Imagen de Portada

Provable In-Context Vector Arithmetic via Retrieving Task Concepts

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Bu, Dake, Huang, Wei, Han, Andi, Nitanda, Atsushi, Zhang, Qingfu, Wong, Hau-San, Suzuki, Taiji
Formato:	Preprint
Publicado:	2025
Materias:	Machine Learning Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Provably Transformers Harness Multi-Concept Word Semantics for Efficient In-Context Learning
por: Bu, Dake, et al.
Publicado: (2024)

DPRM: A Plug-in Doob h transform-induced Token-Ordering Module for Diffusion Language Models
por: Bu, Dake, et al.
Publicado: (2026)

Post-Training as Reweighting: A Stochastic View of Reasoning Trajectories in Language Models
por: Bu, Dake, et al.
Publicado: (2025)

Provable Benefit of Curriculum in Transformer Tree-Reasoning Post-Training
por: Bu, Dake, et al.
Publicado: (2025)

Provably Neural Active Learning Succeeds via Prioritizing Perplexing Samples
por: Bu, Dake, et al.
Publicado: (2024)

Provably Learning Diffusion Models under the Manifold Hypothesis: Collapse and Refine
por: Huang, Wei, et al.
Publicado: (2026)

Direct Distributional Optimization for Provable Alignment of Diffusion Models
por: Kawata, Ryotaro, et al.
Publicado: (2025)

Metastable Dynamics of Chain-of-Thought Reasoning: Provable Benefits of Search, RL and Distillation
por: Kim, Juno, et al.
Publicado: (2025)

On the Learning Dynamics of Two-layer Linear Networks with Label Noise SGD
por: Zhang, Tongcheng, et al.
Publicado: (2026)

Slowly Annealed Langevin Dynamics: Theory and Applications to Training-Free Guided Generation
por: Nitanda, Atsushi, et al.
Publicado: (2026)

Towards a Unified Analysis of Neural Networks in Nonparametric Instrumental Variable Regression: Optimization and Generalization
por: Chen, Zonghao, et al.
Publicado: (2025)

Escaping Optimization Stagnation: Taking Steps Beyond Task Arithmetic via Difference Vectors
por: Wang, Jinping, et al.
Publicado: (2025)

Intrinsic Wasserstein Rates for Score-Based Generative Models on Smooth Manifolds
por: Fu, Guoji, et al.
Publicado: (2026)

Backdoor Vectors: a Task Arithmetic View on Backdoor Attacks and Defenses
por: Pawlak, Stanisław, et al.
Publicado: (2025)

In-Context Learning Is Provably Bayesian Inference: A Generalization Theory for Meta-Learning
por: Wakayama, Tomoya, et al.
Publicado: (2025)

Emergence and Effectiveness of Task Vectors in In-Context Learning: An Encoder Decoder Perspective
por: Han, Seungwook, et al.
Publicado: (2024)

Koopman-based generalization bound: New aspect for full-rank weights
por: Hashimoto, Yuka, et al.
Publicado: (2023)

Bus-Conditioned Zero-Shot Trajectory Generation via Task Arithmetic
por: Liu, Shuai, et al.
Publicado: (2026)

Divide, Reweight, and Conquer: A Logit Arithmetic Approach for In-Context Learning
por: Huang, Chengsong, et al.
Publicado: (2024)

Task-Aware LoRA Adapter Composition via Similarity Retrieval in Vector Databases
por: Adsul, Riya, et al.
Publicado: (2026)

Language Models Do Hard Arithmetic Tasks Easily and Hardly Do Easy Arithmetic Tasks
por: Gambardella, Andrew, et al.
Publicado: (2024)

Provable and Practical In-Context Policy Optimization for Self-Improvement
por: Yu, Tianrun, et al.
Publicado: (2026)

Concept Arithmetics for Circumventing Concept Inhibition in Diffusion Models
por: Petsiuk, Vitali, et al.
Publicado: (2024)

On Task Vectors and Gradients
por: Zhou, Luca, et al.
Publicado: (2025)

Refinement Contrastive Learning of Cell-Gene Associations for Unsupervised Cell Type Identification
por: Peng, Liang, et al.
Publicado: (2025)

VP-VAE: Rethinking Vector Quantization via Adaptive Vector Perturbation
por: Zhai, Linwei, et al.
Publicado: (2026)

Transformers Provably Implement In-Context Reinforcement Learning with Policy Improvement
por: Liang, Haodong, et al.
Publicado: (2026)

Improved Particle Approximation Error for Mean Field Neural Networks
por: Nitanda, Atsushi
Publicado: (2024)

Propagation of Chaos for Mean-Field Langevin Dynamics and its Application to Model Ensemble
por: Nitanda, Atsushi, et al.
Publicado: (2025)

Provable Training Data Identification for Large Language Models
por: Liu, Zhenlong, et al.
Publicado: (2025)

Transformers Provably Solve Parity Efficiently with Chain of Thought
por: Kim, Juno, et al.
Publicado: (2024)

One-Layer Transformer Provably Learns One-Nearest Neighbor In Context
por: Li, Zihao, et al.
Publicado: (2024)

Variational Task Vector Composition
por: Zhang, Boyuan, et al.
Publicado: (2025)

Principled Understanding of Generalization for Generative Transformer Models in Arithmetic Reasoning Tasks
por: Xu, Xingcheng, et al.
Publicado: (2024)

A Relative-Budget Theory for Reinforcement Learning with Verifiable Rewards in Large Language Model Reasoning
por: Wachi, Akifumi, et al.
Publicado: (2026)

Model-agnostic Selective Labeling with Provable Statistical Guarantees
por: Huang, Huipeng, et al.
Publicado: (2025)

Multi-Task Learning for Routing Problem with Cross-Problem Zero-Shot Generalization
por: Liu, Fei, et al.
Publicado: (2024)

Mamba Can Learn Low-Dimensional Targets In-Context via Test-Time Feature Learning
por: Oh, Junsoo, et al.
Publicado: (2025)

CoPS: Empowering LLM Agents with Provable Cross-Task Experience Sharing
por: Yang, Chen, et al.
Publicado: (2024)

Multimodal Task Vectors Enable Many-Shot Multimodal In-Context Learning
por: Huang, Brandon, et al.
Publicado: (2024)