Papers by Matthew Peters

18 papers
Peek Across: Improving Multi-Document Modeling via Cross-Document Question-Answering (2023.acl-long)

Copied to clipboard

Challenge: Among recent NLP research, multi-document processing is gaining increasing attention due to the need to handle and process an increasing amount of textual data and available documents online.
Approach: They propose to pre-train a generic multi-document model from a cross-document question answering pre-training objective by generating salient sentences from one document and challenging it to recover the sentence from which it was generated.
Outcome: The proposed model outperforms zero-shot GPT-3.5 and GPT-4 in multiple document tasks and generates the correct answer and the salient sentence from a salient document.
CDLM: Cross-Document Language Modeling (2021.findings-emnlp)

Copied to clipboard

Challenge: Existing language models (LMs) provide powerful representations for internal text structure, but there are important applications for multi-text tasks.
Approach: They propose a pretraining approach that incorporates two key ideas into the masked language modeling objective.
Outcome: The proposed model improves over existing models and sets of long-range transformers and can be easily applied to multiple multi-text tasks.
Extending a Parser to Distant Domains Using a Few Dozen Partially Annotated Examples (P18-1)

Copied to clipboard

Challenge: Statistical parsers are often criticized for their performance outside of the domain they were trained on . we show that word representations reduce the need for domain adaptation when the target domain is syntactically similar to the source domain.
Approach: They propose a way to adapt a parser to a syntactically similar target domain using partial annotations.
Outcome: The proposed model increases the accuracy of a parser on the Wall Street Journal by 1.7% over the previous state-of-the-art model.
TESS: Text-to-Text Self-Conditioned Simplex Diffusion (2024.eacl-long)

Copied to clipboard

Challenge: Existing models for diffusion generation are expensive and discrete, resulting in a large number of diffusion steps to generate text.
Approach: They propose a text diffusion model that is fully non-autoregressive and employs a new form of self-conditioning and applies the diffusion process on the logit simplex space rather than the learned embedding space.
Outcome: The proposed model outperforms state-of-the-art non-autoregressive models, requires fewer diffusion steps with minimal drop in performance, and is competitive with pretrained autoregressive sequence-to-sequence models.
Efficient Hierarchical Domain Adaptation for Pretrained Language Models (2022.naacl-main)

Copied to clipboard

Challenge: Existing methods to allow domain adaptation to diverse domains are expensive and require continuing training in-domain.
Approach: They propose a method to permit domain adaptation to many diverse domains using a computationally efficient adapter approach.
Outcome: The proposed method allows domain adaptation to many diverse domains while avoiding negative interference between unrelated domains.
Few-Shot Self-Rationalization with Natural Language Prompts (2022.findings-naacl)

Copied to clipboard

Challenge: Existing models that generate free-text explanations for tasks are limited by human-written explanations.
Approach: They propose to use a standardized collection of natural language prompts to create a model that generates free-text explanations for tasks.
Outcome: The proposed model can predict task labels and generate free-text explanations for predictions . plausibility of human explanations is 76%, while human explanation is 51% .
HINT: Hypernetwork Instruction Tuning for Efficient Zero- and Few-Shot Generalisation (2023.acl-long)

Copied to clipboard

Challenge: Recent NLP models have shown the remarkable ability to generalise ‘zero-shot’ to new tasks using only natural language instructions as guidance.
Approach: They introduce Hypernetworks for INstruction Tuning (HINT) which converts task instructions and examples into parameter-efficient modules inserted into an underlying model using a pretrained text encoder.
Outcome: The proposed models outperform strong state-of-the-art models by over 10% when controlling for compute.
Extracting Latent Steering Vectors from Pretrained Language Models (2022.findings-acl)

Copied to clipboard

Challenge: Prior work on controllable text generation has focused on learning how to control language models through trainable decoding, smart-prompt design, or fine-tuning based on a desired objective.
Approach: They propose to extract latent vectors directly from pretrained language model decoders without fine-tuning.
Outcome: The proposed approach generates a target sentence nearly perfectly for English sentences . it outperforms pooled hidden states of models on a textual similarity benchmark .
Construction of the Literature Graph in Semantic Scholar (N18-3)

Copied to clipboard

Challenge: Fig. 1 summarizes a scalable system for organizing published scientific literature into a heterogeneous graph . authors describe methods used to enable semantic features in www.semanticscholar.org .
Approach: They describe a scalable system for organizing published scientific literature into a heterogeneous graph to facilitate algorithmic manipulation and discovery.
Outcome: The proposed system can be deployed on a scalable platform and report empirical results for each task.
Does Self-Rationalization Improve Robustness to Spurious Correlations? (2022.emnlp-main)

