Copertina

Critique-out-Loud Reward Models

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Ankner, Zachary, Paul, Mansheej, Cui, Brandon, Chang, Jonathan D., Ammanabrolu, Prithviraj
Natura:	Preprint
Pubblicazione:	2024
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Simultaneous Multi-objective Alignment Across Verifiable and Non-verifiable Rewards
di: Shen, Yiran, et al.
Pubblicazione: (2025)

Perplexed by Perplexity: Perplexity-Based Data Pruning With Small Reference Models
di: Ankner, Zachary, et al.
Pubblicazione: (2024)

Beyond Needle(s) in the Embodied Haystack: Environment, Architecture, and Training Considerations for Long Context Reasoning
di: Kim, Bosung, et al.
Pubblicazione: (2025)

A Practitioner's Guide to Multi-turn Agentic Reinforcement Learning
di: Wang, Ruiyi, et al.
Pubblicazione: (2025)

How Reasoning Evolves from Post-Training Data: An Empirical Study Using Chess
di: Dionisopoulos, Lucas, et al.
Pubblicazione: (2026)

Preference-Based Learning in Audio Applications: A Systematic Analysis
di: Broukhim, Aaron, et al.
Pubblicazione: (2025)

Scaling Laws for Precision
di: Kumar, Tanishq, et al.
Pubblicazione: (2024)

Hydra: Sequentially-Dependent Draft Heads for Medusa Decoding
di: Ankner, Zachary, et al.
Pubblicazione: (2024)

Introspective X Training: Feedback Conditioning Improves Scaling Across all LLM Training Stages
di: Cui, Brandon, et al.
Pubblicazione: (2026)

$μ$nit Scaling: Simple and Scalable FP8 LLM Training
di: Narayan, Saaketh, et al.
Pubblicazione: (2025)

Does your data spark joy? Performance gains from domain upsampling at the end of training
di: Blakeney, Cody, et al.
Pubblicazione: (2024)

Soup to go: mitigating forgetting during continual learning with model averaging
di: Kleiman, Anat, et al.
Pubblicazione: (2025)

Self-Generated Critiques Boost Reward Modeling for Language Models
di: Yu, Yue, et al.
Pubblicazione: (2024)

In-context Ranking Preference Optimization
di: Wu, Junda, et al.
Pubblicazione: (2025)

Silent Tokens, Loud Effects: Padding in LLMs
di: Himelstein, Rom, et al.
Pubblicazione: (2025)

MASS-DPO: Multi-negative Active Sample Selection for Direct Policy Optimization
di: Surana, Rohan, et al.
Pubblicazione: (2026)

Decoding the Critique Mechanism in Large Reasoning Models
di: Phan, Hoang, et al.
Pubblicazione: (2026)

Behavior Cue Reasoning: Monitorable Reasoning Improves Efficiency and Safety through Oversight
di: Cui, Christopher Z., et al.
Pubblicazione: (2026)

BaNEL: Exploration Posteriors for Generative Modeling Using Only Negative Rewards
di: Lee, Sangyun, et al.
Pubblicazione: (2025)

Helping or Herding? Reward Model Ensembles Mitigate but do not Eliminate Reward Hacking
di: Eisenstein, Jacob, et al.
Pubblicazione: (2023)

RL for Consistency Models: Faster Reward Guided Text-to-Image Generation
di: Oertell, Owen, et al.
Pubblicazione: (2024)

Provably Sample-Efficient Robust Reinforcement Learning with Average Reward
di: Roch, Zachary, et al.
Pubblicazione: (2025)

LoRA Training Provably Converges to a Low-Rank Global Minimum or It Fails Loudly (But it Probably Won't Fail)
di: Kim, Junsu, et al.
Pubblicazione: (2025)

Policy Learning from Large Vision-Language Model Feedback without Reward Modeling
di: Luu, Tung M., et al.
Pubblicazione: (2025)

Noise Injection Systemically Degrades Large Language Model Safety Guardrails
di: Shahani, Prithviraj Singh, et al.
Pubblicazione: (2025)

Dreaming Out Loud: A Self-Synthesis Approach For Training Vision-Language Models With Developmentally Plausible Data
di: AlKhamissi, Badr, et al.
Pubblicazione: (2024)

Prioritize the Process, Not Just the Outcome: Rewarding Latent Thought Trajectories Improves Reasoning in Looped Language Models
di: Williams, Jonathan, et al.
Pubblicazione: (2026)

RewardBench: Evaluating Reward Models for Language Modeling
di: Lambert, Nathan, et al.
Pubblicazione: (2024)

Explanation through Reward Model Reconciliation using POMDP Tree Search
di: Kraske, Benjamin D., et al.
Pubblicazione: (2023)

Reward Generation via Large Vision-Language Model in Offline Reinforcement Learning
di: Lee, Younghwan, et al.
Pubblicazione: (2025)

Reward Models Are Secretly Value Functions: Temporally Coherent Reward Modeling
di: Nikulkov, Alex
Pubblicazione: (2026)

Noise Contrastive Alignment of Language Models with Explicit Rewards
di: Chen, Huayu, et al.
Pubblicazione: (2024)

EvilGenie: A Reward Hacking Benchmark
di: Gabor, Jonathan, et al.
Pubblicazione: (2025)

Generative Reward Models
di: Mahan, Dakota, et al.
Pubblicazione: (2024)

Uncertainty-aware Reward Model: Teaching Reward Models to Know What is Unknown
di: Lou, Xingzhou, et al.
Pubblicazione: (2024)

Bandit Simulation for Average Reward Inference
di: Praharaj, Samya, et al.
Pubblicazione: (2026)

Critiques of World Models
di: Xing, Eric, et al.
Pubblicazione: (2025)

Activation Reward Models for Few-Shot Model Alignment
di: Chai, Tianning, et al.
Pubblicazione: (2025)

Reward Under Attack: Analyzing the Robustness and Hackability of Process Reward Models
di: Tiwari, Rishabh, et al.
Pubblicazione: (2026)

Learning to Keep a Promise: Scaling Language Model Decoding Parallelism with Learned Asynchronous Decoding
di: Jin, Tian, et al.
Pubblicazione: (2025)