangian framework for canonical analysis for the Holst model with $β= 0$
We perform a canonical analysis of the Holst model for General Relativity, within the framework laid out in arXiv:2401.07307 and arXiv:2010.07725, distinguishing our approach by setting the Barbero parameter to $β=0$ and leaving the lapse and shift functions unconstrained. The $β= 0$ choice is of particular interest because it is viable across all dimensions, providing a necessary foundation for e…