Copertina

Learning When to Think: Shaping Adaptive Reasoning in R1-Style Models via Multi-Stage RL

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Tu, Songjun, Lin, Jiahao, Zhang, Qichao, Tian, Xiangyu, Li, Linjing, Lan, Xiangyuan, Zhao, Dongbin
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Computation and Language Artificial Intelligence 68T50 I.2.7
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

PaperAudit-Bench: Benchmarking Error Detection in Research Papers for Critical Automated Peer Review
di: Tu, Songjun, et al.
Pubblicazione: (2026)

Perception-Consistency Multimodal Large Language Models Reasoning via Caption-Regularized Policy Optimization
di: Tu, Songjun, et al.
Pubblicazione: (2025)

Fast Quiet-STaR: Thinking Without Thought Tokens
di: Huang, Wei, et al.
Pubblicazione: (2025)

Bidirectional RAG: Safe Self-Improving Retrieval-Augmented Generation Through Multi-Stage Validation
di: Chinthala, Teja
Pubblicazione: (2025)

Thinking Longer, Not Always Smarter: Evaluating LLM Capabilities in Hierarchical Legal Reasoning
di: Zhang, Li, et al.
Pubblicazione: (2025)

One Agent to Serve All: a Lite-Adaptive Stylized AI Assistant for Millions of Multi-Style Official Accounts
di: Fan, Xingyu, et al.
Pubblicazione: (2025)

Co-NAML-LSTUR: A Combined Model with Attentive Multi-View Learning and Long- and Short-term User Representations for News Recommendation
di: Nguyen, Minh Hoang, et al.
Pubblicazione: (2025)

When Retrieval Succeeds and Fails: Rethinking Retrieval-Augmented Generation for LLMs
di: Wang, Yongjie, et al.
Pubblicazione: (2025)

Can LLMs Compute with Reasons?
di: Sandilya, Harshit, et al.
Pubblicazione: (2024)

Distractor Injection Attacks on Large Reasoning Models: Characterization and Defense
di: Zhang, Zhehao, et al.
Pubblicazione: (2025)

Adaptive Steering and Remasking for Safe Generation in Diffusion Language Models
di: Lee, Yejin, et al.
Pubblicazione: (2026)

GATE: Graph-based Adaptive Tool Evolution Across Diverse Tasks
di: Luo, Jianwen, et al.
Pubblicazione: (2025)

New Skills or Sharper Primitives? A Probabilistic Perspective on the Emergence of Reasoning in RLVR
di: Wang, Zhilin, et al.
Pubblicazione: (2026)

Culturally-Nuanced Story Generation for Reasoning in Low-Resource Languages: The Case of Javanese and Sundanese
di: Pranida, Salsabila Zahirah, et al.
Pubblicazione: (2025)

Dynamic Dual-Granularity Skill Bank for Agentic RL
di: Tu, Songjun, et al.
Pubblicazione: (2026)

A Knowledge Enhanced Learning and Semantic Composition Model for Multi-Claim Fact Checking
di: Wang, Shuai, et al.
Pubblicazione: (2021)

Optimizing What We Trust: Reliability-Guided QUBO Selection of Multi-Agent Weak Framing Signals for Arabic Sentiment Prediction
di: Alkhalifa, Rabab
Pubblicazione: (2026)

CLMN: Concept based Language Models via Neural Symbolic Reasoning
di: Yang, Yibo
Pubblicazione: (2025)

An NLP-Driven Framework for Curriculum-Labor Market Alignment: Schema-Constrained LLM Extraction, ESCO-Anchored Semantic Matching, and Multi-Dimensional Gap Quantification
di: Turaev, Sherzod, et al.
Pubblicazione: (2026)

When Does Content-Based Routing Work? Representation Requirements for Selective Attention in Hybrid Sequence Models
di: Basu, Abhinaba
Pubblicazione: (2026)

Rethinking the Multilingual Reasoning Gap with Layer Swap
di: Lasbordes, Maxence, et al.
Pubblicazione: (2026)

From Noise to Diversity: Random Embedding Injection in LLM Reasoning
di: Kim, Heejun, et al.
Pubblicazione: (2026)

Multi-chain Graph Refinement and Selection for Reliable Reasoning in Large Language Models
di: Yang, Yujiao, et al.
Pubblicazione: (2025)

Adaptive Multi-Stage Patent Claim Generation with Unified Quality Assessment
di: Liang, Chen-Wei, et al.
Pubblicazione: (2026)

ADE: Adaptive Dictionary Embeddings -- Scaling Multi-Anchor Representations to Large Language Models
di: Demirci, Orhan, et al.
Pubblicazione: (2026)

D-SMART: Enhancing LLM Dialogue Consistency via Dynamic Structured Memory And Reasoning Tree
di: Lei, Xiang, et al.
Pubblicazione: (2025)

Emotional Sequential Influence Modeling on False Information
di: Naskar, Debashis, et al.
Pubblicazione: (2024)

Isolating LLM Lexical Bias: A Curation-Free Triangulated Metric for Preference-Stage Learning
di: Ming, Xiaoyang, et al.
Pubblicazione: (2026)

The Knesset Corpus: An Annotated Corpus of Hebrew Parliamentary Proceedings
di: Goldin, Gili, et al.
Pubblicazione: (2024)

Math Natural Language Inference: this should be easy!
di: de Paiva, Valeria, et al.
Pubblicazione: (2025)

Pitfalls in Evaluating Interpretability Agents
di: Haklay, Tal, et al.
Pubblicazione: (2026)

d-TreeRPO: Towards More Reliable Policy Optimization for Diffusion Language Models
di: Pan, Leyi, et al.
Pubblicazione: (2025)

Towards Effective and Efficient Continual Pre-training of Large Language Models
di: Chen, Jie, et al.
Pubblicazione: (2024)

An Unforgeable Publicly Verifiable Watermark for Large Language Models
di: Liu, Aiwei, et al.
Pubblicazione: (2023)

The Unlikely Duel: Evaluating Creative Writing in LLMs through a Unique Scenario
di: Gómez-Rodríguez, Carlos, et al.
Pubblicazione: (2024)

Direct Large Language Model Alignment Through Self-Rewarding Contrastive Prompt Distillation
di: Liu, Aiwei, et al.
Pubblicazione: (2024)

SentiCSE: A Sentiment-aware Contrastive Sentence Embedding Framework with Sentiment-guided Textual Similarity
di: Kim, Jaemin, et al.
Pubblicazione: (2024)

Exploiting Pre-trained Encoder-Decoder Transformers for Sequence-to-Sequence Constituent Parsing
di: Fernández-González, Daniel, et al.
Pubblicazione: (2026)

Parametric Social Identity Injection and Diversification in Public Opinion Simulation
di: Wang, Hexi, et al.
Pubblicazione: (2026)

Trusted Uncertainty in Large Language Models: A Unified Framework for Confidence Calibration and Risk-Controlled Refusal
di: Oehri, Markus, et al.
Pubblicazione: (2025)