Channels - SupraTok: Cross-Boundary Tokenization for Enhanced Language Model Performance :: FRELIP Discovery

Similar Items: SupraTok: Cross-Boundary Tokenization for Enhanced Language Model Performance

Quick Look
SupraTok: Cross-Boundary Tokenization for Enhanced Language Model Performance
Quick Look
Re-evaluating the Word Token for Bilingual Speech Processing: The Case for Intonation Units
Quick Look
Cross-layer Attention Sharing for Pre-trained Large Language Models
Quick Look
Can Language Models Learn Typologically Implausible Languages?
Quick Look
Enhancing spoken language identification on Indian languages with orthogonal task and related task integration: unveiling new dimensions of accuracy
Quick Look
Anthropocentric Bias in Language Model Evaluation
Quick Look
Emotion-Related Language Choice theory in the cross-fire: Evidence from Mexican-American bilinguals
Quick Look
Accelerating Language Model Workflows with Prompt Choreography
Quick Look
Are Formal and Functional Linguistic Mechanisms Dissociated in Language Models?
Quick Look
Human-like Fleeting Memory Improves Language Learning but Impairs Reading Time Prediction in Transformer Language Models
Quick Look
Anthropomimetic Uncertainty: What Verbalized Uncertainty in Language Models is Missing
Quick Look
Data Foundations of Long-Context Language Models: A Survey
Quick Look
A Systematic Assessment of Language Models with Linguistic Minimal Pairs in Chinese
Quick Look
A Survey on Memory-Efficient Fine-Tuning for Large Language Models
Quick Look
Goodnight Gorilla: How Do Second Language Learners’ American Sign Language Narrative Renditions Change after Viewing an ASL Model?
Quick Look
Can Large Language Models Generalize Analogy Solving Like Children Can?
Quick Look
BP-LLM : Belief Propagation for Binary Feedback in Large Language Model Alignment
Quick Look
Fine-tuning Large Language Models with Limited Data: A Survey and Practical Guide
Quick Look
LMEnt: A Suite for Analyzing Knowledge in Language Models from Pretraining Data to Representations
Quick Look
Can a Large Language Model Replace Humans at Rating Lexical Semantic Relations Strength?
Quick Look
From Explicit to Implicit: A Theoretical Framework and Transfer Method for Preference Internalization in Language Models
Quick Look
Language Models Can Resolve Reference Compositionally, But It’s Not Their Native Strength: The Case of the Personal Relation Task
Quick Look
On the Limitations of Language-targeted Pruning: Investigating the Calibration Language Impact in Multilingual LLM Pruning
Quick Look
S HARING B EYOND D ECISION : Deep Collaboration between Large Language Models via Representation Ensemble