Language models (LMs) can produce texts that appear accurate and coherent but contain untruthful or toxic content. Inference-time interventions that edit the hidden activations have shown promising results in steering the LMs towards desirable generations. Existing activation intervention methods often comprise an activation probe to detect undesirable generation, triggering the activation modification to steer subsequent generation. This paper proposes a probe-free intervention method FLORAIN for all attention heads in a specific activation layer. It eliminates the need to train classifiers for probing purposes. The intervention function is parametrized by a sample-wise nonlinear low-rank mapping, which is trained by minimizing the distance between the modified activations and their projection onto the manifold of desirable content. Under specific constructions of the manifold and projection distance, we show that the intervention strategy can be computed efficiently by solving a smooth optimization problem. The empirical results, benchmarked on multiple base models, demonstrate that FLORAIN consistently outperforms several baseline methods in enhancing model truthfulness and quality across generation and multiple-choice tasks. Our implementation can be found at https://github.com/nguyenngocbaocmt02/EFI.

Modern language models (LMs) have significantly advanced generative modeling in natural language processing (NLP). Despite their success, LMs often struggle with adaptation to new contexts in real-time applications. A promising approach to task adaptation is activation intervention, which steers the LMs’ generation process by identifying and manipulating the activations. However, existing interventions rely heavily on heuristic rules or require many prompt inputs to determine effective interventions. In this paper, we propose a layer-wise additive activation intervention framework that optimizes the intervention process, thereby enhancing sample efficiency. We evaluate our framework on various datasets, demonstrating improvements in the accuracy of pretrained LMs and competing intervention baselines.

The SemEval 2021 task 5: Toxic Spans Detection is a task of identifying considered-toxic spans in text, which provides a valuable, automatic tool for moderating online contents. This paper represents the second-place method for the task, an ensemble of two approaches. While one approach relies on combining different embedding methods to extract diverse semantic and syntactic representations of words in context; the other utilizes extra data with a slightly customized Self-training, a semi-supervised learning technique, for sequence tagging problems. Both of our architectures take advantage of a strong language model, which was fine-tuned on a toxic classification task. Although experimental evidence indicates higher effectiveness of the first approach than the second one, combining them leads to our best results of 70.77 F1-score on the test dataset.

This paper proposes an improved custom model for WNUT task 2: Identification of Informative COVID-19 English Tweet. We improve experiment with the effectiveness of fine-tuning methodologies for state-of-the-art language model RoBERTa. We make a preliminary instantiation of this formal model for the text classification approaches. With appropriate training techniques, our model is able to achieve 0.9218 F1-score on public validation set and the ensemble version settles at top 9 F1-score (0.9005) and top 2 Recall (0.9301) on private test set.