5 sonuçlar
Arama Sonuçları
Listeleniyor 1 - 5 / 5
Yayın Geopolitical parallax: beyond Walter Lippmann just after large language models(Cornell Univ, 2025-08-27) Yavuz, Mehmet Can; Kabir, Humza Gohar; Özkan, AylinObjectivity in journalism has long been contested, oscillating between ideals of neutral, fact-based reporting and the inevitability of subjective framing. With the advent of large language models (LLMs), these tensions are now mediated by algorithmic systems whose training data and design choices may themselves embed cultural or ideological biases. This study investigates geopolitical parallax—systematic divergence in news quality and subjectivity assessments—by comparing articlelevel embeddings from Chinese-origin (Qwen, BGE, Jina) and Western-origin (Snowflake, Granite) model families. We evaluate both on a human-annotated news quality benchmark spanning fifteen stylistic, informational, and affective dimensions, and on parallel corpora covering politically sensitive topics, including Palestine and reciprocal China–United States coverage. Using logistic regression probes and matched-topic evaluation, we quantify per-metric differences in predicted positive-class probabilities between model families. Our findings reveal consistent, nonrandom divergences aligned with model origin. In Palestinerelated coverage, Western models assign higher subjectivity and positive emotion scores, while Chinese models emphasize novelty and descriptiveness. Cross-topic analysis shows asymmetries in structural quality metrics—Chinese-on-US scoring notably lower in fluency, conciseness, technicality, and overall quality—contrasted by higher negative emotion scores. These patterns align with media bias theory and our distinction between semantic, emotional, and relational subjectivity, and extend LLM bias literature by showing that geopolitical framing effects persist in downstream quality assessment tasks. We conclude that LLMbased media evaluation pipelines require cultural calibration to avoid conflating content differences with model-induced bias.Yayın Evaluating the efficiency of latent spaces via the coupling-matrix(Cornell Univ, 2025-09-08) Yavuz, Mehmet Can; Yanıkoğlu, BerrinA central challenge in representation learning is constructing latent embeddings that are both expressive and efficient. In practice, deep networks often produce redundant latent spaces where multiple coordinates encode overlapping information, reducing effective capacity and hindering generalization. Standard metrics such as accuracy or reconstruction loss provide only indirect evidence of such redundancy and cannot isolate it as a failure mode. We introduce a redundancy index, denoted ρ(C), that directly quantifies inter-dimensional dependencies by analyzing coupling matrices derived from latent representations and comparing their off-diagonal statistics against a normal distribution via energy distance. The result is a compact, interpretable, and statistically grounded measure of representational quality. We validate ρ(C) across discriminative and generative settings on MNIST variants, Fashion-MNIST, CIFAR-10, and CIFAR-100, spanning multiple architectures and hyperparameter optimization strategies. Empirically, low ρ(C) reliably predicts high classification accuracy or low reconstruction error, while elevated redundancy is associated with performance collapse. Estimator reliability grows with latent dimension, yielding natural lower bounds for reliable analysis. We further show that Treestructured Parzen Estimators (TPE) preferentially explore lowρ regions, suggesting that ρ(C) can guide neural architecture search and serve as a redundancy-aware regularization target. By exposing redundancy as a universal bottleneck across models and tasks, ρ(C) offers both a theoretical lens and a practical tool for evaluating and improving the efficiency of learned representations.Yayın Multivariate variational autoencoder(Cornell Univ, 2025-11-08) Yavuz, Mehmet CanLearning latent representations that are simultaneously expressive, geometrically well-structured, and reliably calibrated remains a central challenge for Variational Autoencoders (VAEs). Standard VAEs typically assume a diagonal Gaussian posterior, which simplifies optimization but rules out correlated uncertainty and often yields entangled or redundant latent dimensions. We introduce the Multivariate Variational Autoencoder (MVAE), a tractable full-covariance extension of the VAE that augments the encoder with sample-specific diagonal scales and a global coupling matrix. This induces a multivariate Gaussian posterior of the form N (µϕ(x), C diag(σ2ϕ(x))C⊤), enabling correlated latent factors while preserving a closedform KL divergence and a simple reparameterization path. Beyond likelihood, we propose a multi-criterion evaluation protocol that jointly assesses reconstruction quality (MSE, ELBO), downstream discrimination (linear probes), probabilistic calibration (NLL, Brier, ECE), and unsupervised structure (NMI, ARI). Across Larochelle-style MNIST variants, Fashion-MNIST, and CIFAR-10/100, MVAE consistently matches or outperforms diagonal-covariance VAEs of comparable capacity, with particularly notable gains in calibration and clustering metrics at both low and high latent dimensions. Qualitative analyses further show smoother, more semantically coherent latent traversals and sharper reconstructions. All code, dataset splits, and evaluation utilities are released to facilitate reproducible comparison and future extensions of multivariate posterior models.Yayın TurkEmbed4Retrieval: Türkçe için geri getirme görevine özel gömme modeli(Cornell Univ, 2025-11-10) Ezerceli, Özay; Gümüşçekiçci, Gizem; Erkoç, Tuğba; Özenç, BerkeBu çalışmada, öncelikle Doğal Dil Çıkarımı (DDÇ) ve Anlamsal Metin Benzerliği (AMB) görevleri için geliştirilen TurkEmbed modelinin, MS-Marco-TR veri seti üzerinde ince ayar yapılarak geri getirme görevlerine uygun hale getirilmesini sağlayan TurkEmbed4Retrieval modelini tanıtıyoruz. Model, Matruşka temsili ögrenme ve özel tasarlanmış negatif çiftlerin sıralanması kayıp fonksiyonu gibi ileri seviye egitim teknikleri kullanılarak optimize edilmiştir. Yapılan kapsamlı deneyler, TurkEmbed4Retrieval’ın, geri getirme metriklerinde TurkishcolBERT modelini Scifact-TR veri kümesinde %19–26 oranında geçtiğini göstermektedir. Bu bağlamda, modelimiz, Türkçe bilgi getirme sistemleri için yeni bir çıtaya ulaşmaktadır.Yayın TurkEmbed: Turkish embedding model on natural language inference & sentence text similarity tasks(Cornell Univ, 2025-11-10) Ezerceli, Özay; Gümüşçekiçci, Gizem; Erkoç, Tuğba; Özenç, BerkeThis paper introduces TurkEmbed, a novel Turkish language embedding model designed to outperform existing models, particularly in Natural Language Inference (NLI) and Semantic Textual Similarity (STS) tasks. Current Turkish embedding models often rely on machine-translated datasets, potentially limiting their accuracy and semantic understanding. TurkEmbed utilizes a combination of diverse datasets and advanced training techniques, including matryoshka representation learning, to achieve more robust and accurate embeddings. This approach enables the model to adapt to various resourceconstrained environments, offering faster encoding capabilities. Our evaluation on the Turkish STS-bTR dataset, using Pearson and Spearman correlation metrics, demonstrates significant improvements in semantic similarity tasks. Furthermore, TurkEmbed surpasses the current state-of-the-art model, Emrecan, on All-NLI-TR and STS-b-TR benchmarks, achieving a 1-4% improvement. TurkEmbed promises to enhance the Turkish NLP ecosystem by providing a more nuanced understanding of language and facilitating advancements in downstream applications.
