Mitigating Reward Over-optimization in Direct Alignment Algorithms with Importance Sampling
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | Nguyen, Phuc Minh, Nguyen, Ngoc-Hieu, Nguyen, Duy H. M., Liu, Anji, Mai, An, Nguyen, Binh T., Sonntag, Daniel, Doan, Khoa D. |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
Documents similaires
The Reasoning Boundary Paradox: How Reinforcement Learning Constrains Language Models
par: Nguyen, Phuc Minh, et autres
Publié: (2025)
par: Nguyen, Phuc Minh, et autres
Publié: (2025)
Cold-start Recommendation by Personalized Embedding Region Elicitation
par: Nguyen, Hieu Trung, et autres
Publié: (2024)
par: Nguyen, Hieu Trung, et autres
Publié: (2024)
Environmental taxes and the economy: New evidence of the shadow economy
par: Canh Phuc Nguyen, et autres
Publié: (2026)
par: Canh Phuc Nguyen, et autres
Publié: (2026)
Why Do Reasoning Models Lose Coverage? The Role of Data and Forks in the Road
par: Nguyen, Ngoc-Hieu, et autres
Publié: (2026)
par: Nguyen, Ngoc-Hieu, et autres
Publié: (2026)
Applications of optimal error bounds for some generalized two-step iterative processes in Banach spaces
par: Nguyen, Tan-Phuc, et autres
Publié: (2025)
par: Nguyen, Tan-Phuc, et autres
Publié: (2025)
A new approach to convergence analysis of iterative models with optimal error bounds
par: Tran, Minh-Phuong, et autres
Publié: (2024)
par: Tran, Minh-Phuong, et autres
Publié: (2024)
A von Neumann-Jordan Constant of Non-Normable Metrics
par: Hieu, Doan Huu, et autres
Publié: (2026)
par: Hieu, Doan Huu, et autres
Publié: (2026)
Reinforcing Trustworthiness in Multimodal Emotional Support Systems
par: Le, Huy M., et autres
Publié: (2025)
par: Le, Huy M., et autres
Publié: (2025)
EFL teachers’ perceptions of professional development activities and their effects in a non-anglosphere context
par: Duy Binh Nguyen
Publié: (2022)
par: Duy Binh Nguyen
Publié: (2022)
Metric constructions and fixed point theorems in product spaces
par: Hieu, Doan Huu, et autres
Publié: (2026)
par: Hieu, Doan Huu, et autres
Publié: (2026)
BSO: Safety Alignment Is Density Ratio Matching
par: Nguyen, Tien-Phat, et autres
Publié: (2026)
par: Nguyen, Tien-Phat, et autres
Publié: (2026)
Risk of Osteoporosis and Degraded Trabecular Bone Score in Rheumatoid Arthritis Patients
par: Huong Nguyen, et autres
Publié: (2025)
par: Huong Nguyen, et autres
Publié: (2025)
Pauli nonlocality and the nucleon effective mass
par: Khoa, Dao T., et autres
Publié: (2024)
par: Khoa, Dao T., et autres
Publié: (2024)
HFedATM: Hierarchical Federated Domain Generalization via Optimal Transport and Regularized Mean Aggregation
par: Nguyen, Thinh, et autres
Publié: (2025)
par: Nguyen, Thinh, et autres
Publié: (2025)
Distributional Surgery for Language Model Activations
par: Nguyen, Bao, et autres
Publié: (2025)
par: Nguyen, Bao, et autres
Publié: (2025)
Disturbance observer based control for passive multi‐actuator systems with aggregation and distribution
par: Binh Minh Nguyen
Publié: (2024)
par: Binh Minh Nguyen
Publié: (2024)
TokenRatio: Principled Token-Level Preference Optimization via Ratio Matching
par: Nguyen, Truong, et autres
Publié: (2026)
par: Nguyen, Truong, et autres
Publié: (2026)
Nuclear Rainbow of Core-Symmetric Systems
par: Phuc, Nguyen Tri Toan, et autres
Publié: (2026)
par: Phuc, Nguyen Tri Toan, et autres
Publié: (2026)
Nuclear rainbow of the symmetric nucleus-nucleus system: Interchange of the nearside and farside scattering
par: Phuc, Nguyen Tri Toan, et autres
Publié: (2024)
par: Phuc, Nguyen Tri Toan, et autres
Publié: (2024)
Extremality of families of sets and set-valued optimization
par: Cuong, Nguyen Duy, et autres
Publié: (2024)
par: Cuong, Nguyen Duy, et autres
Publié: (2024)
Generative Conditional Distributions by Neural (Entropic) Optimal Transport
par: Nguyen, Bao, et autres
Publié: (2024)
par: Nguyen, Bao, et autres
Publié: (2024)
Task-driven Layerwise Additive Activation Intervention
par: Nguyen, Hieu Trung, et autres
Publié: (2025)
par: Nguyen, Hieu Trung, et autres
Publié: (2025)
Bridging the divide between technology and pedagogy: What does a bibliometric analysis reveal about the future of inducing flow in e‐learning?
