Copertina

Decomposing Prediction Mechanisms for In-Context Recall

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Daniels, Sultan, Davis, Dylan, Gautam, Dhruv, Liao, Wentinn, Ranade, Gireeja, Sahai, Anant
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

The Thermodynamic Costs of Simple Linear Regression
di: D'Ambrosia, Samuel H., et al.
Pubblicazione: (2026)

Polynomial Regression as a Task for Understanding In-context Learning Through Finetuning and Alignment
di: Wilcoxson, Max, et al.
Pubblicazione: (2024)

Provable Weak-to-Strong Generalization via Benign Overfitting
di: Wu, David X., et al.
Pubblicazione: (2024)

Precise Asymptotic Generalization for Multiclass Classification with Overparameterized Linear Models
di: Wu, David X., et al.
Pubblicazione: (2023)

LeanTutor: Towards a Verified AI Mathematical Proof Tutor
di: Patel, Manooshree, et al.
Pubblicazione: (2026)

Synthetic Error Injection Fails to Elicit Self-Correction In Language Models
di: Wu, David X., et al.
Pubblicazione: (2025)

Can Custom Models Learn In-Context? An Exploration of Hybrid Architecture Performance on In-Context Learning Tasks
di: Campbell, Ryan, et al.
Pubblicazione: (2024)

LLM In-Context Recall is Prompt Dependent
di: Machlab, Daniel, et al.
Pubblicazione: (2024)

Phase Transitions of Diversity in Stochastic Block Model Dynamics
di: Brânzei, Simina, et al.
Pubblicazione: (2023)

Fine-Tuning Dynamics of In-Context Factual Recall in Transformers
di: Huang, Ruomin, et al.
Pubblicazione: (2026)

DoMINO: A Decomposable Multi-scale Iterative Neural Operator for Modeling Large Scale Engineering Simulations
di: Ranade, Rishikesh, et al.
Pubblicazione: (2025)

Necessary and Sufficient Oracles: Toward a Computational Taxonomy For Reinforcement Learning
di: Rohatgi, Dhruv, et al.
Pubblicazione: (2025)

Constrained Contextual Bandits with Adversarial Contexts
di: Sarkar, Dhruv, et al.
Pubblicazione: (2026)

Decomposing multimodal embedding spaces with group-sparse autoencoders
di: Kaushik, Chiraag, et al.
Pubblicazione: (2026)

Stabilizing a linear system using phone calls when time is information
di: Khojasteh, Mohammad Javad, et al.
Pubblicazione: (2018)

A Multi-Component AI Framework for Computational Psychology: From Robust Predictive Modeling to Deployed Generative Dialogue
di: Pareek, Anant
Pubblicazione: (2025)

A Simple Reduction Scheme for Constrained Contextual Bandits with Adversarial Contexts via Regression
di: Sarkar, Dhruv, et al.
Pubblicazione: (2026)

Decomposing Attention To Find Context-Sensitive Neurons
di: Gibson, Alex
Pubblicazione: (2025)

How Transformers Learn In-Context Recall Tasks? Optimality, Training Dynamics and Generalization
di: Nguyen, Quan, et al.
Pubblicazione: (2025)

Computational-Statistical Tradeoffs at the Next-Token Prediction Barrier: Autoregressive and Imitation Learning under Misspecification
di: Rohatgi, Dhruv, et al.
Pubblicazione: (2025)

The Entropy of Floating-Point Numbers
di: Daniels, Sultan, et al.
Pubblicazione: (2026)

Decomposing and Editing Predictions by Modeling Model Computation
di: Shah, Harshay, et al.
Pubblicazione: (2024)

Summing Up the Facts: Additive Mechanisms Behind Factual Recall in LLMs
di: Chughtai, Bilal, et al.
Pubblicazione: (2024)

ThinkSwitch: Context Distillation with LoRA and Weight Interpolation for Specific-Purpose Reasoning Tasks
di: Saini, Dhruv, et al.
Pubblicazione: (2026)

Secure Generalization through Stochastic Bidirectional Parameter Updates Using Dual-Gradient Mechanism
di: Goel, Shourya, et al.
Pubblicazione: (2025)

Probably Approximately Precision and Recall Learning
di: Cohen, Lee, et al.
Pubblicazione: (2024)

Precision and Recall Reject Curves for Classification
di: Fischer, Lydia, et al.
Pubblicazione: (2023)

Self-Consolidating Language Models: Continual Knowledge Incorporation from Context
di: Wang, Zekun, et al.
Pubblicazione: (2026)

Understanding Contextual Recall in Transformers: How Finetuning Enables In-Context Reasoning over Pretraining Knowledge
di: Vasudeva, Bhavya, et al.
Pubblicazione: (2026)

Exploring Context Window of Large Language Models via Decomposed Positional Vectors
di: Dong, Zican, et al.
Pubblicazione: (2024)

Sense and Sensitivity: Examining the Influence of Semantic Recall on Long Context Code Reasoning
di: Štorek, Adam, et al.
Pubblicazione: (2025)

Heterogeneous Federated Learning Systems for Time-Series Power Consumption Prediction with Multi-Head Embedding Mechanism
di: Syu, Jia-Hao, et al.
Pubblicazione: (2025)

Interpreting Key Mechanisms of Factual Recall in Transformer-Based Language Models
di: Lv, Ang, et al.
Pubblicazione: (2024)

GeoTransolver: Learning Physics on Irregular Domains Using Multi-scale Geometry Aware Physics Attention Transformer
di: Adams, Corey, et al.
Pubblicazione: (2025)

Learning to Recall with Transformers Beyond Orthogonal Embeddings
di: Vural, Nuri Mert, et al.
Pubblicazione: (2026)

FLAME: Adaptive Mixture-of-Experts for Continual Multimodal Multi-Task Learning
di: Han, Xing, et al.
Pubblicazione: (2026)

FullRecall: A Semantic Search-Based Ranking Approach for Maximizing Recall in Patent Retrieval
di: Ali, Amna, et al.
Pubblicazione: (2025)

Decomposable Neuro Symbolic Regression
di: Morales, Giorgio, et al.
Pubblicazione: (2025)

Decomposable Transformer Point Processes
di: Panos, Aristeidis
Pubblicazione: (2024)

Self-Improvement in Language Models: The Sharpening Mechanism
di: Huang, Audrey, et al.
Pubblicazione: (2024)