Copied to clipboard

Challenge: Rationalization is fundamental to human reasoning and learning.
Approach: They evaluate robustness to spurious correlations in encoder-decoder and decoder-only models . authors say explanations can come at the cost of robustness .
Outcome: The proposed model outputs are more interpretable and easier to interact with for end-users than nonrationalizing models.
FiD-ICL: A Fusion-in-Decoder Approach for Efficient In-Context Learning (2023.acl-long)

Copied to clipboard

Challenge: Large pre-trained models are capable of few-shot in-context learning (ICL) however, concatenated demonstrations are often excessively long and require additional computation.
Approach: They propose to apply fusion-in-decoder (FiD) models to perform few-shot in-context learning (ICL) they propose to use concatenation-based, early-fusion, intermediate- and late-fusion methods to improve efficiency .
Outcome: The proposed methods outperform concatenation-based models on 11 held-out tasks.
PIGLeT: Language Grounding Through Neuro-Symbolic Interaction in a 3D World (2021.acl-long)

Copied to clipboard

Challenge: PIGLeT model learns physical commonsense knowledge through interaction, then uses this knowledge to ground language.
Approach: They propose a model that learns physical commonsense knowledge through interaction . they factorize PIGLeT into a physical dynamics model and a separate language model .
Outcome: The proposed model outperforms a 100x larger, text-to-text approach in forecasting language . it can read a sentence, simulate neurally what might happen next, and communicate that result through a literal symbolic representation, or natural language.
Hyperdecoders: Instance-specific decoders for multi-task NLP (2022.findings-emnlp)

Copied to clipboard

Challenge: Recent work in NLP has examined the performance of large pretrained transformer-based models in multi-task settings, where a single model is evaluated on multiple tasks simultaneously.
Approach: They propose a method for multi-tasking using a hypernetwork conditioned on the output of an encoder to generate a unique decoder adaptation for every input instance.
Outcome: The proposed method outperforms previous methods for efficient multi-task fine-tuning and maps from encoder representations to output labels.
AdapterSoup: Weight Averaging to Improve Generalization of Pretrained Language Models (2023.findings-eacl)

Copied to clipboard

Challenge: Pretrained language models often need to specialize to specific domains.
Approach: They propose an approach that performs weight-space averaging of adapters trained on different domains.
Outcome: The proposed approach improves performance to new domains without extra training.
Competency Problems: On Finding and Removing Artifacts in Language Data (2021.emnlp-main)

Copied to clipboard

Challenge: Recent work in NLP has documented dataset artifacts, bias, and spurious correlations . how to tell which features have spurious instead of legitimate correlations is typically left unspecified .
Approach: They propose a class of competency problems to formalize this notion into a classification . they show that realistic datasets will increasingly deviate from competency problems .
Outcome: The proposed model can be used to show that models are inappropriately affected by these less extreme biases.
Tailor: Generating and Perturbing Text with Semantic Controls (2022.acl-long)

Copied to clipboard

Challenge: Existing studies train task-specific generators, relying on training a model for every perturbation.
Approach: They propose a semantically-controlled text generation system that modifies sentences to match target attributes.
Outcome: The proposed system produces textual outputs conditioned on control codes derived from semantic representations.
Explaining NLP Models via Minimal Contrastive Editing (MiCE) (2021.findings-acl)

Copied to clipboard

Challenge: Cognitive science and philosophy research has shown that human explanations are contrastive . a contrast case plays a key role in modulating what explanations can be given .
Approach: They propose a method for producing contrastive explanations of model predictions . they edit models' outputs to change model outputs, and then edit them to the contrast case .
Outcome: a new method produces contrastive explanations of model predictions in the form of edits . the edits are minimal and fluent, consistent with human contrastive edits.
ATTEMPT: Parameter-Efficient Multi-task Tuning via Attentional Mixtures of Soft Prompts (2022.emnlp-main)

Copied to clipboard

Challenge: a new multi-task, parameter-efficient language model tuning method learns to transfer knowledge across different tasks via a mixture of soft prompts.
Approach: They propose a multi-task, parameter-efficient language model tuning method that uses soft prompts to learn to transfer knowledge across different tasks.
Outcome: The proposed method outperforms prompt tuning and outperfies or matches fully fine-tuned tuning approaches that use 10 times more parameters.

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