Image de couverture de livre

Context Copying Modulation: The Role of Entropy Neurons in Managing Parametric and Contextual Knowledge Conflicts

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Tighidet, Zineddine, Mogini, Andrea, Ben-younes, Hedi, Mei, Jiali, Gallinari, Patrick, Piwowarski, Benjamin
Format:	Preprint
Publié:	2025
Sujets:	Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Probing Language Models on Their Knowledge Source
par: Tighidet, Zineddine, et autres
Publié: (2024)

Classification EM-PCA for clustering and embedding
par: Tighidet, Zineddine, et autres
Publié: (2025)

MEXMA: Token-level objectives improve sentence representations
par: Janeiro, João Maria, et autres
Publié: (2024)

AdaCAD: Adaptively Decoding to Balance Conflicts between Contextual and Parametric Knowledge
par: Wang, Han, et autres
Publié: (2024)

Contextual Categorization Enhancement through LLMs Latent-Space
par: Bettouche, Zineddine, et autres
Publié: (2024)

Mitigating Copy Bias in In-Context Learning through Neuron Pruning
par: Ali, Ameen, et autres
Publié: (2024)

Towards Lossless Token Pruning in Late-Interaction Retrieval Models
par: Zong, Yuxuan, et autres
Publié: (2025)

CopyNE: Better Contextual ASR by Copying Named Entities
par: Zhou, Shilin, et autres
Publié: (2023)

IRCAN: Mitigating Knowledge Conflicts in LLM Generation via Identifying and Reweighting Context-Aware Neurons
par: Shi, Dan, et autres
Publié: (2024)

Survey on Question Answering over Visually Rich Documents: Methods, Challenges, and Trends
par: Barboule, Camille, et autres
Publié: (2025)

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding
par: Jin, Mingyu, et autres
Publié: (2025)

Do Large Language Models Know Conflict? Investigating Parametric vs. Non-Parametric Knowledge of LLMs for Conflict Forecasting
par: Nemkova, Apollinaire Poli, et autres
Publié: (2025)

Structural Deep Encoding for Table Question Answering
par: Mouravieff, Raphaël, et autres
Publié: (2025)

On the Representations of Entities in Auto-regressive Large Language Models
par: Morand, Victor, et autres
Publié: (2025)

RoleConflictBench: A Benchmark of Role Conflict Scenarios for Evaluating LLMs' Contextual Sensitivity
par: Shin, Jisu, et autres
Publié: (2025)

Understanding the Interplay between Parametric and Contextual Knowledge for Large Language Models
par: Cheng, Sitao, et autres
Publié: (2024)

In-Context Learning Without Copying
par: Sahin, Kerem, et autres
Publié: (2025)

Resisting Contextual Interference in RAG via Parametric-Knowledge Reinforcement
par: Lin, Chenyu, et autres
Publié: (2025)

ToMMeR -- Efficient Entity Mention Detection from Large Language Models
par: Morand, Victor, et autres
Publié: (2025)

Understanding QA generation: Extracting Parametric and Contextual Knowledge with CQA for Low Resource Bangla Language
par: Azmary, Umme Abira, et autres
Publié: (2026)

Give it Space! Explicit Disentangling of Positional and Semantic Representations in Encoders
par: Lequeu, Pierre-Antoine, et autres
Publié: (2026)

Three Regimes of Context-Parametric Conflict: A Predictive Framework and Empirical Validation
par: Venkata, Pruthvinath Jeripity
Publié: (2026)

Understanding the Interplay between LLMs' Utilisation of Parametric and Contextual Knowledge: A keynote at ECIR 2025
par: Augenstein, Isabelle
Publié: (2026)

To Copy or Not to Copy: Copying Is Easier to Induce Than Recall
par: Farahani, Mehrdad, et autres
Publié: (2026)

Updating Parametric Knowledge with Context Distillation Retains Post-Training Capabilities
par: Padmanabhan, Shankar, et autres
Publié: (2026)

Resona: Improving Context Copying in Linear Recurrence Models with Retrieval
par: Wang, Xinyu, et autres
Publié: (2025)

Rational Retrieval Acts: Leveraging Pragmatic Reasoning to Improve Sparse Retrieval
par: Satouf, Arthur, et autres
Publié: (2025)

Context Parametrization with Compositional Adapters
par: Jukić, Josip, et autres
Publié: (2025)

Zebra: In-Context Generative Pretraining for Solving Parametric PDEs
par: Serrano, Louis, et autres
Publié: (2024)

How Training Data Shapes the Use of Parametric and In-Context Knowledge in Language Models
par: Kim, Minsung, et autres
Publié: (2025)

Blinded by Generated Contexts: How Language Models Merge Generated and Retrieved Contexts When Knowledge Conflicts?
par: Tan, Hexiang, et autres
Publié: (2024)

Task Matters: Knowledge Requirements Shape LLM Responses to Context-Memory Conflict
par: Sun, Kaiser, et autres
Publié: (2025)

From Tokens to Concepts: Leveraging SAE for SPLADE
par: Zong, Yuxuan, et autres
Publié: (2026)

A Systematic Evaluation of Large Language Models for PTSD Severity Estimation: The Role of Contextual Knowledge and Modeling Strategies
par: Kaliosis, Panagiotis, et autres
Publié: (2026)

Structured Knowledge Representation through Contextual Pages for Retrieval-Augmented Generation
par: Li, Xinze, et autres
Publié: (2026)

Context Retrieval via Normalized Contextual Latent Interaction for Conversational Agent
par: Liu, Junfeng, et autres
Publié: (2023)

CoRect: Context-Aware Logit Contrast for Hidden State Rectification to Resolve Knowledge Conflicts
par: Ma, Xuhua, et autres
Publié: (2026)

Does RAG Know When Retrieval Is Wrong? Diagnosing Context Compliance under Knowledge Conflict
par: Chen, Yihang, et autres
Publié: (2026)

What does the Knowledge Neuron Thesis Have to do with Knowledge?
par: Niu, Jingcheng, et autres
Publié: (2024)

CoCoA: Confidence and Context-Aware Adaptive Decoding for Resolving Knowledge Conflicts in Large Language Models
par: Khandelwal, Anant, et autres
Publié: (2025)