Regularized Best-of-N Sampling with Minimum Bayes Risk Objective for Language Model Alignment (2025.naacl-long)
Copied to clipboard
| Challenge: | Best-of-N (BoN) sampling is an effective strategy for aligning Large Language Models (LLMs) to human preferences at the time of decoding. |
| Approach: | They propose a variant of BoN that incorporates the Minimum Bayes Risk objective as a proximity regularizer for BoN sampling. |
| Outcome: | The proposed method outperforms both BoN sampling and MBR decoding on the AlpacaFarm and Anthropic datasets. |
Similar Papers
Epsilon Sampling Rocks: Investigating Sampling Strategies for Minimum Bayes Risk Decoding for Machine Translation (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Recent advances in machine translation (MT) have shown that minimum bayes risk decoding can be a powerful alternative to beam search. |
| Approach: | They propose to use epsilon-sampling to prune away all tokens with a smaller probability mass. |
| Outcome: | The proposed method outperforms beam search decoding and other methods in four languages. |
Direct Preference Optimization for Neural Machine Translation with Minimum Bayes Risk Decoding (2024.naacl-short)
Copied to clipboard
| Challenge: | Recent work shows that MBR decoding can significantly improve translation performance of Multilingual Large Language Models. |
| Approach: | They propose a method that uses a monolingual fine-tuning set to fine- tune MLLMs to get the gains of MBR without additional computation in inference. |
| Outcome: | The proposed method outperforms greedy decoding and beam search on multiple NMT tests. |
Faster Minimum Bayes Risk Decoding with Confidence-based Pruning (2023.emnlp-main)
Copied to clipboard
| Challenge: | Minimum Bayes risk (MBR) decoding is a decision rule for conditional sequence generation tasks. |
| Approach: | They propose an algorithm which grows the number of samples used to estimate utility . it prunes hypotheses that are unlikely to have the highest utility based on bootstrap sampling . |
| Outcome: | The proposed method outperforms beam search in conditional language generation and neural machine translation tasks while being statistically indistinguishable from other proposed methods. |
Structure-Conditional Minimum Bayes Risk Decoding (2025.emnlp-main)
Copied to clipboard
| Challenge: | Minimum Bayes Risk (MBR) decoding has been used in machine translation for many years. |
| Approach: | They propose three adaptations to the minimum bayes risk utility function to make it more sensitive to structural variability in the outcome space. |
| Outcome: | The proposed adaptations significantly improve generation quality by up to 13.7 percentage points in win rate. |
On the True Distribution Approximation of Minimum Bayes-Risk Decoding (2024.naacl-short)
Copied to clipboard
| Challenge: | Minimum Bayes-risk (MBR) decoding has recently gained renewed attention in text generation. |
| Approach: | They propose to use anomaly detection to measure the degree of approximation by sampling texts from a model and selecting the text with the highest similarity to the others. |
| Outcome: | The proposed method shows that previous hypotheses about samples do not correlate well with the variation, but the results support the core assumption of MBR decoding. |
Hyperparameter-Free Approach for Faster Minimum Bayes Risk Decoding (2024.findings-acl)
Copied to clipboard
| Challenge: | Minimum Bayes-Risk (MBR) decoding is a powerful alternative to beam search for text generation tasks. |
| Approach: | They propose a method that uses a hyperparameter-free method to run MBR decoding efficiently. |
| Outcome: | The proposed method reduces the amount of computation but requires hyperparameter tuning to be effective. |
Efficient Safety Alignment of Large Language Models via Preference Re-ranking and Representation-based Reward Modeling (2025.acl-long)
Copied to clipboard
| Challenge: | Existing safety alignment methods for Large Language Models (LLMs) face the distribution shift issue, which requires significant computational resources. |
| Approach: | They propose a framework that leverages the model’s intrinsic safety judgment capability to extract reward signals, which are then used to calculate label confidence for preference reordering. |
| Outcome: | The proposed framework improves safety performance while avoiding 300x computational overheads. |
Theoretical Guarantees for Minimum Bayes Risk Decoding (2025.acl-long)
Copied to clipboard
| Challenge: | Minimum Bayes Risk (MBR) decoding is a decision rule used to generate sequences from autoregressive probability models (e.g., LLMs). |
| Approach: | They propose to use minimum bayes risk (MBR) decoding to optimize output selection by maximizing expected utility value of an underlying human distribution. |
| Outcome: | The proposed method is effective even though the language space Y is larger than the hypothesis set. |
RS-DPO: A Hybrid Rejection Sampling and Direct Preference Optimization Method for Alignment of Large Language Models (2024.findings-naacl)
Copied to clipboard
| Challenge: | Reinforcement learning with human feedback (RLHF) is widely employed to align large language models with user intent. |
| Approach: | They propose to combine rejection sampling and direct preference optimization to improve alignment with user intent by identifying pairs of contrastive samples from human annotator and alternative LLMs. |
| Outcome: | The proposed method outperforms existing methods including RS, PPO, and DPO in a limited resource environment. |
Unveiling the Power of Source: Source-based Minimum Bayes Risk Decoding for Neural Machine Translation (2025.acl-long)
Copied to clipboard
| Challenge: | Maximum a posteriori decoding aims to maximize the estimated posterior probability, but high estimated probability does not always lead to high translation quality. |
| Approach: | They propose a method that seeks hypotheses with the highest expected utility by using quasi-sources as “support hypothese . they propose sMBR decoding which utilizes a reference-free quality estimation metric as the utility function. |
| Outcome: | The proposed approach outperforms QE reranking and the standard MBR decoding. |