The Archive

Theoretically explains why embedding norms in scale-invariant contrastive models encode semantic specificity despite being ignored by cosine similarity metrics.

Single-stage geo-localization framework with 220k+ ground-satellite image pairs for cross-view object matching.

Linguistic Firewall improves multi-agent routing by using geometric profiling to verify agent competence empirically rather than self-description.

Clover tool tracks student engagement with AI code completion via behavioral metrics and attention checks to measure critical evaluation.

Non-parametric causal diffusion recovery from cross-sectional steady-state data; application to gene expression without requiring longitudinal observation.

Curvature-Weighted Gradient Diversity (CWGD) refines SGD convergence analysis by weighting gradient noise by inverse Hessian to better model effective optimization dynamics.

EMPATH: multilingual multi-turn safety benchmark for emotional-support chatbots across 140 scenarios, 34 personas, and 19 metrics on crisis handling.

LLM evaluation and AI safety face a shared measurement problem: benchmark scores, reward-model signals, and reported safety metrics can improve while the latent properties they are meant to represent remain difficult to verify. This paper combines a hybrid survey - a systematic…

SHOVIR benchmark evaluates whether radiology report generation models rely on image evidence or learned priors via spatially-grounded metrics.

Nuclear physics paper using interpretable neural networks to model binding energies via SU(3)/SU(4) symmetries; domain-specific, not AI-frontier work.

Dimensionality reduction via autoencoders and PCA for compressing wearable runner telemetry into scalar performance scores.

Mechanism-driven internal monitors detect LLM training instability before loss/gradient metrics surface failures, reducing costly compute waste.

BiDeMem uses bidirectional degradation memory with explainable queries for interpretable image restoration beyond PSNR metrics.

Study shows LLM self-evaluation is not uniformly easier than generation; multi-hop QA reveals attention-based mechanistic differences.

Lifted Causal Inference applies relational lifted inference to causal graphical models via parametric causal factor graphs.

Shaders are GPU programs that process visual data—such as rays, pixels, geometry, and textures—to produce specific rendering effects. Shaders find necessary... Shaders are GPU programs that process visual data—such as rays, pixels, geometry, and textures—to produce specific…

Language-based digital twin framework using LLMs detects Mild Cognitive Impairment via stylometric analysis and conversation patterns.

Study identifies co-failure ceiling limiting accuracy gains in multi-model LLM routing/voting systems across 67 frontier models; proposes beta metric beyond pairwise correlation.

Training-free source selection for LLMs using Fisher alignment metrics at vocabulary scale for scientific domains (SMILES, proteins).

fTNN tensor neural network for fractional PDEs using geometry-adapted integration split on bounded domains.

Synthetic data augmentation is widely used to mitigate class imbalance, but its theoretical effects on score-based classification remain poorly understood. This paper develops a framework for characterizing when synthetic minority augmentation can improve threshold-integrated…

Midway through an ordinary pretraining run, a small language model learns the pronoun-gender rule: cued with a girl's name ("Sue cried because"), it resolves the next pronoun to she, generalizing to held-out probes (0.94 by step 925). By step 3,500 the same model scores near…

AI translation of literary works is increasingly common. While the content may be rendered adequately, we do not know enough about how readers experience it in terms of immersiveness and literary effect, aspects poorly captured by automatic machine translation metrics or human…

Training-time data poisoning during fine-tuning poses a significant threat to large language models (LLMs) deployed for abstractive text summarization, where small task-specific datasets exert disproportionate influence on model behavior. In this setting, adversaries manipulate…

Text detoxification, the automated detection and mitigation of abusive and harmful content, is essential for ensuring the safety of online communities and protecting users. However, low resource languages such as Tatar have received little research attention. In this paper we…

Variational Monte Carlo (VMC) is a central algorithm in electronic structure theory and has gained renewed importance through modern neural-network ansätze such as FermiNet. At its core, VMC seeks ground states by minimizing the Rayleigh quotient by stochastic optimization. In…

As machine learning models and datasets continue to grow, developing complex models has become increasingly computationally demanding. Knowledge distillation reduces deployment cost by compressing a large, well-trained teacher model into a compact student model, but it does not…

Existing texture datasets for tactile sensing primarily consist of sensor readings from a specific sensor interacting with available surfaces/objects rather than describing the textures themselves, limiting fair comparison between tactile sensors and hindering reproducible…

DGPs are probabilistic models with remarkable prediction performance that concatenate GPs across several layers. Exact inference in DGPs is intractable, and variational inference is often used to approximate the posterior with a parametric distribution tuned by minimizing the…

A new update for Google Home could make it less likely your smart home cameras mistake you for someone else, just because you're facing away from the camera. Starting June 23rd, Google's expanding its facial recognition feature so that people you've tagged in your Familiar…