Are Generative Models Underconfident? Better Quality Estimation with Boosted Model Probability (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have shown that text-generation models can be overconfident when there are multiple correct options. |
| Approach: | They propose a QE approach called BoostedProb which boosts the model’s confidence in cases where there are multiple viable output options. |
| Outcome: | The proposed approach achieves on average +0.194 improvement in Pearson correlation to ground-truth quality and outperforms more costly approaches like supervised or ensemble-based QE in certain settings. |
Similar Papers
Improving the Calibration of Confidence Scores in Text Generation Using the Output Distribution’s Characteristics (2025.acl-short)
Copied to clipboard
| Challenge: | Existing methods for estimating confidence in text generation do not account for many valid answers in generation tasks. |
| Approach: | They propose task-agnostic confidence metrics that rely solely on model probabilities without the need for further fine-tuning or heuristics. |
| Outcome: | The proposed models improve the accuracy of BART and Flan-T5 on summarization, translation, and question answering datasets. |
Classification-based Quality Estimation: Small and Efficient Models for Real-world Applications (2021.emnlp-main)
Copied to clipboard
| Challenge: | Sentence-level Quality estimation (QE) is traditionally a regression task . but large multilingual contextualized language models are expensive and infeasible for real-world applications. |
| Approach: | They evaluate several model compression techniques for QE and find they are inefficient . they argue that a full model parameterization is required to achieve SoTA results . |
| Outcome: | The proposed models are poorly expressive in a regression task, the authors argue . they show that reframing QE as a classification problem and evaluating models would improve their performance in real-world applications. |
Supervised Optimism Correction: Be Confident When LLMs Are Sure (2025.findings-acl)
Copied to clipboard
| Challenge: | Recent advances in Large Language Models (LLMs) have demonstrated remarkable success across diverse tasks such as instruction following, code generation, and medical diagnosis. |
| Approach: | They propose a supervised fine-tuning-based auxiliary loss for Q-value estimations during supervised refinement. |
| Outcome: | The proposed method outperforms beam search on GSM8K, MATH, and GAOKAO on reasoning benchmarks. |
Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback (2023.emnlp-main)
Copied to clipboard
Katherine Tian, Eric Mitchell, Allan Zhou, Archit Sharma, Rafael Rafailov, Huaxiu Yao, Chelsea Finn, Christopher Manning
| Challenge: | Recent studies have shown that unsupervised pre-training produces large language models whose conditional probabilities are remarkably well-calibrated. |
| Approach: | They propose to use verbalized confidences to extract confidence from large language models with reinforcement learning from human feedback to improve their accuracy. |
| Outcome: | The proposed methods reduce the expected calibration error by 50% for RLHF-LMs such as ChatGPT, GPT-4, and Claude. |
Unsupervised Quality Estimation for Neural Machine Translation (2020.tacl-1)
Copied to clipboard
Marina Fomicheva, Shuo Sun, Lisa Yankovskaya, Frédéric Blain, Francisco Guzmán, Mark Fishel, Nikolaos Aletras, Vishrav Chaudhary, Lucia Specia
| Challenge: | Existing approaches require large amounts of expert annotated data, computation, and time for training. |
| Approach: | They propose an unsupervised approach to QE where no training is required . they use a dataset that enables work on both black-box and glass-box approaches . |
| Outcome: | The proposed approach rivals state-of-the-art supervised QE models in terms of correlation with human judgments of quality. |
Are we Estimating or Guesstimating Translation Quality? (2020.acl-main)
Copied to clipboard
| Challenge: | A carefully engineered ensemble of pre-trained multilingual language models won the QE shared task at WMT19. |
| Approach: | They propose to use pre-trained multilingual language models to train quality estimation for machine translation. |
| Outcome: | A carefully engineered ensemble of pre-trained language models wins the QE shared task at WMT19. |
Alleviating Distribution Shift in Synthetic Data for Machine Translation Quality Estimation (2025.acl-long)
Copied to clipboard
| Challenge: | Quality Estimation (QE) models evaluate the quality of machine translations without reference translations, serving as the reward models for the translation task. |
| Approach: | They propose a framework for alleviating distribution shift in synthetic QE data . they employ a constrained beam search algorithm and distinct generation models to enhance translation diversity. |
| Outcome: | The proposed framework outperforms SOTA baselines like CometKiwi in supervised and unsupervised settings. |
Knowledge Distillation for Quality Estimation (2021.findings-acl)
Copied to clipboard
Amit Gajbhiye, Marina Fomicheva, Fernando Alva-Manchego, Frédéric Blain, Abiola Obamuyide, Nikolaos Aletras, Lucia Specia
| Challenge: | Recent success in Quality Estimation stems from the use of multilingual pre-trained models, where large models lead to impressive results. |
| Approach: | They propose to transfer knowledge from a strong QE teacher model to a much smaller model with a different, shallower architecture. |
| Outcome: | The proposed model performs better than distilled models with 8x fewer parameters. |
Improving Machine Translation with Human Feedback: An Exploration of Quality Estimation as a Reward Model (2024.naacl-long)
Copied to clipboard
| Challenge: | Existing methods to improve translation quality using human feedback have not been validated. |
| Approach: | They propose to use quality estimation to predict human preferences for feedback training . they propose to detect incorrect translations and assign a penalty term to the reward scores . |
| Outcome: | The proposed method outperforms systems using larger parallel corpora by a small amount of monolingual data. |
Assessing Quality Estimation Models for Sentence-Level Prediction (C18-1)
Copied to clipboard
| Challenge: | Using a relevant QE model is also very important in QE. |
| Approach: | They evaluate a wide range of advanced sentence-level Quality Estimation models including Support Vector Regression, Ride Regression and Bayesian Neural Networks. |
| Outcome: | The proposed models behave differently in evaluation settings depending on whether test data come from the same domain as the training data or not. |