Papers by Dani Yogatama
On the Cross-lingual Transferability of Monolingual Representations (2020.acl-main)
Copied to clipboard
| Challenge: | State-of-the-art unsupervised multilingual models generalize in zero-shot cross-lingual setting . generalization ability attributed to shared subword vocabulary and joint training across multiple languages . |
| Approach: | They propose an approach that transfers a monolingual model to new languages at the lexical level. |
| Outcome: | The proposed approach is competitive with multilingual BERT on cross-lingual classification benchmarks and on a new cross-linguistic question answering dataset. |
ABC: Attention with Bounded-memory Control (2022.acl-long)
Copied to clipboard
Hao Peng, Jungo Kasai, Nikolaos Pappas, Dani Yogatama, Zhaofeng Wu, Lingpeng Kong, Roy Schwartz, Noah A. Smith
| Challenge: | Existing approaches to attention with bounded-memory control (ABC) have a quadratic complexity in sequence lengths, making it prohibitive for long sequences. |
| Approach: | They propose a new abstraction that bounds memory size to improve efficiency . they propose bounded-memory control, which connects several efficient attention variants . |
| Outcome: | The proposed approach outperforms existing approaches on language modeling, machine translation, and masked language model finetuning. |
Questions Are All You Need to Train a Dense Passage Retriever (2023.tacl-1)
Copied to clipboard
| Challenge: | Existing methods for dense retrieval require large supervised datasets with custom hard-negative mining and denoising of positive examples. |
| Approach: | They propose a new corpus-level autoencoding approach for training dense retrieval models that does not require labeled training data. |
| Outcome: | The proposed method matches or surpasses strong supervised performance levels on multiple QA benchmarks with no labeled training data or task-specific losses. |
Pelican Soup Framework: A Theoretical Framework for Language Model Capabilities (2026.findings-eacl)
Copied to clipboard
| Challenge: | Existing theoretical frameworks for large language models (LLMs) do not explain how pretraining leads to in-context learning. |
| Approach: | They propose a theoretical framework that allows LLMs to generalize to unseen instructions and perform in-context learning even when verbalizers are irrelevant to the task. |
| Outcome: | The proposed framework can be used to analyze LLMs' ability to perform in-context learning . it can be applied to linguistic, psychology, and philosophy tasks . |
FOL-Traces: Verified First-Order Logic Reasoning Traces at Scale (2026.findings-eacl)
Copied to clipboard
| Challenge: | Existing approaches to evaluate language models fail to provide structural clarity and verifiable inference. |
| Approach: | They propose to use a large-scale dataset of programmatically verified reasoning traces to evaluate structured logical inference. |
| Outcome: | The proposed model achieves 45.7% accuracy on masked operation prediction and 27% on two-step completion. |
Scaling Laws vs Model Architectures: How does Inductive Bias Influence Scaling? (2023.findings-emnlp)
Copied to clipboard
Yi Tay, Mostafa Dehghani, Samira Abnar, Hyung Chung, William Fedus, Jinfeng Rao, Sharan Narang, Vinh Tran, Dani Yogatama, Donald Metzler
| Challenge: | Existing studies on the scaling properties of model architectures have not explored the impact of inductive biases on scaling behaviour. |
| Approach: | They conduct extensive experiments to understand scaling behaviour of ten different model architectures. |
| Outcome: | The results show that the best performing model can fluctuate at different scales. |
Relational Memory-Augmented Language Models (2022.tacl-1)
Copied to clipboard
| Challenge: | Existing language models rely on word correlation and are difficult to interpret . existing models often lack explicit representations for such information . |
| Approach: | They propose a memory-augmented approach to condition autoregressive language models on knowledge graphs. |
| Outcome: | The proposed model improves perplexity and bits per character in an autoregressive language model . it is complementary to token-based memory and enables causal interventions . |
The Distributional Hypothesis Does Not Fully Explain the Benefits of Masked Language Model Pretraining (2023.emnlp-main)
Copied to clipboard
| Challenge: | Despite the rise of the prompting paradigm with the scaling breakthrough of very large language models, understanding the mechanism of model fine-tuning remains an important endeavor. |
| Approach: | They analyze the masked language modeling pretraining objective function from the perspective of the Distributional Hypothesis and examine whether the distributional property leads to better sample efficiency and better generalization capability of pretrained models. |
| Outcome: | The proposed model pretraining objective function improves sample efficiency and generalization capability but does not explain the generalization ability of natural language models. |
Achieving Verified Robustness to Symbol Substitutions via Interval Bound Propagation (D19-1)
Copied to clipboard
Po-Sen Huang, Robert Stanforth, Johannes Welbl, Chris Dyer, Dani Yogatama, Sven Gowal, Krishnamurthy Dvijotham, Pushmeet Kohli
| Challenge: | Recent work has exposed the vulnerabilities of neural NLP models, e.g. with small, semantically invariant input alterations. |
