Neeraj Gangwar, Rishabh Deshmukh, Michael Shavlovsky, Hancao Li, Vivek Mittal, Lexing Ying, Nickvash Kani
As model sizes continue to grow, parameter-efficient fine-tuning has emerged as a powerful alternative to full fine-tuning. While LoRA is widely adopted among these methods, recent research has explored vector-based adaptation methods due to their extreme parameter efficiency. However, these methods typically require substantially higher ranks than LoRA to match its performance, leading to increased training costs. This work introduces GiVA, a gradient-based initialization strategy for vector-based adaptation. It achieves training times comparable to LoRA and maintains the extreme parameter efficiency of vector-based adaptation. We evaluate GiVA across diverse benchmarks, including natural language understanding, natural language generation, and image classification. Experiments show that our approach consistently outperforms or achieves performance competitive with existing vector-based adaptation methods and LoRA while reducing rank requirements by a factor of eight ($8\times$).
Neeraj Gangwar, Nickvash Kani
Operations research deals with modeling and solving real-world problems as mathematical optimization problems. While solving mathematical systems is accomplished by analytical software, formulating a problem as a set of mathematical operations has been typically done manually by domain experts. Recent machine learning methods have shown promise in converting textual problem descriptions to corresponding mathematical formulations. This paper presents an approach that converts linear programming word problems into mathematical formulations. We leverage the named entities in the input and augment the input to highlight these entities. Our approach achieves the highest accuracy among all submissions to the NL4Opt Competition, securing first place in the generation track.
Neeraj Gangwar, Suma P Bhat, Nickvash Kani
While large models pre-trained on high-quality data exhibit excellent performance on mathematical reasoning (e.g., GSM8k, MultiArith), it remains challenging to specialize smaller models for these tasks. Common approaches to address this challenge include knowledge distillation from large teacher models and data augmentation (e.g., rephrasing questions and generating synthetic solutions). Despite these efforts, smaller models struggle with arithmetic computations, leading to errors in mathematical reasoning. In this work, we leverage a synthetic arithmetic dataset generated programmatically to enhance the reasoning capabilities of smaller models. We investigate two key approaches to incorporate this dataset: (1) intermediate fine-tuning, in which a model is fine-tuned on the arithmetic dataset before training it on a reasoning dataset, and (2) integrating the arithmetic dataset into an instruction-tuning mixture, allowing the model to learn arithmetic skills alongside general instruction-following abilities. Our experiments on multiple reasoning benchmarks demonstrate that incorporating an arithmetic dataset, whether through targeted fine-tuning or within an instruction-tuning mixture, enhances models' arithmetic capabilities, thereby improving their mathematical reasoning performance.
Neeraj Gangwar, Nickvash Kani
Mathematical notation makes up a large portion of STEM literature, yet finding semantic representations for formulae remains a challenging problem. Because mathematical notation is precise, and its meaning changes significantly with small character shifts, the methods that work for natural text do not necessarily work well for mathematical expressions. This work describes an approach for representing mathematical expressions in a continuous vector space. We use the encoder of a sequence-to-sequence architecture, trained on visually different but mathematically equivalent expressions, to generate vector representations (or embeddings). We compare this approach with a structural approach that considers visual layout to embed an expression and show that our proposed approach is better at capturing mathematical semantics. Finally, to expedite future research, we publish a corpus of equivalent transcendental and algebraic expression pairs.
Neeraj Gangwar, Anshuka Rangi, Rishabh Deshmukh, Holakou Rahmanian, Yesh Dattatreya, Nickvash Kani
Parameter-efficient fine-tuning methods have emerged as a promising solution for adapting pre-trained models to various downstream tasks. While these methods perform well in single-task learning, extending them to multi-task learning exacerbates common challenges, such as task interference and negative transfer, due to the limited number of trainable parameters. To address these issues, we introduce progressive task-specific multi-task adaptation, a novel parameter-efficient approach for multi-task learning. This approach introduces adapter modules in a pre-trained model such that these modules are shared across all tasks in the initial layers and become progressively more task-specific in the later layers. The motivation is to reduce the conflicts among tasks by allowing transfer learning across all tasks in the initial layers and enabling task-specific learning toward the prediction heads. Additionally, we propose a gradient-based approach for computing task similarity and use this measure to allocate similar tasks to the shared adapter modules. Our task similarity method introduces minimal overhead in the pipeline. We evaluate our approach by adapting the Swin Transformer for dense prediction tasks. Experiments on the PASCAL and NYUD-v2 datasets demonstrate that our approach outperforms a fully fine-tuned multi-task model while requiring only one-fifth of the trainable parameters. This approach achieves better relative improvement to single-task fine-tuning while reducing the number of trainable parameters and surpasses the current state-of-the-art methods for parameter-efficient multi-task learning.
Hongbo Zheng, Suyuan Wang, Neeraj Gangwar, Nickvash Kani
Vector representations have been pivotal in advancing natural language processing (NLP), with prior research focusing on embedding techniques for mathematical expressions using mathematically equivalent formulations. While effective, these approaches are constrained by the size and diversity of training data. In this work, we address these limitations by introducing E-Gen, a novel e-graph-based dataset generation scheme that synthesizes large and diverse mathematical expression datasets, surpassing prior methods in size and operator variety. Leveraging this dataset, we train embedding models using two strategies: (1) generating mathematically equivalent expressions, and (2) contrastive learning to explicitly group equivalent expressions. We evaluate these embeddings on both in-distribution and out-of-distribution mathematical language processing tasks, comparing them against prior methods. Finally, we demonstrate that our embedding-based approach outperforms state-of-the-art large language models (LLMs) on several tasks, underscoring the necessity of optimizing embedding methods for the mathematical data modality. The source code and datasets are available at https://github.com/MLPgroup/E-Gen.