Copertina

Can RL Teach Long-Horizon Reasoning to LLMs? Expressiveness Is Key

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Wang, Tianle, Wang, Zhaoyang, Lan, Guangchen, Wei, Xinpeng, Zhang, Sipeng, Qiu, Guanwen, Saparov, Abulhair
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Artificial Intelligence Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Do Language Models Follow Occam's Razor? An Evaluation of Parsimony in Inductive and Abductive Reasoning
di: Sun, Yunxin, et al.
Pubblicazione: (2025)

Transformers Can Learn Connectivity in Some Graphs but Not Others
di: Roy, Amit, et al.
Pubblicazione: (2025)

Language Models Might Not Understand You: Evaluating Theory of Mind via Story Prompting
di: Getachew, Nathaniel, et al.
Pubblicazione: (2025)

LLMs Are Prone to Fallacies in Causal Inference
di: Joshi, Nitish, et al.
Pubblicazione: (2024)

Reasoning Models Reason Well, Until They Don't
di: Rameshkumar, Revanth, et al.
Pubblicazione: (2025)

Learning to Reason Efficiently with A* Post-Training
di: Opedal, Andreas, et al.
Pubblicazione: (2026)

MaPPO: Maximum a Posteriori Preference Optimization with Prior Knowledge
di: Lan, Guangchen, et al.
Pubblicazione: (2025)

MathGAP: Out-of-Distribution Evaluation on Problems with Arbitrarily Complex Proofs
di: Opedal, Andreas, et al.
Pubblicazione: (2024)

Personas as a Way to Model Truthfulness in Language Models
di: Joshi, Nitish, et al.
Pubblicazione: (2023)

A Practical Review of Mechanistic Interpretability for Transformer-Based Language Models
di: Rai, Daking, et al.
Pubblicazione: (2024)

Can LLMs Learn by Teaching for Better Reasoning? A Preliminary Study
di: Ning, Xuefei, et al.
Pubblicazione: (2024)

Does RAG Know When Retrieval Is Wrong? Diagnosing Context Compliance under Knowledge Conflict
di: Chen, Yihang, et al.
Pubblicazione: (2026)

Are Language Models Efficient Reasoners? A Perspective from Logic Programming
di: Opedal, Andreas, et al.
Pubblicazione: (2025)

Efficient Long CoT Reasoning in Small Language Models
di: Wang, Zhaoyang, et al.
Pubblicazione: (2025)

Long-context LLMs Struggle with Long In-context Learning
di: Li, Tianle, et al.
Pubblicazione: (2024)

Reinforcement Learning for Scalable and Trustworthy Intelligent Systems
di: Lan, Guangchen
Pubblicazione: (2026)

Do Language Models Exhibit the Same Cognitive Biases in Problem Solving as Human Learners?
di: Opedal, Andreas, et al.
Pubblicazione: (2024)

R-Horizon: How Far Can Your Large Reasoning Model Really Go in Breadth and Depth?
di: Lu, Yi, et al.
Pubblicazione: (2025)

Can LLMs Reason Over Non-Text Modalities in a Training-Free Manner? A Case Study with In-Context Representation Learning
di: Zhang, Tianle, et al.
Pubblicazione: (2025)

Beyond Ten Turns: Unlocking Long-Horizon Agentic Search with Large-Scale Asynchronous RL
di: Gao, Jiaxuan, et al.
Pubblicazione: (2025)

ExpressivityBench: Can LLMs Communicate Implicitly?
di: Tint, Joshua, et al.
Pubblicazione: (2024)

LightReasoner: Can Small Language Models Teach Large Language Models Reasoning?
di: Wang, Jingyuan, et al.
Pubblicazione: (2025)

EcoGym: Evaluating LLMs for Long-Horizon Plan-and-Execute in Interactive Economies
di: Hu, Xavier, et al.
Pubblicazione: (2026)

COMPASS: Enhancing Agent Long-Horizon Reasoning with Evolving Context
di: Wan, Guangya, et al.
Pubblicazione: (2025)

From Long to Short: LLMs Excel at Trimming Own Reasoning Chains
di: Han, Wei, et al.
Pubblicazione: (2025)

AgentFugue: Agent Scaling for Long-Horizon Tasks through Collective Reasoning
di: Hu, Yuyang, et al.
Pubblicazione: (2026)

RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Horizon Generation
di: Wang, Zihao, et al.
Pubblicazione: (2024)

LLMs Can Teach Themselves to Better Predict the Future
di: Turtel, Benjamin, et al.
Pubblicazione: (2025)

Teaching LLMs According to Their Aptitude: Adaptive Reasoning for Mathematical Problem Solving
di: Xu, Xin, et al.
Pubblicazione: (2025)

UltraHorizon: Benchmarking Agent Capabilities in Ultra Long-Horizon Scenarios
di: Luo, Haotian, et al.
Pubblicazione: (2025)

Activation-aware Probe-Query: Effective Key-Value Retrieval for Long-Context LLMs Inference
di: Xiao, Qingfa, et al.
Pubblicazione: (2025)

Transformers Struggle to Learn to Search
di: Saparov, Abulhair, et al.
Pubblicazione: (2024)

Miner:Mining Intrinsic Mastery for Data-Efficient RL in Large Reasoning Models
di: Jiang, Shuyang, et al.
Pubblicazione: (2026)

Word Form Matters: LLMs' Semantic Reconstruction under Typoglycemia
di: Wang, Chenxi, et al.
Pubblicazione: (2025)

The Optimal Token Baseline: Variance Reduction for Long-Horizon LLM-RL
di: Li, Yingru, et al.
Pubblicazione: (2026)

The Challenge of Teaching Reasoning to LLMs Without RL or Distillation
di: Du, Wei, et al.
Pubblicazione: (2025)

CLARity: Reasoning Consistency Alone Can Teach Reinforced Experts
di: Lin, Jiuheng, et al.
Pubblicazione: (2025)

HorizonBench: Long-Horizon Personalization with Evolving Preferences
di: Li, Shuyue Stella, et al.
Pubblicazione: (2026)

CHI-Bench: Can AI Agents Automate End-to-End, Long-Horizon, Policy-Rich Healthcare Workflows?
di: Chen, Haolin, et al.
Pubblicazione: (2026)

OPSD Compresses What RLVR Teaches: A Post-RL Compaction Stage for Reasoning Models
di: Kim, Jaehoon, et al.
Pubblicazione: (2026)