Cover Image

Sensivity of LLMs' Explanations to the Training Randomness:Context, Class & Task Dependencies

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Loncour, Romain, Bogaert, Jérémie, Standaert, François-Xavier
Format:	Preprint
Published:	2026
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

A Question on the Explainability of Large Language Models and the Word-Level Univariate First-Order Plausibility Assumption
by: Bogaert, Jeremie, et al.
Published: (2024)

Explanation sensitivity to the randomness of large language models: the case of journalistic text classification
by: Bogaert, Jeremie, et al.
Published: (2024)

Guess or Recall? Training CNNs to Classify and Localize Memorization in LLMs
by: Dentan, Jérémie, et al.
Published: (2025)

Facilitating Cognitive Accessibility with LLMs: A Multi-Task Approach to Easy-to-Read Text Generation
by: Ledoyen, François, et al.
Published: (2025)

Assessing the Reliability of LLMs Annotations in the Context of Demographic Bias and Model Explanation
by: Mohammadi, Hadi, et al.
Published: (2025)

How does Multi-Task Training Affect Transformer In-Context Capabilities? Investigations with Function Classes
by: Bhasin, Harmon, et al.
Published: (2024)

Attention Consistency for LLMs Explanation
by: Lan, Tian, et al.
Published: (2025)

Neither Stochastic Parroting nor AGI: LLMs Solve Tasks through Context-Directed Extrapolation from Training Data Priors
by: Madabushi, Harish Tayyar, et al.
Published: (2025)

Stochastic Chameleons: Irrelevant Context Hallucinations Reveal Class-Based (Mis)Generalization in LLMs
by: Cheng, Ziling, et al.
Published: (2025)

Are Long-LLMs A Necessity For Long-Context Tasks?
by: Qian, Hongjin, et al.
Published: (2024)

Explanation Generation for Contradiction Reconciliation with LLMs
by: Chan, Jason, et al.
Published: (2026)

Class Distillation with Mahalanobis Contrast: An Efficient Training Paradigm for Pragmatic Language Understanding Tasks
by: Wang, Chenlu, et al.
Published: (2025)

Embedding-based In-Context Prompt Training for Enhancing LLMs as Text Encoders
by: Lin, Ailiang, et al.
Published: (2026)

Training LLMs for EHR-Based Reasoning Tasks via Reinforcement Learning
by: Lin, Jiacheng, et al.
Published: (2025)

Towards Faithful Explanations for Text Classification with Robustness Improvement and Explanation Guided Training
by: Li, Dongfang, et al.
Published: (2023)

In-Context Learning and Fine-Tuning GPT for Argument Mining
by: Cabessa, Jérémie, et al.
Published: (2024)

DRIV-EX: Counterfactual Explanations for Driving LLMs
by: Cardiel, Amaia, et al.
Published: (2026)

Argument-Based Consistency in Toxicity Explanations of LLMs
by: Mothilal, Ramaravind Kommiya, et al.
Published: (2025)

Investigating the Pre-Training Dynamics of In-Context Learning: Task Recognition vs. Task Learning
by: Wang, Xiaolei, et al.
Published: (2024)

DynamicKV: Task-Aware Adaptive KV Cache Compression for Long Context LLMs
by: Zhou, Xiabin, et al.
Published: (2024)

Context-Aware Machine Translation with Source Coreference Explanation
by: Vu, Huy Hien, et al.
Published: (2024)

LADM: Long-context Training Data Selection with Attention-based Dependency Measurement for LLMs
by: Chen, Jianghao, et al.
Published: (2025)

Let's (not) just put things in Context: Test-Time Training for Long-Context LLMs
by: Bansal, Rachit, et al.
Published: (2025)

Musketeer: Joint Training for Multi-task Vision Language Model with Task Explanation Prompts
by: Zhang, Zhaoyang, et al.
Published: (2023)

Training-Inference Consistent Segmented Execution for Long-Context LLMs
by: Shang, Xianpeng, et al.
Published: (2026)

Local Explanations and Self-Explanations for Assessing Faithfulness in black-box LLMs
by: Fragkathoulas, Christos, et al.
Published: (2024)

LCIRC: A Recurrent Compression Approach for Efficient Long-form Context and Query Dependent Modeling in LLMs
by: An, Sumin, et al.
Published: (2025)

GRKV: Global Regression for Training-Free KV Cache Compression in Long-Context LLMs
by: Peng, Junjie, et al.
Published: (2026)

Evaluation Framework for Highlight Explanations of Context Utilisation in Language Models
by: Sun, Jingyi, et al.
Published: (2025)

Simple Mechanistic Explanations for Out-Of-Context Reasoning
by: Wang, Atticus, et al.
Published: (2025)

ATACompressor: Adaptive Task-Aware Compression for Efficient Long-Context Processing in LLMs
by: Li, Xuancheng, et al.
Published: (2026)

Model Tells You Where to Merge: Adaptive KV Cache Merging for LLMs on Long-Context Tasks
by: Wang, Zheng, et al.
Published: (2024)

Counterfactual Simulatability of LLM Explanations for Generation Tasks
by: Limpijankit, Marvin, et al.
Published: (2025)

Generating Medically-Informed Explanations for Depression Detection using LLMs
by: Chen, Xiangyong, et al.
Published: (2025)

Culinary Class Wars: Evaluating LLMs using ASH in Cuisine Transfer Task
by: Lee, Hoonick, et al.
Published: (2024)

Explanation based In-Context Demonstrations Retrieval for Multilingual Grammatical Error Correction
by: Li, Wei, et al.
Published: (2025)

Retrieval-Augmented Hierarchical in-Context Reinforcement Learning and Hindsight Modular Reflections for Task Planning with LLMs
by: Sun, Chuanneng, et al.
Published: (2024)

Unveiling Decision-Making in LLMs for Text Classification : Extraction of influential and interpretable concepts with Sparse Autoencoders
by: Bail, Mathis Le, et al.
Published: (2025)

InfLLM: Training-Free Long-Context Extrapolation for LLMs with an Efficient Context Memory
by: Xiao, Chaojun, et al.
Published: (2024)

Out of the Memory Barrier: A Highly Memory Efficient Training System for LLMs with Million-Token Contexts
by: Li, Wenhao, et al.
Published: (2026)