Imagen de Portada

Universal Transformers Need Memory: Depth-State Trade-offs in Adaptive Recursive Reasoning

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autor principal:	Sapunov, Grigory
Formato:	Preprint
Publicado:	2026
Materias:	Machine Learning Artificial Intelligence Computation and Language I.2.6
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Graph Memory Transformer (GMT)
por: Zanarini, Nicola, et al.
Publicado: (2026)

Beyond Pass@k: Breadth-Depth Metrics for Reasoning Boundaries
por: Dragoi, Marius, et al.
Publicado: (2025)

The Mirror Loop: Recursive Non-Convergence in Generative Reasoning Systems
por: DeVilling, Bentley
Publicado: (2025)

Forget Attention: Importance-Aware Attention Is All You Need
por: Shin, Soohyeong, et al.
Publicado: (2026)

Synthius-Mem: Brain-Inspired Hallucination-Resistant Persona Memory Achieving 94.4% Memory Accuracy and 99.6% Adversarial Robustness on LoCoMo
por: Gadzhiev, Artem, et al.
Publicado: (2026)

Deep Memory Search: A Metaheuristic Approach for Optimizing Heuristic Search
por: Hedar, Abdel-Rahman, et al.
Publicado: (2024)

Contextual Integrity in LLMs via Reasoning and Reinforcement Learning
por: Lan, Guangchen, et al.
Publicado: (2025)

Counterfactual Likelihood Tests for Indirect Influence in Private Reasoning Channels
por: Lorup, Alexander Boesgaard
Publicado: (2026)

No Free Swap: Protocol-Dependent Layer Redundancy in Transformers
por: Garcia, Gabriel
Publicado: (2026)

Weakly Supervised Distillation of Hallucination Signals into Transformer Representations
por: Salehmohamed, Shoaib Sadiq, et al.
Publicado: (2026)

Prototype Transformer: Towards Language Model Architectures Interpretable by Design
por: Yordanov, Yordan, et al.
Publicado: (2026)

The Deterministic Horizon: When Extended Reasoning Fails and Tool Delegation Becomes Necessary
por: Guo, Dongxin, et al.
Publicado: (2026)

Cognitive Load Limits in Large Language Models: Benchmarking Multi-Hop Reasoning
por: Adapala, Sai Teja Reddy
Publicado: (2025)

Harnessing Negative Signals: Reinforcement Distillation from Teacher Data for LLM Reasoning
por: Xu, Shuyao, et al.
Publicado: (2025)

Generalizing Numerical Reasoning in Table Data through Operation Sketches and Self-Supervised Learning
por: Cho, Hanjun, et al.
Publicado: (2026)

Perturbation Dose Responses in Recursive LLM Loops: Raw Switching, Stochastic Floors, and Persistent Escape under Append, Replace, and Dialog Updates
por: Kaplanski, Pawel
Publicado: (2026)

Entropy-Based Measurement of Value Drift and Alignment Work in Large Language Models
por: Fadli, Samih
Publicado: (2025)

Grokking in the Wild: Data Augmentation for Real-World Multi-Hop Reasoning with Transformers
por: Abramov, Roman, et al.
Publicado: (2025)

Continuous-Depth Transformers with Learned Control Dynamics
por: Jemley, Peter
Publicado: (2026)

Cross-Entropy Is Load-Bearing: A Pre-Registered Scope Test of the K-Way Energy Probe on Bidirectional Predictive Coding
por: Cacioli, Jon-Paul
Publicado: (2026)

Towards Understanding Sycophancy in Language Models
por: Sharma, Mrinank, et al.
Publicado: (2023)

Characterizing Pattern Matching and Its Limits on Compositional Task Structures
por: Chang, Hoyeon, et al.
Publicado: (2025)

Ouroboros: Dynamic Weight Generation for Recursive Transformers via Input-Conditioned LoRA Modulation
por: Jaber, Jaber, et al.
Publicado: (2026)

QuIDE: Mastering the Quantized Intelligence Trade-off via Active Optimization
por: Jiang, Xiantao
Publicado: (2026)

Dynamic Policy Induction for Adaptive Prompt Optimization: Bridging the Efficiency-Accuracy Gap via Lightweight Reinforcement Learning
por: Xu, Jiexi
Publicado: (2025)

Adaptive Activation Cancellation for Hallucination Mitigation in Large Language Models
por: Yocam, Eric, et al.
Publicado: (2026)

Control Reinforcement Learning: Interpretable Token-Level Steering of LLMs via Sparse Autoencoder Features
por: Cho, Seonglae, et al.
Publicado: (2026)

In-Context Fixation: When Demonstrated Labels Override Semantics in Few-Shot Classification
por: Liu, Ming
Publicado: (2026)

The Last Word Often Wins: A Format Confound in Chain-of-Thought Corruption Studies
por: Garcia, Gabriel
Publicado: (2026)

TIAR: Trajectory-Informed Advantage Reweighting for LLM Abstention Learning
por: Pan, Muyu, et al.
Publicado: (2026)

Pressure-Testing Deception Probes in LLMs: Scaling, Robustness, and the Geometry of Deceptive Representations
por: Kumar, Sachin
Publicado: (2026)

Structured Prompt Optimization Meets Reinforcement Learning for Global and Local Interpretability over Complex Text
por: Zhou, Tianyang, et al.
Publicado: (2026)

AMEL: Accumulated Message Effects on LLM Judgments
por: Temkit, Sid-Ali
Publicado: (2026)

Turning the TIDE: Cross-Architecture Distillation for Diffusion Large Language Models
por: Zhang, Gongbo, et al.
Publicado: (2026)

Model Collapse as Cultural Evolution
por: Guo, Dongxin, et al.
Publicado: (2026)

Alternating Reinforcement Learning with Contextual Rubric Rewards: Beyond the Scalarization Strategy
por: Lan, Guangchen, et al.
Publicado: (2026)

Self-Training Doesn't Flatten Language -- It Restructures It: Surface Markers Amplify While Deep Syntax Dies
por: Liu, Ming
Publicado: (2026)

The Readout Shortcut: Positional Number Copying Dominates Arithmetic CoT Readout in Small Language Models
por: Liu, Ming
Publicado: (2026)

Revisiting Intermediate-Layer Matching in Knowledge Distillation: Layer-Selection Strategy Doesn't Matter (Much)
por: Yu, Zony, et al.
Publicado: (2025)

Painless Activation Steering: An Automated, Lightweight Approach for Post-Training Large Language Models
por: Cui, Sasha, et al.
Publicado: (2025)