Cover Image

Beyond the Final Layer: Intermediate Representations for Better Multilingual Calibration in Large Language Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhou, Ej, Zhang, Caiqi, Hu, Tiancheng, Li, Chengzu, Collier, Nigel, Vulić, Ivan, Korhonen, Anna
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

TopViewRS: Vision-Language Models as Top-View Spatial Reasoners
by: Li, Chengzu, et al.
Published: (2024)

Quantifying Language Disparities in Multilingual Large Language Models
by: Hu, Songbo, et al.
Published: (2025)

Large Language Models are Miscalibrated In-Context Learners
by: Li, Chengzu, et al.
Published: (2023)

LoVeC: Reinforcement Learning for Better Verbalized Confidence in Long-Form Generations
by: Zhang, Caiqi, et al.
Published: (2025)

Analyzing and Adapting Large Language Models for Few-Shot Multilingual NLU: Are We There Yet?
by: Razumovskaia, Evgeniia, et al.
Published: (2024)

Fairer Preferences Elicit Improved Human-Aligned Large Language Model Judgments
by: Zhou, Han, et al.
Published: (2024)

Visual Planning: Let's Think Only with Images
by: Xu, Yi, et al.
Published: (2025)

Value of Information: A Framework for Human-Agent Communication
by: Dong, Yijiang River, et al.
Published: (2026)

Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluators
by: Liu, Yinhong, et al.
Published: (2024)

DIALIGHT: Lightweight Multilingual Development and Evaluation of Task-Oriented Dialogue Systems with Large Language Models
by: Hu, Songbo, et al.
Published: (2024)

Dial HEALTHDIAL for Advice: A Multilingual and Multi-Parallel Spoken Dialogue Dataset for Knowledge-Grounded Information Seeking
by: Hu, Songbo, et al.
Published: (2026)

Lost in Embeddings: Information Loss in Vision-Language Models
by: Li, Wenyan, et al.
Published: (2025)

Polyglot Teachers: Evaluating Language Models for Multilingual Synthetic Data Generation
by: Miranda, Lester James V., et al.
Published: (2026)

On Bilingual Lexicon Induction with Large Language Models
by: Li, Yaoyiran, et al.
Published: (2023)

Improving Word Translation via Two-Stage Contrastive Learning
by: Li, Yaoyiran, et al.
Published: (2022)

Confidence Estimation for LLMs in Multi-turn Interactions
by: Zhang, Caiqi, et al.
Published: (2026)

All Roads Lead to Rome: Graph-Based Confidence Estimation for Large Language Model Reasoning
by: Zhang, Caiqi, et al.
Published: (2025)

Navigating the Alignment-Calibration Trade-off: A Pareto-Superior Frontier via Model Merging
by: Hu, Tiancheng, et al.
Published: (2025)

Conformity in Large Language Models
by: Zhu, Xiaochen, et al.
Published: (2024)

iNews: A Multimodal Dataset for Modeling Personalized Affective Responses to News
by: Hu, Tiancheng, et al.
Published: (2025)

Quantifying the Persona Effect in LLM Simulations
by: Hu, Tiancheng, et al.
Published: (2024)

Decoupling the Effect of Chain-of-Thought Reasoning: A Human Label Variation Perspective
by: Chen, Beiduo, et al.
Published: (2026)

Scaling Sparse Fine-Tuning to Large Language Models
by: Ansell, Alan, et al.
Published: (2024)

Aligning with Logic: Measuring, Evaluating and Improving Logical Preference Consistency in Large Language Models
by: Liu, Yinhong, et al.
Published: (2024)

LUQ: Long-text Uncertainty Quantification for LLMs
by: Zhang, Caiqi, et al.
Published: (2024)

Self-Augmented In-Context Learning for Unsupervised Word Translation
by: Li, Yaoyiran, et al.
Published: (2024)

Atomic Calibration of LLMs in Long-Form Generations
by: Zhang, Caiqi, et al.
Published: (2024)

Translation-Enhanced Multilingual Text-to-Image Generation
by: Li, Yaoyiran, et al.
Published: (2023)

Agentic Policy Optimization via Instruction-Policy Co-Evolution
by: Zhou, Han, et al.
Published: (2025)

AutoPEFT: Automatic Configuration Search for Parameter-Efficient Fine-Tuning
by: Zhou, Han, et al.
Published: (2023)

Multi-Trigger Poisoning Amplifies Backdoor Vulnerabilities in LLMs
by: Sivapiromrat, Sanhanat, et al.
Published: (2025)

SQATIN: Supervised Instruction Tuning Meets Question Answering for Improved Dialogue NLU
by: Razumovskaia, Evgeniia, et al.
Published: (2023)

Can LLM be a Personalized Judge?
by: Dong, Yijiang River, et al.
Published: (2024)

Bias Beyond English: Evaluating Social Bias and Debiasing Methods in a Low-Resource Setting
by: Zhou, Ej, et al.
Published: (2025)

Steer Model beyond Assistant: Controlling System Prompt Strength via Contrastive Decoding
by: Dong, Yijiang River, et al.
Published: (2026)

SimBench: Benchmarking the Ability of Large Language Models to Simulate Human Behaviors
by: Hu, Tiancheng, et al.
Published: (2025)

Improving Bilingual Lexicon Induction with Cross-Encoder Reranking
by: Li, Yaoyiran, et al.
Published: (2022)

Retrofitting Large Language Models with Dynamic Tokenization
by: Feher, Darius, et al.
Published: (2024)

Calibrating Beyond English: Language Diversity for Better Quantized Multilingual LLM
by: Chimoto, Everlyn Asiko, et al.
Published: (2026)

Emergent Communication Pretraining for Few-Shot Machine Translation
by: Li, Yaoyiran, et al.
Published: (2020)