Cover Image

Learning When to Plan: Efficiently Allocating Test-Time Compute for LLM Agents

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Paglieri, Davide, Cupiał, Bartłomiej, Cook, Jonathan, Piterbarg, Ulyana, Tuyls, Jens, Grefenstette, Edward, Foerster, Jakob Nicolaus, Parker-Holder, Jack, Rocktäschel, Tim
Format:	Preprint
Published:	2025
Subjects:	Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

BALROG: Benchmarking Agentic LLM and VLM Reasoning On Games
by: Paglieri, Davide, et al.
Published: (2024)

Outliers and Calibration Sets have Diminishing Effect on Quantization of Modern LLMs
by: Paglieri, Davide, et al.
Published: (2024)

Multi-Agent Diagnostics for Robustness via Illuminated Diversity
by: Samvelyan, Mikayel, et al.
Published: (2024)

Preference-Based Alignment of Discrete Diffusion Models
by: Borso, Umberto, et al.
Published: (2025)

TICKing All the Boxes: Generated Checklists Improve LLM Evaluation and Generation
by: Cook, Jonathan, et al.
Published: (2024)

Programming by Backprop: An Instruction is Worth 100 Examples When Finetuning LLMs
by: Cook, Jonathan, et al.
Published: (2025)

Imagined Autocurricula
by: Güzel, Ahmet H., et al.
Published: (2025)

minimax: Efficient Baselines for Autocurricula in JAX
by: Jiang, Minqi, et al.
Published: (2023)

Scaling Opponent Shaping to High Dimensional Games
by: Khan, Akbir, et al.
Published: (2023)

diff History for Neural Language Agents
by: Piterbarg, Ulyana, et al.
Published: (2023)

Training Language Models on Synthetic Edit Sequences Improves Code Synthesis
by: Piterbarg, Ulyana, et al.
Published: (2024)

Assessing the Zero-Shot Capabilities of LLMs for Action Evaluation in RL
by: Pignatelli, Eduardo, et al.
Published: (2024)

Open-Endedness is Essential for Artificial Superhuman Intelligence
by: Hughes, Edward, et al.
Published: (2024)

Infusion: Shaping Model Behavior by Editing Training Data via Influence Functions
by: Rosser, J, et al.
Published: (2026)

DéjàQ: Open-Ended Evolution of Diverse, Learnable and Verifiable Problems
by: Röpke, Willem, et al.
Published: (2026)

Learning Multi-Agent Coordination via Sheaf-ADMM
by: Seely, Jeffrey, et al.
Published: (2026)

Rainbow Teaming: Open-Ended Generation of Diverse Adversarial Prompts
by: Samvelyan, Mikayel, et al.
Published: (2024)

Artificial Generational Intelligence: Cultural Accumulation in Reinforcement Learning
by: Cook, Jonathan, et al.
Published: (2024)

Learning to Learn from Language Feedback with Social Meta-Learning
by: Cook, Jonathan, et al.
Published: (2026)

Representation-Based Exploration for Language Models: From Test-Time to Post-Training
by: Tuyls, Jens, et al.
Published: (2025)

Mechanistically analyzing the effects of fine-tuning on procedurally defined tasks
by: Jain, Samyak, et al.
Published: (2023)

Interaction Dynamics as a Reward Signal for LLMs
by: Gooding, Sian, et al.
Published: (2025)

Debating with More Persuasive LLMs Leads to More Truthful Answers
by: Khan, Akbir, et al.
Published: (2024)

Procedural Knowledge in Pretraining Drives Reasoning in Large Language Models
by: Ruis, Laura, et al.
Published: (2024)

Synthetic Data is Sufficient for Zero-Shot Visual Generalization from Offline Data
by: Güzel, Ahmet H., et al.
Published: (2025)

LLM-First Search: Self-Guided Exploration of the Solution Space
by: Herr, Nathan, et al.
Published: (2025)

Representation Invariance and Allocation: When Subgroup Balance Matters
by: Alloula, Anissa, et al.
Published: (2025)

Can a MISL Fly? Analysis and Ingredients for Mutual Information Skill Learning
by: Zheng, Chongyi, et al.
Published: (2024)

Computing the Cousin-Zuckerman Resolution and the Lusztig-Vogan Bijection
by: Cook, Jack A.
Published: (2026)

AI & Human Co-Improvement for Safer Co-Superintelligence
by: Weston, Jason, et al.
Published: (2025)

AgentBreeder: Mitigating the AI Safety Risks of Multi-Agent Scaffolds via Self-Improvement
by: Rosser, J, et al.
Published: (2025)

Improving Interactive In-Context Learning from Natural Language Feedback
by: Klissarov, Martin, et al.
Published: (2026)

Investigating Non-Transitivity in LLM-as-a-Judge
by: Xu, Yi, et al.
Published: (2025)

When to Ponder: Adaptive Compute Allocation for Code Generation via Test-Time Training
by: Sim, Gihyeon
Published: (2025)

Writing as a testbed for open ended agents
by: Gooding, Sian, et al.
Published: (2025)

Language-Guided World Models: A Model-Based Approach to AI Control
by: Zhang, Alex, et al.
Published: (2024)

Multi-Agent Craftax: Benchmarking Open-Ended Multi-Agent Reinforcement Learning at the Hyperscale
by: Omari, Bassel Al, et al.
Published: (2025)

Select to Perfect: Imitating desired behavior from large multi-agent data
by: Franzmeyer, Tim, et al.
Published: (2024)

Ad-Hoc Human-AI Coordination Challenge
by: Dizdarević, Tin, et al.
Published: (2025)

Fine-tuning Reinforcement Learning Models is Secretly a Forgetting Mitigation Problem
by: Wołczyk, Maciej, et al.
Published: (2024)