Image de couverture de livre

Activation Manifold Projection: Liberating Task-Specific Behaviors from LLM Architectures

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteur principal:	Kari, Al
Format:	Preprint
Publié:	2025
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Projected Task-Specific Layers for Multi-Task Reinforcement Learning
par: Roberts, Josselin Somerville, et autres
Publié: (2023)

Task-Specific Activation Functions for Neuroevolution using Grammatical Evolution
par: Winter, Benjamin David, et autres
Publié: (2025)

Simplifying Multi-Task Architectures Through Task-Specific Normalization
par: Suteu, Mihai, et autres
Publié: (2025)

Local Manifold Approximation and Projection for Manifold-Aware Diffusion Planning
par: Lee, Kyowoon, et autres
Publié: (2025)

Manifold of Failure: Behavioral Attraction Basins in Language Models
par: Munshi, Sarthak, et autres
Publié: (2026)

Layers at Similar Depths Generate Similar Activations Across LLM Architectures
par: Wolfram, Christopher, et autres
Publié: (2025)

Structured Progressive Knowledge Activation for LLM-Driven Neural Architecture Search
par: Liu, Zhen, et autres
Publié: (2026)

Implicit Federated In-context Learning For Task-Specific LLM Fine-Tuning
par: Li, Dongcheng, et autres
Publié: (2025)

AgentRx: A Benchmark Study of LLM Agents for Multimodal Clinical Prediction Tasks
par: Jorf, Baraa Al, et autres
Publié: (2026)

Robo-Troj: Attacking LLM-based Task Planners
par: Nahian, Mohaiminul Al, et autres
Publié: (2025)

Toward Modeling Player-Specific Chess Behaviors
par: Sogliuzzo, Loris, et autres
Publié: (2026)

Towards Interpretable Foundation Models of Robot Behavior: A Task Specific Policy Generation Approach
par: Sheidlower, Isaac, et autres
Publié: (2024)

Decoupled Behavioral Cloning for Scalable Inductive Generalization in RL from Specifications
par: Subramanian, Vignesh, et autres
Publié: (2026)

Meta-Models: An Architecture for Decoding LLM Behaviors Through Interpreted Embeddings and Natural Language
par: Costarelli, Anthony, et autres
Publié: (2024)

Motion Manifold Flow Primitives for Task-Conditioned Trajectory Generation under Complex Task-Motion Dependencies
par: Lee, Yonghyeon, et autres
Publié: (2024)

Manifold-Matching Autoencoders
par: Cheret, Laurent, et autres
Publié: (2026)

Mano: Restriking Manifold Optimization for LLM Training
par: Gu, Yufei, et autres
Publié: (2026)

Fusion Steering: Prompt-Specific Activation Control
par: Chang, Waldemar, et autres
Publié: (2025)

Understanding Artificial Neural Network's Behavior from Neuron Activation Perspective
par: Zhang, Yizhou, et autres
Publié: (2024)

Understanding LLM Behavior When Encountering User-Supplied Harmful Content in Harmless Tasks
par: Chu, Junjie, et autres
Publié: (2026)

Modeling Task Immersion based on Goal Activation Mechanism
par: Nagashima, Kazuma, et autres
Publié: (2024)

One Request, Multiple Experts: LLM Orchestrates Domain Specific Models via Adaptive Task Routing
par: Yang, Xu, et autres
Publié: (2025)

Deep Extrinsic Manifold Representation for Vision Tasks
par: Zhang, Tongtong, et autres
Publié: (2024)

Chain-of-Thought in Large Language Models: Decoding, Projection, and Activation
par: Yang, Hao, et autres
Publié: (2024)

Watch your steps: Dormant Adversarial Behaviors that Activate upon LLM Finetuning
par: Gloaguen, Thibaud, et autres
Publié: (2025)

Interfaze: The Future of AI is built on Task-Specific Small Models
par: Khurdula, Harsha Vardhan, et autres
Publié: (2026)

How Chain-of-Thought Works? Tracing Information Flow from Decoding, Projection, and Activation
par: Yang, Hao, et autres
Publié: (2025)

Identity as Attractor: Geometric Evidence for Persistent Agent Architecture in LLM Activation Space
par: Vasilenko, Vladimir
Publié: (2026)

Hallucination Behavior in Multimodal LLMs Across Agricultural Image Interpretation and Generation Tasks
par: Ghose, Partho, et autres
Publié: (2026)

Outlier Dimensions Encode Task-Specific Knowledge
par: Rudman, William, et autres
Publié: (2023)

Spectral Manifold Regularization for Stable and Modular Routing in Deep MoE Architectures
par: Delibasoglu, Ibrahim
Publié: (2026)

Unveiling Project-Specific Bias in Neural Code Models
par: Li, Zhiming, et autres
Publié: (2022)

ROSE: A Reward-Oriented Data Selection Framework for LLM Task-Specific Instruction Tuning
par: Wu, Yang, et autres
Publié: (2024)

Demystifying Manifold Constraints in LLM Pre-training
par: An, Kang, et autres
Publié: (2026)

Autonomous Agency and Persistent Intent Architecture Specification v0.1
par: Brown, Cameron
Publié: (2026)

Specification and Detection of LLM Code Smells
par: Mahmoudi, Brahim, et autres
Publié: (2025)

How Many Parameters Does Your Task Really Need? Task Specific Pruning with LLM-Sieve
par: Reda, Waleed, et autres
Publié: (2025)

Distinguishing Task-Specific and General-Purpose AI in Regulation
par: Wang, Jennifer, et autres
Publié: (2025)

Provable Benefits of Task-Specific Prompts for In-context Learning
par: Chang, Xiangyu, et autres
Publié: (2025)

Task-Specific Knowledge Distillation via Intermediate Probes
par: Brown, Ryan, et autres
Publié: (2026)