| Approach: | They propose to model text classification under synonym replacements or character flip perturbations and then use a formal model verification method to verify its robustness. |
| Outcome: | The proposed models show little difference in terms of nominal accuracy, but have much improved verified accuracy under perturbations and come with an efficiently computable formal guarantee on worst case adversaries. |
Adaptive Semiparametric Language Models (2021.tacl-1)
Copied to clipboard
| Challenge: | Existing language models that use a large parametric neural network with episodic memory are not efficient. |
| Approach: | They propose a language model that combines a large parametric neural network with a non-parametric episodic memory component in an integrated architecture. |
| Outcome: | The proposed model can predict local context, short-term memory, or long-term memories on an ad hoc basis depending on the context. |
Syntactic Structure Distillation Pretraining for Bidirectional Encoders (2020.tacl-1)
Copied to clipboard
Adhiguna Kuncoro, Lingpeng Kong, Daniel Fried, Dani Yogatama, Laura Rimell, Chris Dyer, Phil Blunsom
| Challenge: | Textual representation learners trained on large amounts of data have been successful on downstream tasks. |
| Approach: | They propose a knowledge distillation strategy for injecting syntactic biases into BERT pretraining by distilling the approximate marginal distribution over words in context from the syntaktic LM. |
| Outcome: | The proposed method reduces relative error by 2–21% on a diverse set of structured prediction tasks. |
Finetuning Pretrained Transformers into RNNs (2021.emnlp-main)
Copied to clipboard
Jungo Kasai, Hao Peng, Yizhe Zhang, Dani Yogatama, Gabriel Ilharco, Nikolaos Pappas, Yi Mao, Weizhu Chen, Noah A. Smith
| Challenge: | Efficient transformers outperform recurrent neural networks in natural language generation, but this comes with significant computational cost and memory footprint during generation. |
| Approach: | They propose to convert a pretrained transformer into its efficient recurrent counterpart, improving efficiency while maintaining accuracy. |
| Outcome: | The proposed transformers outperform recurrent neural networks in natural language generation but come with significant computational and memory footprint during generation. |
A Call for More Rigor in Unsupervised Cross-lingual Learning (2020.acl-main)
Copied to clipboard
| Challenge: | Existing research on unsupervised cross-lingual learning has focused on purely unsupervised learning without any parallel data for most of the world's languages. |
| Approach: | They propose to define "multilingual learning" as learning a common model for two or more languages from raw text, without any downstream task labels. |
| Outcome: | The proposed model is based on a model with no parallel data and abundant monolingual data. |
LSTMs Can Learn Syntax-Sensitive Dependencies Well, But Modeling Structure Makes Them Better (P18-1)
Copied to clipboard
| Challenge: | a recent study found that language models fail to learn long-range syntax sensitive dependencies. |
| Approach: | They propose to use a subject-verb agreement diagnostic to determine whether language models can learn long-range syntax sensitive dependencies. |
| Outcome: | The proposed model outperforms left-corner and bottom-up variants in learning non-local dependencies. |
On Retrieval Augmentation and the Limitations of Language Model Training (2024.naacl-short)
Copied to clipboard
| Challenge: | Recent efforts to improve the performance of language models (LMs) have focused on scaling up model and training data size, though with steep accompanying energy and compute resource costs. |
| Approach: | They propose to augment a language model with k-nearest neighbors retrieval on its training data to reduce its perplexity. |
| Outcome: | The proposed model reduces storage costs by over 25x compared to traditional retrieval methods for GPT-2 and Mistral 7B . |
The Rotary Position Embedding May Cause Dimension Inefficiency in Attention Heads for Long-Distance Retrieval (2025.findings-acl)
Copied to clipboard
| Challenge: | We hypothesize that the wide range of rotation angles may prevent LLMs from utilizing certain dimensions. |
| Approach: | They propose to use the Rotary Position Embedding (RoPE) for long context modeling . they hypothesize that the wide range of rotation angles may prevent LLMs from utilizing those dimensions. |
| Outcome: | The proposed model may not be useful for long-context modeling. |
Reducing Sentiment Bias in Language Models via Counterfactual Evaluation (2020.findings-emnlp)
Copied to clipboard
Po-Sen Huang, Huan Zhang, Ray Jiang, Robert Stanforth, Johannes Welbl, Jack Rae, Vishal Maini, Dani Yogatama, Pushmeet Kohli
| Challenge: | Language modeling has advanced rapidly due to efficient model architectures and the availability of large text corpora. |
| Approach: | They propose to embed and regularize sentiment prediction-derived regularizations on the language model’s latent representations to reduce bias in the sentiment of generated text. |
| Outcome: | The proposed methods reduce bias in the sentiment of generated text by adopting individual and group fairness metrics from the fair machine learning literature. |