Copertina

Improving Reward Models with Synthetic Critiques

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Ye, Zihuiwen, Greenlee-Scott, Fraser, Bartolo, Max, Blunsom, Phil, Campos, Jon Ander, Gallé, Matthias
Natura:	Preprint
Pubblicazione:	2024
Soggetti:	Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Uncertainty-Aware Step-wise Verification with Generative Reward Models
di: Ye, Zihuiwen, et al.
Pubblicazione: (2025)

Human Feedback is not Gold Standard
di: Hosking, Tom, et al.
Pubblicazione: (2023)

Reverse Engineering Human Preferences with Reinforcement Learning
di: Alazraki, Lisa, et al.
Pubblicazione: (2025)

No Need for Explanations: LLMs can implicitly learn from mistakes in-context
di: Alazraki, Lisa, et al.
Pubblicazione: (2025)

LLMCRIT: Teaching Large Language Models to Use Criteria
di: Yuan, Weizhe, et al.
Pubblicazione: (2024)

Aya 23: Open Weight Releases to Further Multilingual Progress
di: Aryabumi, Viraat, et al.
Pubblicazione: (2024)

Fishing for Magikarp: Automatically Detecting Under-trained Tokens in Large Language Models
di: Land, Sander, et al.
Pubblicazione: (2024)

When to Retrieve: Teaching LLMs to Utilize Information Retrieval Effectively
di: Labruna, Tiziano, et al.
Pubblicazione: (2024)

Rope to Nope and Back Again: A New Hybrid Attention Strategy
di: Yang, Bowen, et al.
Pubblicazione: (2025)

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

On the Trade-off between Redundancy and Local Coherence in Summarization
di: Cardenas, Ronald, et al.
Pubblicazione: (2022)

West-of-N: Synthetic Preferences for Self-Improving Reward Models
di: Pace, Alizée, et al.
Pubblicazione: (2024)

Improving Model Factuality with Fine-grained Critique-based Evaluator
di: Xie, Yiqing, et al.
Pubblicazione: (2024)

`Keep it Together': Enforcing Cohesion in Extractive Summaries by Simulating Human Memory
di: Cardenas, Ronald, et al.
Pubblicazione: (2024)

Beyond Sparse Rewards: Enhancing Reinforcement Learning with Language Model Critique in Text Generation
di: Cao, Meng, et al.
Pubblicazione: (2024)

Critique-Coder: Enhancing Coder Models by Critique Reinforcement Learning
di: Ruan, Chi, et al.
Pubblicazione: (2025)

Uncertainty Quantification for LLM Function-Calling
di: Ye, Zihuiwen, et al.
Pubblicazione: (2026)

KARMA: Karma-Aligned Reward Model Adaptation
di: Scott, Jared, et al.
Pubblicazione: (2026)

The Critique of Critique
di: Sun, Shichao, et al.
Pubblicazione: (2024)

Colombian Waitresses y Jueces canadienses: Gender and Country Biases in Occupation Recommendations from LLMs
di: Rodríguez, Elisa Forcada, et al.
Pubblicazione: (2025)

LM2: Large Memory Models
di: Kang, Jikun, et al.
Pubblicazione: (2025)

Reward Models Can Improve Themselves: Reward-Guided Adversarial Failure Mode Discovery for Robust Reward Modeling
di: Pathmanathan, Pankayaraj, et al.
Pubblicazione: (2025)

Distilled Self-Critique of LLMs with Synthetic Data: a Bayesian Perspective
di: Gallego, Victor
Pubblicazione: (2023)

Critique-RL: Training Language Models for Critiquing through Two-Stage Reinforcement Learning
di: Xi, Zhiheng, et al.
Pubblicazione: (2025)

Process Supervision via Verbal Critique Improves Reasoning in Large Language Models
di: Chen, Hao-Yuan
Pubblicazione: (2026)

Boosting Reward Model with Preference-Conditional Multi-Aspect Synthetic Data Generation
di: Shen, Jiaming, et al.
Pubblicazione: (2024)

Merging Improves Self-Critique Against Jailbreak Attacks
di: Gallego, Victor
Pubblicazione: (2024)

Synthetic Eggs in Many Baskets: The Impact of Synthetic Data Diversity on LLM Fine-Tuning
di: Schaffelder, Max, et al.
Pubblicazione: (2025)

Libra: Assessing and Improving Reward Model by Learning to Think
di: Zhou, Meng, et al.
Pubblicazione: (2025)

Agents Explore but Agents Ignore: LLMs Lack Environmental Curiosity
di: Engländer, Leon, et al.
Pubblicazione: (2026)

Training Language Model to Critique for Better Refinement
di: Yu, Tianshu, et al.
Pubblicazione: (2025)

OpenRubrics: Towards Scalable Synthetic Rubric Generation for Reward Modeling and LLM Alignment
di: Liu, Tianci, et al.
Pubblicazione: (2025)

CritiqueLLM: Towards an Informative Critique Generation Model for Evaluation of Large Language Model Generation
di: Ke, Pei, et al.
Pubblicazione: (2023)

Training Language Models with Language Feedback at Scale
di: Scheurer, Jérémy, et al.
Pubblicazione: (2023)

ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline
di: Xu, Yifan, et al.
Pubblicazione: (2024)

Improving Code Generation by Training with Natural Language Feedback
di: Chen, Angelica, et al.
Pubblicazione: (2023)

Grounding Spatial Relations in Text-Only Language Models
di: Azkune, Gorka, et al.
Pubblicazione: (2024)

Dancing with Critiques: Enhancing LLM Reasoning with Stepwise Natural Language Self-Critique
di: Li, Yansi, et al.
Pubblicazione: (2025)

Critique Fine-Tuning: Learning to Critique is More Effective than Learning to Imitate
di: Wang, Yubo, et al.
Pubblicazione: (2025)

Training Language Models to Critique With Multi-agent Feedback
di: Lan, Tian, et al.
Pubblicazione: (2024)