Image de couverture de livre

Beyond Reproducibility: Token Probabilities Expose Large Language Model Nondeterminism

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Fu, Tairan, Martínez, Gonzalo, Conde, Javier, Arriaga, Carlos, Reviriego, Pedro, Qi, Xiuyuan, Liu, Shanshan
Format:	Preprint
Publié:	2026
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Multiple Choice Questions: Reasoning Makes Large Language Models (LLMs) More Self-Confident, Especially When They are Wrong
par: Fu, Tairan, et autres
Publié: (2025)

Is There a Case for Conversation Optimized Tokenizers in Large Language Models?
par: Ferrando, Raquel, et autres
Publié: (2025)

The Generative Energy Arena (GEA): Incorporating Energy Awareness in Large Language Model (LLM) Human Evaluations
par: Arriaga, Carlos, et autres
Publié: (2025)

Lost in Sampling: Assessing Lexical Reachability in LLMs via the Word Coverage Score (WCS)
par: Awad, Samer, et autres
Publié: (2026)

Speed and Conversational Large Language Models: Not All Is About Tokens per Second
par: Conde, Javier, et autres
Publié: (2025)

Why Do Large Language Models (LLMs) Struggle to Count Letters?
par: Fu, Tairan, et autres
Publié: (2024)

Can ChatGPT Learn to Count Letters?
par: Conde, Javier, et autres
Publié: (2025)

Stochastic Streets: A Walk Through Random LLM Address Generation in four European Cities
par: Fu, Tairan, et autres
Publié: (2025)

Concurrent Linguistic Error Detection (CLED): a New Methodology for Error Detection in Large Language Models
par: Zhu, Jinhua, et autres
Publié: (2024)

To Words and Beyond: Probing Large Language Models for Sentence-Level Psycholinguistic Norms of Memorability and Reading Times
par: Clark, Thomas Hikaru, et autres
Publié: (2026)

Understanding the Impact of Artificial Intelligence in Academic Writing: Metadata to the Rescue
par: Conde, Javier, et autres
Publié: (2025)

Playing with words: Comparing the vocabulary and lexical diversity of ChatGPT and humans
par: Reviriego, Pedro, et autres
Publié: (2023)

Psycholinguistic Word Features: a New Approach for the Evaluation of LLMs Alignment with Humans
par: Conde, Javier, et autres
Publié: (2025)

How does fine-tuning improve sensorimotor representations in large language models?
par: Wu, Minghua, et autres
Publié: (2026)

Energy-Efficient Stochastic Computing (SC) Neural Networks for Internet of Things Devices With Layer-Wise Adjustable Sequence Length (ASL)
par: Wang, Ziheng, et autres
Publié: (2025)

Have Multimodal Large Language Models (MLLMs) Really Learned to Tell the Time on Analog Clocks?
par: Fu, Tairan, et autres
Publié: (2025)

Large Language Models and Book Summarization: Reading or Remembering, Which Is Better?
par: Fu, Tairan, et autres
Publié: (2026)

Real-time Spatial Retrieval Augmented Generation for Urban Environments
par: Campo, David Nazareno, et autres
Publié: (2025)

Lost in the Vibrations: Vision Language Models Fail the Dynamic Gauges Test
par: Fu, Tairan, et autres
Publié: (2026)

DiFR: Inference Verification Despite Nondeterminism
par: Karvonen, Adam, et autres
Publié: (2025)

Backdoor Token Unlearning: Exposing and Defending Backdoors in Pretrained Language Models
par: Jiang, Peihai, et autres
Publié: (2025)

Optimistic Verifiable Training by Controlling Hardware Nondeterminism
par: Srivastava, Megha, et autres
Publié: (2024)

Spanish and LLM Benchmarks: is MMLU Lost in Translation?
par: Plaza, Irene, et autres
Publié: (2024)

Evaluating SAT and SMT Solvers on Large-Scale Sudoku Puzzles
par: Davis, Liam, et autres
Publié: (2025)

Assessing Latency in ASR Systems: A Methodological Perspective for Real-Time Use
par: Arriaga, Carlos, et autres
Publié: (2024)

Detecting Hallucinations in Large Language Model Generation: A Token Probability Approach
par: Quevedo, Ernesto, et autres
Publié: (2024)

Adding LLMs to the psycholinguistic norming toolbox: A practical guide to getting the most out of human ratings
par: Conde, Javier, et autres
Publié: (2025)

Distributed Specialization: Rare-Token Neurons in Large Language Models
par: Liu, Jing, et autres
Publié: (2025)

BrainBench: Exposing the Commonsense Reasoning Gap in Large Language Models
par: Tang, Yuzhe
Publié: (2026)

Calibrating Verbalized Probabilities for Large Language Models
par: Wang, Cheng, et autres
Publié: (2024)

Incoherent Probability Judgments in Large Language Models
par: Zhu, Jian-Qiao, et autres
Publié: (2024)

Beyond Next Token Prediction: Patch-Level Training for Large Language Models
par: Shao, Chenze, et autres
Publié: (2024)

Reproducibility Study of "XRec: Large Language Models for Explainable Recommendation"
par: Mishra, Ranjan, et autres
Publié: (2025)

Token Signature: Predicting Chain-of-Thought Gains with Token Decoding Feature in Large Language Models
par: Liu, Peijie, et autres
Publié: (2025)

Reproducing and Extending Experiments in Behavioral Strategy with Large Language Models
par: Albert, Daniel, et autres
Publié: (2024)

Tokenization Matters! Degrading Large Language Models through Challenging Their Tokenization
par: Wang, Dixuan, et autres
Publié: (2024)

Entropy-KL Divergence-based Token Masking: A Novel Approach for Selective Fine-tuning of Large Language Models
par: Liu, Qi, et autres
Publié: (2026)

Transient Turn Injection: Exposing Stateless Multi-Turn Vulnerabilities in Large Language Models
par: Rayhan, Naheed, et autres
Publié: (2026)

Re-Initialization Token Learning for Tool-Augmented Large Language Models
par: Li, Chenghao, et autres
Publié: (2025)

findsylls: A Language-Agnostic Toolkit for Syllable-Level Speech Tokenization and Embedding
par: Martínez, Héctor Javier Vázquez
Publié: (2026)