A Natural Bias for Language Generation Models (2023.acl-short)

Copied to clipboard

Challenge: a standard probabilistic model for language generation has likely not yet learnt many semantic or syntactic rules of natural language, making it difficult to estimate the probability distribution over next tokens.
Approach: They propose to initialise bias terms in a model's final linear layer with the log-unigram distribution and use it to output the unigram frequency statistics as prior knowledge.
Outcome: The proposed method improves learning efficiency and improves overall performance.

Similar Papers

Do Neural Language Models Overcome Reporting Bias? (2020.coling-main)

Copied to clipboard

Challenge: Recent studies show that pre-trained language models can overcome reporting bias by estimating the plausibility of rare but unspoken facts.
Approach: They revisit the experiments conducted by Gordon and Van Durme (2013) . they find that pre-trained language models overestimate the very rare .
Outcome: The proposed approach overestimates the rare at the expense of the rare, while minimizing reporting bias.
End-to-End Bias Mitigation by Modelling Biases in Corpora (2020.acl-main)

Copied to clipboard

Challenge: Recent studies have shown that strong natural language understanding models are prone to relying on unwanted dataset biases without learning the underlying task.
Approach: They propose two learning strategies to train neural models that are more robust to dataset biases and transfer better to out-of-domain datasets.
Outcome: The proposed methods improve robustness in all settings and transfer better to out-of-domain datasets.
Predictive Biases in Natural Language Processing Models: A Conceptual Framework and Overview (2020.acl-main)

Copied to clipboard

Challenge: a growing number of studies address the effect of bias on predictions, but no unifying framework exists . a general phenomenon of biased predictive models in NLP is not recent, authors say .
Approach: They propose a unifying framework for identifying and reducing bias in natural language processing . they propose to differentiate two consequences of bias and four potential origins of bias .
Outcome: The proposed framework provides an overview of predictive bias in natural language processing . it differentiates two consequences of bias and four potential origins of bias: label bias, selection bias, model overamplification, and semantic bias.
Quantifying Bias from Decoding Techniques in Natural Language Generation (2022.coling-1)

Copied to clipboard

Challenge: Natural language generation (NLG) models can propagate social bias towards particular demography.
Approach: They propose to examine whether bias metrics like toxicity and sentiment are impacted by decoding techniques that use stochastic decoding.
Outcome: The proposed methods reveal the imperative of testing inference time bias and provide evidence on the usefulness of inspecting the entire decoding spectrum.
Overestimation of Syntactic Representation in Neural Language Models (2020.acl-main)

Copied to clipboard

Challenge: Several testing methodologies have been developed to probe models’ syntactic representations.
Approach: They propose a method to determine syntactic structure by training a model on strings generated according to a template and testing its ability to distinguish between similar ones with different syntax.
Outcome: The proposed method reproduces positive results with two non-syntactic baseline language models: an n-gram model and an LSTM model trained on scrambled inputs.
Towards Zero-shot Language Modeling (D19-1)

Copied to clipboard

Challenge: a number of natural questions have been asked about the inductive biases of neural networks on core NLP tasks.
Approach: They construct an informative prior for held-out languages on a task of character-level, open-vocabulary language modelling.
Outcome: The proposed model outperforms baseline models with an uninformative prior in both zero-shot and few-shot settings, showing that it is imbued with universal linguistic knowledge.
Avoiding the Hypothesis-Only Bias in Natural Language Inference via Ensemble Adversarial Training (2020.emnlp-main)

Copied to clipboard

Challenge: Neural models pick up on annotation artefacts and spurious correlations, resulting in learning sentences that suffer from the same biases.
Approach: They propose to tackle this problem by using adversarial training to reduce the bias in sentence representations by using an ensemble of adversaries.
Outcome: The proposed approach produces more robust models outperforming previous de-biasing efforts when generalised to 12 other NLI datasets.
MiLe Loss: a New Loss for Mitigating the Bias of Learning Difficulties in Generative Language Models (2024.findings-naacl)

Copied to clipboard

Challenge: Existing generative language models neglect an inherent challenge in text corpus during training, i.e., the imbalance between frequent tokens and infrequent ones.
Approach: They propose a function to mitigate the imbalance between frequent and infrequent tokens . authors propose 'MiLe Loss' function to assess learning difficulty of tokens during training .
Outcome: Experiments show that models with proposed model can improve on downstream benchmarks.
End-to-End Self-Debiasing Framework for Robust NLU Training (2021.findings-acl)

Copied to clipboard

Challenge: Existing models incorporate dataset biases leading to strong performance on in-distribution test sets but poor performance on out-of-distortion (OOD) tests.
Approach: They propose a debiasing framework where the shallow representations of the main model are used to derive a bias model and both models are trained simultaneously.
Outcome: The proposed framework outperforms existing approaches on three well-studied NLU tasks while still delivering high in-distribution performance.
Transparency at the Source: Evaluating and Interpreting Language Models With Access to the True Distribution (2023.findings-emnlp)

Copied to clipboard

Challenge: a new approach to train, evaluate and interpret neural language models uses artificial, language-like data.
Approach: They propose a setup for training, evaluating and interpreting neural language models that uses artificial, language-like data.
Outcome: The proposed model is based on a massive probabilistic grammar and a large natural language corpus, and provides complete control over the generative process.

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