Papers by Canasai Kruengkrai

10 papers
Teaching Text Agents to Learn Sequential Decision Making from Failure (2025.acl-long)

Copied to clipboard

Challenge: Existing methods to improve text-based reinforcement-learning agents' performance contain failed actions that reinforce incorrect behaviors and reduce task success rates.
Approach: They propose a failed action-aware objective that suppresses negative impact of failed actions . they propose 'failed action-based' perturbation method that leverages unsuccessful trajectories to construct new successful ones .
Outcome: The proposed method outperforms baselines and generalizes across environments.
Improving Low-Resource Named Entity Recognition using Joint Sentence and Token Labeling (2020.acl-main)

Copied to clipboard

Challenge: Existing models for named entity recognition (NER) use sentence-level labels, which are expensive to obtain, to improve NER.
Approach: They propose a sentence-level named entity recognition model that uses sentence-based labels that are easy to obtain.
Outcome: The proposed model produces 3.78%, 4.20%, 2.08% improvements in F1 over the baseline on e-commerce product titles in Vietnamese, Thai, and Indonesian, respectively.
A Multi-Level Attention Model for Evidence-Based Fact Checking (2021.findings-acl)

Copied to clipboard

Challenge: Recent state-of-the-art approaches have developed increasingly sophisticated models based on graph structures.
Approach: They propose a simple model that can be trained on sequence structures and can benefit from joint training.
Outcome: The proposed model outperforms the graph-based models on a large-scale dataset for Fact Extraction and VERification.
Revisiting Pathologies of Neural Models under Input Reduction (2023.findings-acl)

Copied to clipboard

Challenge: Recent studies have shown that modern neural models tend to be miscalibrated.
Approach: They examine why models produce high-confidence predictions on inputs that appear nonsensical to humans . previous work suggested that models fail to assign low probabilities due to model overconfidence .
Outcome: The proposed methods can be extended to reduce the number of examples but with the cost of miscalibration.
DAGA: Data Augmentation with a Generation Approach for Low-resource Tagging Tasks (2020.emnlp-main)

Copied to clipboard

Challenge: Data augmentation techniques are widely used to improve machine learning performance . however, due to the complexity of language, it is difficult to generalize such rules for languages.
Approach: They propose a method to generate high quality synthetic data for low-resource tagging tasks . they use unlabeled data only and unlabelled data plus a knowledge base .
Outcome: The proposed method outperforms baselines on NER, part of speech and target based sentiment analysis tasks.
Better Exploiting Latent Variables in Text Modeling (P19-1)

Copied to clipboard

Challenge: Consistent gains in performance on two datasets, Penn Treebank and Yahoo, indicate the generalizability of our method.
Approach: They propose a method to exploit latent variables through hidden state averaging by sampling latent variable multiple times at a gradient step.
Outcome: The proposed method shows consistent gains on two datasets showing that it is generalizable.
Mitigating the Diminishing Effect of Elastic Weight Consolidation (2022.coling-1)

Copied to clipboard

Challenge: Existing work addresses catastrophic forgetting in sequential training by fine-tuning pre-trained language models on different datasets.
Approach: They propose to rescale the components of EWC to mitigate catastrophic forgetting by mixing new and old training data and retraining the model from scratch.
Outcome: The proposed method requires smaller values for the trade-off parameters to achieve comparable results to EWC on natural language inference and fact-checking tasks.
Learning to Flip the Sentiment of Reviews from Non-Parallel Corpora (D19-1)

Copied to clipboard

Challenge: Existing methods for flipping sentiment are costly and require parallel data.
Approach: They propose a method for acquiring imperfectly aligned sentences from non-parallel corpora and propose 'sensational' model that learns to minimize sentiment and content losses in a fully end-to-end manner.
Outcome: The proposed model offers well-balanced results across Yelp restaurant and Amazon product reviews.
Outlier-Aware Training for Improving Group Accuracy Disparities (2022.aacl-srw)

Copied to clipboard

Challenge: Methods addressing spurious correlations such as Just Train Twice involve reweighting a subset of the training set to maximize the worst-group accuracy.
Approach: They propose to reweight a subset of a training set to maximize the worst-group accuracy by detecting outliers and removing them before reweighing.
Outcome: The proposed method achieves competitive or better accuracy compared with JTT and can detect and remove annotation errors in the subset being reweighted in JTT.
Bridging Textual and Tabular Worlds for Fact Verification: A Lightweight, Attention-Based Model (2024.lrec-main)

Copied to clipboard

Challenge: Existing fact-checking systems rely on extensive preprocessing and rule-based transformations, leading to potential context loss or misleading encodings.
Approach: They propose a simple yet powerful model that nullifies the need for modality conversion, thereby preserving the original evidence’s context.
Outcome: The proposed model nullifies the need for modality conversion, preserving the original evidence’s context.

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