Challenge: Modern neural networks do not always produce wellcalibrated predictions . post-hoc calibration methods require a held-out calibration dataset, which may not be available in all circumstances.
Approach: They validate ensemble distillation framework for producing well-calibrated structured prediction models without the prohibitive inference-time cost of ensembles.
Outcome: The proposed framework produces well-calibrated predictions without the prohibitive inference-time cost of ensembles.

Similar Papers

Distilling Knowledge for Search-based Structured Prediction (P18-1)

Copied to clipboard

Challenge: Existing studies have focused on the performance of structured prediction models, but they are often limited by the ambiguities of the reference policy.
Approach: They propose to distill an ensemble of multiple models trained with different initializations into a single model and use it to explore the search space.
Outcome: The proposed model outperforms the greedy models on two typical search-based structured prediction tasks and achieves 1.32 in LAS and 2.65 in BLEU over strong baselines.
Calibrating Structured Output Predictors for Natural Language Processing (2020.acl-main)

Copied to clipboard

Challenge: Several modern machine-learning based NLP systems can provide a confidence score with their output predictions.
Approach: They propose a general calibration scheme for output entities of interest in NLP applications that can be used to calibrate confidence scores.
Outcome: The proposed calibration scheme outperforms current calibration techniques for Named Entity Recognition, Part-of-speech tagging and Question Answering systems.
Structural Knowledge Distillation: Tractably Distilling Information for Structured Predictor (2021.acl-long)

Copied to clipboard

Challenge: Knowledge distillation is a technique to transfer knowledge between models, typically from a large model (the teacher) to a more fine-grained one (the student).
Approach: They propose a factorized form of the knowledge distillation objective for structured prediction which is tractable for many typical choices of the teacher and student models.
Outcome: The proposed model is able to transfer knowledge between teacher and student models without loss of accuracy under four different scenarios.
Distillation of encoder-decoder transformers for sequence labelling (2023.findings-eacl)

Copied to clipboard

Challenge: despite the strong trend in NLP to explore the use of large language models, there is still limited work evaluating prompting and decoding mechanisms for SL tasks.
Approach: They propose a hallucination-free framework for sequence tagging that is especially suited for distillation.
Outcome: The proposed framework performs well across multiple sequence labelling datasets and in a few-shot learning scenario.
Generation-Distillation for Efficient Natural Language Understanding in Low-Data Settings (D19-61)

Copied to clipboard

Challenge: Recent research points to knowledge distillation as a potential solution for NLU tasks.
Approach: They propose a training approach that distills large finetuned LMs into a small network using unlabeled training examples.
Outcome: The proposed approach outperforms BERT training approaches while using 300 times fewer parameters.
Multi-Grained Knowledge Distillation for Named Entity Recognition (2021.naacl-main)

Copied to clipboard

Challenge: Pre-trained big models have delivered top performance in Seq2seq modeling, but their deployments in real-world applications are often hindered by excessive computations and memory demands.
Approach: They propose a distillation scheme to efficiently transfer knowledge from big models to their cheaper counterparts.
Outcome: The proposed scheme maximizes the assimilation of knowledge from the teacher model to the student model.
Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback (2023.emnlp-main)

Copied to clipboard

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.
Accurate Knowledge Distillation via n-best Reranking (2024.naacl-long)

Copied to clipboard

Challenge: Existing studies using sequencelevel knowledge distillation (KD) have adopted this approach.
Approach: They propose to utilize n-best reranking to enhance Sequence-Level Knowledge Distillation by utilizing a diverse set of models with different inductive biases, objective functions or architectures to pick the highest-quality hypotheses as labels.
Outcome: The proposed approach is validated on the WMT’21 German English and Chinese english translation tasks.
XtremeDistil: Multi-stage Distillation for Massive Multilingual Models (2020.acl-main)

Copied to clipboard

Challenge: Existing work on pre-trained language models focuses on reducing the size of these models into shallow ones.
Approach: They propose a knowledge distillation technique that leverages teacher internal representations to reduce the size of pre-trained language models.
Outcome: The proposed method outperforms previous methods in multilingual Named Entity Recognition (NER) it reduces the size of teacher models by 35x while retaining 95% of its F1 score.
Calibrating Zero-shot Cross-lingual (Un-)structured Predictions (2022.emnlp-main)

Copied to clipboard

Challenge: Existing need for model calibration when natural language models are deployed in critical tasks.
Approach: They compare model calibration methods in a context of zero-shot cross-lingual transfer with pre-trained language models.
Outcome: The proposed method fails to calibrate more complex confidence estimations in structured predictions compared to expressive alternatives like Gaussian Process Calibration.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations