Copertina

Where Hindsight Credit Can Reside: A Signed-Capacity View of Token Updates in RLVR

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	He, Yuhang, Wu, Haodong, Liu, Siyi, Ge, Hongyu, Zhou, Hange, Wu, Keyi, Zheng, Zhuo, Lin, Qihong, Zhong, Zixin, Zhang, Yongqi
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Machine Learning Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Hindsight Credit Assignment for Long-Horizon LLM Agents
di: Tan, Hui-Ze, et al.
Pubblicazione: (2026)

Where Rollouts Begin: Low-Load, High-Leverage First-Token Diversification for RLVR
di: Kim, Soeun, et al.
Pubblicazione: (2026)

Beyond Where to Look: Trajectory-Guided Reinforcement Learning for Multimodal RLVR
di: Lu, Jinda, et al.
Pubblicazione: (2026)

On the Direction of RLVR Updates for LLM Reasoning: Identification and Exploitation
di: Huang, Kexin, et al.
Pubblicazione: (2026)

Bridging Perception and Reasoning: Token Reweighting for RLVR in Multimodal LLMs
di: Lu, Jinda, et al.
Pubblicazione: (2026)

Observational Constraints and Cosmological Dynamics of Interacting Fractional Holographic Dark Energy in Light of DESI DR2
di: Huang, Qihong, et al.
Pubblicazione: (2026)

StructSynth: Leveraging LLMs for Structure-Aware Tabular Data Synthesis in Low-Data Regimes
di: Liu, Siyi, et al.
Pubblicazione: (2025)

Beyond Uniform Credit Assignment: Selective Eligibility Traces for RLVR
di: Mou, Chaoli, et al.
Pubblicazione: (2026)

From Instance Selection to Fixed-Pool Data Recipe Search for Supervised Fine-Tuning
di: Wu, Haodong, et al.
Pubblicazione: (2026)

Can Interlinked Credit and Insurance Contracts Boost Farmers' Adoption of Improved Rice Varieties? Field Experimental Evidence from China
di: Haixia Wu, et al.
Pubblicazione: (2025)

The Obfuscation Atlas: Mapping Where Honesty Emerges in RLVR with Deception Probes
di: Taufeeque, Mohammad, et al.
Pubblicazione: (2026)

Utilizing and Calibrating Hindsight Process Rewards via Reinforcement with Mutual Information Self-Evaluation
di: Yao, Jiashu, et al.
Pubblicazione: (2026)

Orchestrating Tokens and Sequences: Dynamic Hybrid Policy Optimization for RLVR
di: Min, Zijun, et al.
Pubblicazione: (2026)

Stabilizing Knowledge, Promoting Reasoning: Dual-Token Constraints for RLVR
di: Wang, Jiakang, et al.
Pubblicazione: (2025)

Mixture of Length and Pruning Experts for Knowledge Graphs Reasoning
di: Du, Enjun, et al.
Pubblicazione: (2025)

GraphOracle: Efficient Fully-Inductive Knowledge Graph Reasoning via Relation-Dependency Graphs
di: Du, Enjun, et al.
Pubblicazione: (2025)

Credit Where Credit is due: Recommendations for Inclusive and Ethical Authorship Practices in Ecology
di: Kevin C. Elliott, et al.
Pubblicazione: (2026)

Where to Spend Rollouts: Hit-Utility Optimal Rollout Allocation for Group-Based RLVR
di: Wang, Tao, et al.
Pubblicazione: (2026)

Hindsight on High John
di: Moses, Richard
Pubblicazione: (1972)

Quantile Advantage Estimation: Stabilizing RLVR for LLM Reasoning
di: Wu, Junkang, et al.
Pubblicazione: (2025)

Exploring Multi-Temperature Strategies for Token- and Rollout-Level Control in RLVR
di: Zhuang, Haomin, et al.
Pubblicazione: (2025)

Generalization of RLVR Using Causal Reasoning as a Testbed
di: Lu, Brian, et al.
Pubblicazione: (2025)

DelTA: Discriminative Token Credit Assignment for Reinforcement Learning from Verifiable Rewards
di: Zhang, Kaiyi, et al.
Pubblicazione: (2026)

Credit Where Credit Is Due: Considering Ethics, "Ethos", and Process in Library Instruction on Attribution
di: Harris, Benjamin R.
Pubblicazione: (2005)

How Far Can Unsupervised RLVR Scale LLM Training?
di: He, Bingxiang, et al.
Pubblicazione: (2026)

Combinatorial Synthesis: Scaling Code RLVR via Atomic Decomposition and Recombination
di: Zheng, Jiasheng, et al.
Pubblicazione: (2026)

Mobile Cold Energy Storage: Coupling Food Distribution and Energy Systems
di: Lao, Hange, et al.
Pubblicazione: (2026)

Self-Distilled RLVR
di: Yang, Chenxu, et al.
Pubblicazione: (2026)

Tokens All the Way Down: A Money View of Decentralized Finance
di: Wu, Wenbin
Pubblicazione: (2026)

Hindsight-Anchored Policy Optimization: Turning Failure into Feedback in Sparse Reward Settings
di: Wu, Yuning, et al.
Pubblicazione: (2026)

Generalized Capacity Planning for the Hospital-Residents Problem
di: Balasundaram, Haricharan, et al.
Pubblicazione: (2025)

From Hindsight to Foresight: Self-Encouraged Hindsight Distillation for Knowledge-based Visual Question Answering
di: Zhao, Yu, et al.
Pubblicazione: (2025)

Quantifying Empirical Compute-Supervision Tradeoffs in RLVR
di: Mitsuhashi, Ryo, et al.
Pubblicazione: (2026)

RLVR-World: Training World Models with Reinforcement Learning
di: Wu, Jialong, et al.
Pubblicazione: (2025)

White Dwarfs with Infrared Excess from LAMOST Data Release 11
di: Wang, Keyi, et al.
Pubblicazione: (2026)

Leave or Stay? Community Environment Change and Resident Mobility in Tourism Gentrification
di: Zixin Feng, et al.
Pubblicazione: (2025)

Translating Flow to Policy via Hindsight Online Imitation
di: Zheng, Yitian, et al.
Pubblicazione: (2025)

Evaluating Parameter Efficient Methods for RLVR
di: Yin, Qingyu, et al.
Pubblicazione: (2025)

Foreign Residency Rights and Corporate Bond Yield Spreads
di: Zhong‐qin Su, et al.
Pubblicazione: (2025)

Can Central Bank Digital Currencies Promote the Internationalization of Currencies?
di: Haodong Gu
Pubblicazione: (2025)