Image de couverture de livre

Mixing Mechanisms: How Language Models Retrieve Bound Entities In-Context

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Gur-Arieh, Yoav, Geva, Mor, Geiger, Atticus
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

Enhancing Automated Interpretability with Output-Centric Feature Descriptions
par: Gur-Arieh, Yoav, et autres
Publié: (2025)

Faithfulness Metrics Don't Measure Faithfulness: A Meta-Evaluation with Ground Truth
par: Gur-Arieh, Yoav, et autres
Publié: (2026)

Disentangling MLP Neuron Weights in Vocabulary Space
par: Avrahamy, Asaf, et autres
Publié: (2026)

Precise In-Parameter Concept Erasure in Large Language Models
par: Gur-Arieh, Yoav, et autres
Publié: (2025)

Constructing Interpretable Features from Compositional Neuron Groups
par: Shafran, Or, et autres
Publié: (2025)

LMEnt: A Suite for Analyzing Knowledge in Language Models from Pretraining Data to Representations
par: Gottesman, Daniela, et autres
Publié: (2025)

RAVEL: Evaluating Interpretability Methods on Disentangling Language Model Representations
par: Huang, Jing, et autres
Publié: (2024)

From Directions to Regions: Decomposing Activations in Language Models via Local Geometry
par: Shafran, Or, et autres
Publié: (2026)

Performance Gap in Entity Knowledge Extraction Across Modalities in Vision Language Models
par: Cohen, Ido, et autres
Publié: (2024)

Friends and Grandmothers in Silico: Localizing Entity Cells in Language Models
par: Yona, Itay, et autres
Publié: (2026)

Estimating Knowledge in Large Language Models Without Generating a Single Token
par: Gottesman, Daniela, et autres
Publié: (2024)

Can Large Language Models Faithfully Express Their Intrinsic Uncertainty in Words?
par: Yona, Gal, et autres
Publié: (2024)

ScoNe: Benchmarking Negation Reasoning in Language Models With Fine-Tuning and In-Context Learning
par: She, Jingyuan Selena, et autres
Publié: (2023)

Detecting (Un)answerability in Large Language Models with Linear Directions
par: Lavi, Maor Juliet, et autres
Publié: (2025)

Inferring Functionality of Attention Heads from their Parameters
par: Elhelo, Amit, et autres
Publié: (2024)

How Causal Abstraction Underpins Computational Explanation
par: Geiger, Atticus, et autres
Publié: (2025)

How Do Transformers Learn Variable Binding in Symbolic Programs?
par: Wu, Yiwei, et autres
Publié: (2025)

Indications of Belief-Guided Agency and Meta-Cognitive Monitoring in Large Language Models
par: Yalon, Noam Steinmetz, et autres
Publié: (2026)

From Loops to Oops: Fallback Behaviors of Language Models Under Uncertainty
par: Ivgi, Maor, et autres
Publié: (2024)

Do Large Language Models Latently Perform Multi-Hop Reasoning?
par: Yang, Sohee, et autres
Publié: (2024)

Eliciting Textual Descriptions from Representations of Continuous Prompts
par: Ramati, Dana, et autres
Publié: (2024)

Narrowing the Knowledge Evaluation Gap: Open-Domain Question Answering with Multi-Granularity Answers
par: Yona, Gal, et autres
Publié: (2024)

Hallucinations Undermine Trust; Metacognition is a Way Forward
par: Yona, Gal, et autres
Publié: (2026)

Interpretability at Scale: Identifying Causal Mechanisms in Alpaca
par: Wu, Zhengxuan, et autres
Publié: (2023)

Backward Lens: Projecting Language Model Gradients into the Vocabulary Space
par: Katz, Shahar, et autres
Publié: (2024)

Hopping Too Late: Exploring the Limitations of Large Language Models on Multi-Hop Queries
par: Biran, Eden, et autres
Publié: (2024)

Do Large Language Models Perform Latent Multi-Hop Reasoning without Exploiting Shortcuts?
par: Yang, Sohee, et autres
Publié: (2024)

How Well Can Reasoning Models Identify and Recover from Unhelpful Thoughts?
par: Yang, Sohee, et autres
Publié: (2025)

Preventing Rogue Agents Improves Multi-Agent Collaboration
par: Barbi, Ohav, et autres
Publié: (2025)

Routers Learn the Geometry of Their Experts: Geometric Coupling in Sparse Mixture-of-Experts
par: Ahrac, Sagi, et autres
Publié: (2026)

Thinking to Recall: How Reasoning Unlocks Parametric Knowledge in LLMs
par: Gekhman, Zorik, et autres
Publié: (2026)

The Hidden Space of Transformer Language Adapters
par: Alabi, Jesujoba O., et autres
Publié: (2024)

NER Retriever: Zero-Shot Named Entity Retrieval with Type-Aware Embeddings
par: Shachar, Or, et autres
Publié: (2025)

Patchscopes: A Unifying Framework for Inspecting Hidden Representations of Language Models
par: Ghandeharioun, Asma, et autres
Publié: (2024)

How do Language Models Bind Entities in Context?
par: Feng, Jiahai, et autres
Publié: (2023)

Rethinking Selective Knowledge Distillation
par: Tavor, Almog, et autres
Publié: (2026)

Jump to Conclusions: Short-Cutting Transformers With Linear Transformations
par: Din, Alexander Yom, et autres
Publié: (2023)

Language Models use Lookbacks to Track Beliefs
par: Prakash, Nikhil, et autres
Publié: (2025)

Are language models aware of the road not taken? Token-level uncertainty and hidden state dynamics
par: Zur, Amir, et autres
Publié: (2025)

MoNaCo: More Natural and Complex Questions for Reasoning Across Dozens of Documents
par: Wolfson, Tomer, et autres
Publié: (2025)