Copertina

Min-$k$ Sampling: Decoupling Truncation from Temperature Scaling via Relative Logit Dynamics

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Ding, Yuanhao, Li, Meimingwei, Arias, Esteban Garces, Aßenmacher, Matthias, Heumann, Christian, Zhang, Chongsheng
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Artificial Intelligence Computation and Language Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Beyond Temperature: Hyperfitting as a Late-Stage Geometric Expansion
di: Li, Meimingwei, et al.
Pubblicazione: (2026)

Decoding Decoded: Understanding Hyperparameter Effects in Open-Ended Text Generation
di: Arias, Esteban Garces, et al.
Pubblicazione: (2024)

GUARD: Glocal Uncertainty-Aware Robust Decoding for Effective and Efficient Open-Ended Text Generation
di: Ding, Yuanhao, et al.
Pubblicazione: (2025)

Adaptive Contrastive Search: Uncertainty-Guided Decoding for Open-Ended Text Generation
di: Arias, Esteban Garces, et al.
Pubblicazione: (2024)

The Truncation Blind Spot: How Decoding Strategies Systematically Exclude Human-Like Token Choices
di: Arias, Esteban Garces, et al.
Pubblicazione: (2026)

Towards Better Open-Ended Text Generation: A Multicriteria Evaluation Framework
di: Arias, Esteban Garces, et al.
Pubblicazione: (2024)

Unveiling Factors for Enhanced POS Tagging: A Study of Low-Resource Medieval Romance Languages
di: Schöffel, Matthias, et al.
Pubblicazione: (2025)

Lost in Translation? Exploring the Shift in Grammatical Gender from Latin to Occitan
di: Chatterjee, Ahan, et al.
Pubblicazione: (2026)

How Prevalent is Gender Bias in ChatGPT? -- Exploring German and English ChatGPT Responses
di: Urchs, Stefanie, et al.
Pubblicazione: (2023)

Modern Models, Medieval Texts: A POS Tagging Study of Old Occitan
di: Schöffel, Matthias, et al.
Pubblicazione: (2025)

From Traditional Taggers to LLMs: A Comparative Study of POS Tagging for Medieval Romance Languages
di: Schöffel, Matthias, et al.
Pubblicazione: (2026)

Self-Reinforcing Controllable Synthesis of Rare Relational Data via Bayesian Calibration
di: Zhang, Chongsheng, et al.
Pubblicazione: (2026)

Explainable Coarse-to-Fine Ancient Manuscript Duplicates Discovery
di: Zhang, Chongsheng, et al.
Pubblicazione: (2025)

Can OpenSource beat ChatGPT? -- A Comparative Study of Large Language Models for Text-to-Code Generation
di: Mayer, Luis, et al.
Pubblicazione: (2024)

The Geometry of Creative Variability: How Credal Sets Expose Calibration Gaps in Language Models
di: Arias, Esteban Garces, et al.
Pubblicazione: (2025)

Are All Genders Equal in the Eyes of Algorithms? -- Analysing Search and Retrieval Algorithms for Algorithmic Gender Fairness
di: Urchs, Stefanie, et al.
Pubblicazione: (2025)

From Calculation to Adjudication: Examining LLM judges on Mathematical Reasoning Tasks
di: Stephan, Andreas, et al.
Pubblicazione: (2024)

Scaling Textual Gradients via Sampling-Based Momentum
di: Ding, Zixin, et al.
Pubblicazione: (2025)

On the Role of Temperature Sampling in Test-Time Scaling
di: Wu, Yuheng, et al.
Pubblicazione: (2025)

On Giant's Shoulders: Effortless Weak to Strong by Dynamic Logits Fusion
di: Fan, Chenghao, et al.
Pubblicazione: (2024)

Distilling the Essence: Efficient Reasoning Distillation via Sequence Truncation
di: Chen, Wei-Rui, et al.
Pubblicazione: (2025)

taz2024full: Analysing German Newspapers for Gender Bias and Discrimination across Decades
di: Urchs, Stefanie, et al.
Pubblicazione: (2025)

Anti-Length Shift: Dynamic Outlier Truncation for Training Efficient Reasoning Models
di: Wu, Wei, et al.
Pubblicazione: (2026)

DynScaling: Efficient Verifier-free Inference Scaling via Dynamic and Integrated Sampling
di: Wang, Fei, et al.
Pubblicazione: (2025)

Fewer Truncations Improve Language Modeling
di: Ding, Hantian, et al.
Pubblicazione: (2024)

No Free Lunch in Active Learning: LLM Embedding Quality Dictates Query Strategy Success
di: Rauch, Lukas, et al.
Pubblicazione: (2025)

Scaled and Inter-token Relation Enhanced Transformer for Sample-restricted Residential NILM
di: Rahman, Minhajur, et al.
Pubblicazione: (2024)

Statistical Multicriteria Evaluation of LLM-Generated Text
di: Arias, Esteban Garces, et al.
Pubblicazione: (2025)

Free Lunch for Pass@$k$? Low Cost Diverse Sampling for Diffusion Language Models
di: Lamont, Sean, et al.
Pubblicazione: (2026)

Decoupling Understanding from Reasoning via Problem Space Mapping for Small-Scale Model Reasoning
di: Wang, Li, et al.
Pubblicazione: (2025)

A Decoupling and Aggregating Framework for Joint Extraction of Entities and Relations
di: Wang, Yao, et al.
Pubblicazione: (2024)

Spectral Logit Sculpting: Adaptive Low-Rank Logit Transformation for Controlled Text Generation
di: Li, Jin, et al.
Pubblicazione: (2025)

ROME: Memorization Insights from Text, Logits and Representation
di: Li, Bo, et al.
Pubblicazione: (2024)

Fair Play in the Newsroom: Actor-Based Filtering Gender Discrimination in Text Corpora
di: Urchs, Stefanie, et al.
Pubblicazione: (2025)

Enhancing Low-Resource Relation Representations through Multi-View Decoupling
di: Fan, Chenghao, et al.
Pubblicazione: (2023)

Steering Language Models Before They Speak: Logit-Level Interventions
di: An, Hyeseon, et al.
Pubblicazione: (2026)

Logit Arithmetic Elicits Long Reasoning Capabilities Without Training
di: Zhang, Yunxiang, et al.
Pubblicazione: (2025)

Private Language Models via Truncated Laplacian Mechanism
di: Huang, Tianhao, et al.
Pubblicazione: (2024)

LongAgent: Scaling Language Models to 128k Context through Multi-Agent Collaboration
di: Zhao, Jun, et al.
Pubblicazione: (2024)

Sparse Logit Sampling: Accelerating Knowledge Distillation in LLMs
di: Anshumann, et al.
Pubblicazione: (2025)