Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 389 for “"gemma"”.
-
A preliminary study of the polyembryonic cutworm parasite berecyntus Bakeri var. gemma Girault
As early as 1923, the attention of Mr. K. M. King Was drawn to the occurrence in SasKatchewan of a hymenopterous insect parasite of the family Encyrtidae, Berecyntus bak.eri How. which was parasitizing noctuid larvae. Because of the great number of adults developing in a single host larva, it was …
-
Ανάπτυξη αναλυτικών μεθόδων φασματομετρίας κινητικότητας ιόντων για τον προσδιορισμό μεγέθους και σύστασης νανοσωματιδίων
… In several of the relevant publications the term GEMMA (gas-phase electrophoretic mobility molecular analyzer) is used instead of IMS. GEMMA has been successfully used to determine the size and Relative Molecular Mass of macromolecules of biological origin and a plurality of inorganic engineered …
-
Feverdreams: Dream Affect in Women who Have Experienced Sexual Trauma
… we compared the ability of two emerging LLMs, GEMMA and C-RoBERTa, to build construct validity. Our results suggested that trauma may influence the overall tone of dreams instead of nightmares alone, and using positive and negative indices of emotion may be useful in evaluating dream emotion. …
-
Evolution of life history strategies in Lophoziaceae
… (sexuality) and between diaspore (spore or gemma) frequency and sexuality. Data from the field were used to establish diaspore (spore and gemma) sizes and their production per capsule or shoot and to test for relationships between diaspore size and production per shoot/capsule and also …
-
Occupational gender bias in large language models : a multi-level analysis using the OccuBias Dataset
… LLMs, Meta’s LLaMA-3-8B-Instruct and Google’s Gemma-7B-IT, using a purpose-built evaluation dataset named OccuBias, comprising 6,300 prompts across gendered, gender-neutral, and stereotype-challenging occupational contexts. Bias is analyzed at both the output level and the embedding level using …
-
Structure of a Subtidal Benthic Community Near Cape Charles, Virginia
… energy shoal assemblage numerically dominated by Gemma gemma and haustoriid amphipods, and (3) an offshore assemblage composed of Spiophanes bombyx and other euryhaline marine species characteristic of inner continental shelf sand communities in temperate areas of the eastern United States.
-
Multivariate genome-wide association studies and genomic predictions in multiple breeds and crossbred animals
… animals in the second research project. Using GEMMA, we fit traits in a Bayesian Sparse Linear Mixed Models (BSLMM). The SNPs with the largest effects are found by calculating allele substitution effect from the SNP based BSLMM analysis. Using the four flanking SNPs of each identified core SNP …
-
The ecology of fishes on submerged pinnacle coral reefs
Gemma Galbraith studied the ecology of reef fishes on submerged pinnacle coral reefs. She found that these unique but understudied coral reef habitats supported distinct, diverse and highly productive fish communities. Gemma's research demonstrates that small submerged patch reefs are valuable …
-
MULTILINGUAL EVENT EXTRACTION USING GENERATIVE LARGE LANGUAGE MODELS
… from four different families (Phi-3, Qwen 2.5, Gemma 2, and Mistral) across three size tiers. The study employs a generative extraction approach using a single prompt across all models to extract four key components from news articles describing protest events: actors, targets, location, and …
-
Cross-layer SAE feature drift for early detection and mitigation of multi-turn jailbreaks
… can restore safe behavior upon detection. Using Gemma-3-4B-IT with Gemma Scope SAEs across four transformer layers, a cross-layer MLP classifier is trained on per-turn aggregated SAE activations to detect escalation before harmful output appears. The detector successfully identified jailbreak …
-
Ancient Greek to Modern Greek Machine Translation: A novel benchmark and fine-tuning experiments on LLMs and NMT models
… και Μεγάλα Γλωσσικά Μοντέλα (Llama-Krikri-8B, Gemma 3). Τα πειράματά μας δείχνουν ότι η πλήρης βελτιστοποίηση (full fine-tuning) του M2M100-1.2B επιτυγχάνει την υψηλότερη απόδοση σε επίπεδο πρότασης (67,15 BLEU). Μεταξύ των LLMs, το πολυγλωσσικό μοντέλο Gemma-3-4B-it με χρήση LoRA επιτυγχάνει …
-
Advanced Retrieval-Augmented Generation: Optimization and Evaluation
… κειμένου. Αξιολογήθηκαν δύο μοντέλα retrieval (Gemma-300m έναντι Jina-v3) και δύο μοντέλα generation (Mistral-7B έναντι Llama-3-8B). Παράλληλα, εξετάστηκαν τεχνικές βελτιστοποίησης, όπως είναι η σημασιολογική τμηματοποίηση (semantic chunking), η επανακατάταξη μέσω cross-encoder και τα …
-
Synthesising Summaries: A novel Retrieval-Augmented Generation-based pipeline for multi-document summarisation
… language models (LLMs), specifically Google's Gemma 2B and 7B models, to the task of scientific literature summarisation. A key innovative approach is the integration of Retrieval-Augmented Generation (RAG) techniques to enhance the summarisation process. The study employs a mixed-methods …
-
Waste Relations in Young Adult Literature
… Games Trilogy, Aldous Huxley’s Brave New World, Gemma Malley’s The Declaration, and Catherine Sinclair’s Uncle David’s Nonsensical Story about Giants and Fairies. It discusses the consequences that evolve from contemporary notions of waste relations, the ways we define waste, and what boundaries …
-
Využití velkých předtrénovaných jazykových modelů pro konfiguraci a podporu klinického informačního systému
… Výsledky ukazují, že metoda RAG s modelem Gemma a embedding modelem bge-m3 poskytuje nejrelevantnější odpovědi, ale má potíže s porozuměním složitějších otázek. Metoda dotrénování modelu nepřináší očekávané výsledky, a to ani po úpravách parametrů trénování.
-
LLMs in Network Intrusion Detection – A Comprehensive Analysis
… as well as large-scale generative models like Gemma (2B, 9B, and 27B parameter versions) for intrusion detection tasks. Our study begins by implementing standard machine learning models - k-Nearest Neighbors (kNN), Random Forest, XGBoost, Support Vector Machines (SVM), and Logistic Regression …
-
Detekce falešných recenzí generovaných AI na online platformách
… s recenziami generovanými jazykovým modelom Gemma 3, ktorý je natrénovaný na pôvodných recenziách. Potom je implementovaný a natrénovaný klasifikátor falošných recenzií. Navrhnutý klasifikátor je stacked ensemble model ktorý využíva štýlometriu, TF-IDF a BERT-like model. Model dosahuje …
Page 1 of 20