Copertina

Replacing thinking with tool usage enables reasoning in small language models

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Rainone, Corrado, Bakker, Tim, Memisevic, Roland
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Machine Learning Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Language hooks: a modular framework for augmenting LLM reasoning that decouples tool usage from the model and its prompt
di: de Mijolla, Damien, et al.
Pubblicazione: (2024)

Look, Remember and Reason: Grounded reasoning in videos with language models
di: Bhattacharyya, Apratim, et al.
Pubblicazione: (2023)

CodeIt: Self-Improving Language Models with Prioritized Hindsight Replay
di: Butt, Natasha, et al.
Pubblicazione: (2024)

Context information can be more important than reasoning for time series forecasting with a large language model
di: Yang, Janghoon
Pubblicazione: (2025)

A note on the impossibility of conditional PAC-efficient reasoning in large language models
di: Zeng, Hao
Pubblicazione: (2025)

Multi-step retrieval and reasoning improves radiology question answering with large language models
di: Wind, Sebastian, et al.
Pubblicazione: (2025)

Performance of AI agents based on reasoning language models on ALD process optimization tasks
di: Yanguas-Gil, Angel
Pubblicazione: (2026)

Brain alignment of reasoning and action representations from vision-language and action models during naturalistic gameplay
di: Oota, Subba Reddy, et al.
Pubblicazione: (2026)

Vision-language models lag human performance on physical dynamics and intent reasoning
di: Gu, Tianjun, et al.
Pubblicazione: (2026)

Self-rewarding correction for mathematical reasoning
di: Xiong, Wei, et al.
Pubblicazione: (2025)

Insights into resource utilization of code small language models serving with runtime engines and execution providers
di: Durán, Francisco, et al.
Pubblicazione: (2024)

Agentic retrieval-augmented reasoning reshapes collective reliability under model variability in radiology question answering
di: Farajiamiri, Mina, et al.
Pubblicazione: (2026)

Linguistic properties and model scale in brain encoding: from small to compressed language models
di: Oota, Subba Reddy, et al.
Pubblicazione: (2026)

How predictable is language model benchmark performance?
di: Owen, David
Pubblicazione: (2024)

Emergent temporal abstractions in autoregressive models enable hierarchical reinforcement learning
di: Kobayashi, Seijin, et al.
Pubblicazione: (2025)

Bilinear representation mitigates reversal curse and enables consistent model editing
di: Kim, Dong-Kyum, et al.
Pubblicazione: (2025)

Reinforcement Learning in hyperbolic space for multi-step reasoning
di: Xu, Tao, et al.
Pubblicazione: (2025)

ProvMind: Provenance-grounded reasoning for materials synthesis
di: Zhang, Yiming, et al.
Pubblicazione: (2026)

Alignment faking in large language models
di: Greenblatt, Ryan, et al.
Pubblicazione: (2024)

Efficiency optimization of large-scale language models based on deep learning in natural language processing tasks
di: Mei, Taiyuan, et al.
Pubblicazione: (2024)

When Cloud Agents Meet Device Agents: Lessons from Hybrid Multi-Agent Systems
di: Rainone, Corrado, et al.
Pubblicazione: (2026)

Language models show human-like content effects on reasoning tasks
di: Dasgupta, Ishita, et al.
Pubblicazione: (2022)

Fitness aligned structural modeling enables scalable virtual screening with AuroBind
di: Zhang, Zhongyue, et al.
Pubblicazione: (2025)

Quantifying construct validity in large language model evaluations
di: Kearns, Ryan Othniel
Pubblicazione: (2026)

Applying sparse autoencoders to unlearn knowledge in language models
di: Farrell, Eoin, et al.
Pubblicazione: (2024)

Exploring System 1 and 2 communication for latent reasoning in LLMs
di: Coda-Forno, Julian, et al.
Pubblicazione: (2025)

Slm-mux: Orchestrating small language models for reasoning
di: Wang, Chenyu, et al.
Pubblicazione: (2025)

Deliberative Alignment is Deep, but Uncertainty Remains: Inference time safety improvement in reasoning via attribution of unsafe behavior to base model
di: Pathmanathan, Pankayaraj, et al.
Pubblicazione: (2026)

The language of time: a language model perspective on time-series foundation models
di: Xie, Yi, et al.
Pubblicazione: (2025)

Auditing language models for hidden objectives
di: Marks, Samuel, et al.
Pubblicazione: (2025)

Large language models as uncertainty-calibrated optimizers for experimental discovery
di: Ranković, Bojana, et al.
Pubblicazione: (2025)

Asymmetric Proximal Policy Optimization: mini-critics boost LLM reasoning
di: Liu, Jiashun, et al.
Pubblicazione: (2025)

Reinforcing privacy reasoning in LLMs via normative simulacra from fiction
di: Franchi, Matt, et al.
Pubblicazione: (2026)

Math Takes Two: A test for emergent mathematical reasoning in communication
di: Cooper, Michael, et al.
Pubblicazione: (2026)

DiPT: Enhancing LLM reasoning through diversified perspective-taking
di: Just, Hoang Anh, et al.
Pubblicazione: (2024)

Can machines think efficiently?
di: Winchell, Adam
Pubblicazione: (2025)

GRADE: Replacing Policy Gradients with Backpropagation for LLM Alignment
di: Nel, Lukas Abrie
Pubblicazione: (2025)

SCORE: Replacing Layer Stacking with Contractive Recurrent Depth
di: Godin, Guillaume
Pubblicazione: (2026)

Improving the Diversity of Bootstrapped DQN by Replacing Priors With Noise
di: Meng, Li, et al.
Pubblicazione: (2022)

Code-enabled language models can outperform reasoning models on diverse tasks
di: Zhang, Cedegao E., et al.
Pubblicazione: (2025)