par: Nguyen Binh Phuong Duy
Publié: (2026)
par: Nguyen Binh Phuong Duy
Publié: (2026)
Robust Aggregation for Federated Sequential Recommendation with Sparse and Poisoned Data
par: Nguyen, Minh Hieu
Publié: (2026)
par: Nguyen, Minh Hieu
Publié: (2026)
OpenEvents V1: Large-Scale Benchmark Dataset for Multimodal Event Grounding
par: Nguyen, Hieu, et autres
Publié: (2025)
par: Nguyen, Hieu, et autres
Publié: (2025)
Wicked Oddities: Selectively Poisoning for Effective Clean-Label Backdoor Attacks
par: Nguyen, Quang H., et autres
Publié: (2024)
par: Nguyen, Quang H., et autres
Publié: (2024)
Overcoming Catastrophic Forgetting in Federated Class-Incremental Learning via Federated Global Twin Generator
par: Nguyen, Thinh, et autres
Publié: (2024)
par: Nguyen, Thinh, et autres
Publié: (2024)
Retrospective Feature Estimation for Continual Learning
par: Nguyen, Nghia D., et autres
Publié: (2024)
par: Nguyen, Nghia D., et autres
Publié: (2024)
Evaluating the Prognostic Accuracy of New Scores for In‐Hospital Outcomes in Cirrhotic Patients With Esophageal Variceal Bleeding
par: Khoa Phuoc Nguyen, et autres
Publié: (2026)
par: Khoa Phuoc Nguyen, et autres
Publié: (2026)
Are you SURE? Enhancing Multimodal Pretraining with Missing Modalities through Uncertainty Estimation
par: Nguyen, Duy A., et autres
Publié: (2025)
par: Nguyen, Duy A., et autres
Publié: (2025)
EventCap
par: Nguyen, Phuc-Tan, et autres
Publié: (2024)
par: Nguyen, Phuc-Tan, et autres
Publié: (2024)
Revisiting Incremental Stochastic Majorization-Minimization Algorithms with Applications to Mixture of Experts
par: Tran, TrungKhang, et autres
Publié: (2026)
par: Tran, TrungKhang, et autres
Publié: (2026)
Deep-Wide Learning Assistance for Insect Pest Classification
par: Nguyen, Toan, et autres
Publié: (2024)
par: Nguyen, Toan, et autres
Publié: (2024)
MIC: Maximizing Informational Capacity in Adaptive Representations via Isotropic Subspace Alignment
par: Hong, Dang Nguyen, et autres
Publié: (2026)
par: Hong, Dang Nguyen, et autres
Publié: (2026)
ViSpeechFormer: A Phonemic Approach for Vietnamese Automatic Speech Recognition
par: Nguyen, Khoa Anh, et autres
Publié: (2026)
par: Nguyen, Khoa Anh, et autres
Publié: (2026)
On Zero-Initialized Attention: Optimal Prompt and Gating Factor Estimation
par: Diep, Nghiem T., et autres
Publié: (2025)
par: Diep, Nghiem T., et autres
Publié: (2025)
Enhancing DC Microgrids Stability by Integrating DAB Converters with Consensus Algorithms for Bus Voltage Drop Mitigation
par: Minh Duc Pham, et autres
Publié: (2024)
par: Minh Duc Pham, et autres
Publié: (2024)
Energy-Aware Resource Allocation for Energy Harvesting Powered Wireless Sensor Nodes
par: Ngo, Ngoc M., et autres
Publié: (2025)
par: Ngo, Ngoc M., et autres
Publié: (2025)
Coverage-Validity-Aware Algorithmic Recourse
par: Bui, Ngoc, et autres
Publié: (2023)
par: Bui, Ngoc, et autres
Publié: (2023)
A new benzophenone from the bark of Garcinia lanessanii
par: Hieu T. Nguyen, et autres
Publié: (2025)
par: Hieu T. Nguyen, et autres
Publié: (2025)
Documents similaires
-
The Reasoning Boundary Paradox: How Reinforcement Learning Constrains Language Models
par: Nguyen, Phuc Minh, et autres
Publié: (2025) -
Cold-start Recommendation by Personalized Embedding Region Elicitation
par: Nguyen, Hieu Trung, et autres
Publié: (2024) -
Environmental taxes and the economy: New evidence of the shadow economy
par: Canh Phuc Nguyen, et autres
Publié: (2026) -
Why Do Reasoning Models Lose Coverage? The Role of Data and Forks in the Road
par: Nguyen, Ngoc-Hieu, et autres
Publié: (2026) -
Applications of optimal error bounds for some generalized two-step iterative processes in Banach spaces
par: Nguyen, Tan-Phuc, et autres
Publié: (2025)