Copertina

When Is Compositional Reasoning Learnable from Verifiable Rewards?

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Barzilai, Daniel, Wolf, Yotam, Basri, Ronen
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Controlling the Inductive Bias of Wide Neural Networks by Modifying the Kernel's Spectrum
di: Geifman, Amnon, et al.
Pubblicazione: (2023)

Querying Kernel Methods Suffices for Reconstructing their Training Data
di: Barzilai, Daniel, et al.
Pubblicazione: (2025)

When Models Don't Collapse: On the Consistency of Iterative MLE
di: Barzilai, Daniel, et al.
Pubblicazione: (2025)

Likelihood Training of Cascaded Diffusion Models via Hierarchical Volume-preserving Maps
di: Li, Henry, et al.
Pubblicazione: (2025)

Reward Hacking Mitigation using Verifiable Composite Rewards
di: Tarek, Mirza Farhan Bin, et al.
Pubblicazione: (2025)

Limitations of SGD for Multi-Index Models Beyond Statistical Queries
di: Barzilai, Daniel, et al.
Pubblicazione: (2026)

SpectralNet: Spectral Clustering using Deep Neural Networks
di: Shaham, Uri, et al.
Pubblicazione: (2018)

Generalization in Kernel Regression Under Realistic Assumptions
di: Barzilai, Daniel, et al.
Pubblicazione: (2023)

Simple Relative Deviation Bounds for Covariance and Gram Matrices
di: Barzilai, Daniel, et al.
Pubblicazione: (2024)

Random Policy Valuation is Enough for LLM Reasoning with Verifiable Rewards
di: He, Haoran, et al.
Pubblicazione: (2025)

Verifying Meta-Awareness via Predictive Rewards in Reasoning Models
di: Kim, Yoonjeon, et al.
Pubblicazione: (2025)

Rewarding Progress: Scaling Automated Process Verifiers for LLM Reasoning
di: Setlur, Amrith, et al.
Pubblicazione: (2024)

Fully Learnable Neural Reward Machines
di: Dewidar, Hazem, et al.
Pubblicazione: (2025)

Milestones over Outcome: Unlocking Geometric Reasoning with Sub-Goal Verifiable Reward
di: Chen, Jianlong, et al.
Pubblicazione: (2026)

Prediction-sharing During Training and Inference
di: Gafni, Yotam, et al.
Pubblicazione: (2024)

When Self-Belief Misleads: Active Label Acquisition for Reinforcement Learning with Verifiable Rewards
di: Wang, Li, et al.
Pubblicazione: (2026)

When Sharpening Becomes Collapse: Sampling Bias and Semantic Coupling in RL with Verifiable Rewards
di: Fan, Mingyuan, et al.
Pubblicazione: (2026)

Online Learnability of Chain-of-Thought Verifiers: Soundness and Completeness Trade-offs
di: Balcan, Maria-Florina, et al.
Pubblicazione: (2026)

REASONING GYM: Reasoning Environments for Reinforcement Learning with Verifiable Rewards
di: Stojanovski, Zafir, et al.
Pubblicazione: (2025)

Identifying and Evaluating Inactive Heads in Pretrained LLMs
di: Sandoval-Segura, Pedro, et al.
Pubblicazione: (2025)

Noisy Data is Destructive to Reinforcement Learning with Verifiable Rewards
di: Zhu, Yuxuan, et al.
Pubblicazione: (2026)

Decoupling Reasoning and Confidence: Resurrecting Calibration in Reinforcement Learning from Verifiable Rewards
di: Ma, Zhengzhao, et al.
Pubblicazione: (2026)

Beyond Benign Overfitting in Nadaraya-Watson Interpolators
di: Barzilai, Daniel, et al.
Pubblicazione: (2025)

Chart-RVR: Reinforcement Learning with Verifiable Rewards for Explainable Chart Reasoning
di: Sinha, Sanchit, et al.
Pubblicazione: (2025)

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

RLVMR: Reinforcement Learning with Verifiable Meta-Reasoning Rewards for Robust Long-Horizon Agents
di: Zhang, Zijing, et al.
Pubblicazione: (2025)

Reinforcement Learning with Verifiable yet Noisy Rewards under Imperfect Verifiers
di: Cai, Xin-Qiang, et al.
Pubblicazione: (2025)

Asymmetric Prompt Weighting for Reinforcement Learning with Verifiable Rewards
di: Heckel, Reinhard, et al.
Pubblicazione: (2026)

Generative Verifiers: Reward Modeling as Next-Token Prediction
di: Zhang, Lunjun, et al.
Pubblicazione: (2024)

When To Solve, When To Verify: Compute-Optimal Problem Solving and Generative Verification for LLM Reasoning
di: Singhi, Nishad, et al.
Pubblicazione: (2025)

STRIDE: Learnable Stepwise Language Feedback for LLM Reasoning
di: Zhang, Junjie, et al.
Pubblicazione: (2026)

Shrinking the Variance: Shrinkage Baselines for Reinforcement Learning with Verifiable Rewards
di: Zeng, Guanning, et al.
Pubblicazione: (2025)

Learning to Reason at the Frontier of Learnability
di: Foster, Thomas, et al.
Pubblicazione: (2025)

CinePile: A Long Video Question Answering Dataset and Benchmark
di: Rawal, Ruchit, et al.
Pubblicazione: (2024)

Minerva: Reinforcement Learning with Verifiable Rewards for Cyber Threat Intelligence LLMs
di: Alam, Md Tanvirul, et al.
Pubblicazione: (2026)

An Imperfect Verifier is Good Enough: Learning with Noisy Rewards
di: Plesner, Andreas, et al.
Pubblicazione: (2026)

Contextual Rollout Bandits for Reinforcement Learning with Verifiable Rewards
di: Lu, Xiaodong, et al.
Pubblicazione: (2026)

A Relative-Budget Theory for Reinforcement Learning with Verifiable Rewards in Large Language Model Reasoning
di: Wachi, Akifumi, et al.
Pubblicazione: (2026)

Revisiting Reinforcement Learning with Verifiable Rewards from a Contrastive Perspective
di: Zhang, Feng, et al.
Pubblicazione: (2026)

Are Convex Optimization Curves Convex?
di: Barzilai, Guy, et al.
Pubblicazione: (2025)