Bharat Runwal

Oliver Sieberling, Bharat Runwal, Rameswar Panda et al.

Transformers have become the dominant architecture for large language models, largely due to the scalability and flexibility of attention, feed-forward layers, residual connections, and normalization. This paper introduces dynamic short convolutions as an additional neural network primitive for improving Transformers. Unlike static short convolutions, dynamic convolutions use input-dependent filters, which preserves the locality bias of convolution while increasing expressivity. Motivating experiments show that applying dynamic short convolutions to key, query, and value representations improves performance on challenging associative recall tasks compared with static convolutional variants. Across language-modeling experiments ranging from 150M to 2B parameters, dynamic convolutions consistently outperform standard Transformers and Transformers augmented with static short convolutions. Fitting scaling laws indicates a 1.33$\times$ compute advantage over compute-matched Transformers when dynamic convolutions are applied to the key, query, and value vectors, and a 1.60$\times$ advantage when adding dynamic convolutions after every linear layer. Dynamic convolutions also offer improvements on linear RNNs (Mamba-2/Gated DeltaNet) and mixture-of-experts architectures. We make these gains practical with custom Triton kernels that enable efficient training with a manageable end-to-end slowdown. These results suggest that dynamic short convolutions are a scalable, hardware-efficient, and expressive primitive for advancing Transformer-based language models.

Diganta Misra, Bharat Runwal, Tianlong Chen et al.

With the latest advances in deep learning, there has been a lot of focus on the online learning paradigm due to its relevance in practical settings. Although many methods have been investigated for optimal learning settings in scenarios where the data stream is continuous over time, sparse networks training in such settings have often been overlooked. In this paper, we explore the problem of training a neural network with a target sparsity in a particular case of online learning: the anytime learning at macroscale paradigm (ALMA). We propose a novel way of progressive pruning, referred to as \textit{Anytime Progressive Pruning} (APP); the proposed approach significantly outperforms the baseline dense and Anytime OSP models across multiple architectures and datasets under short, moderate, and long-sequence training. Our method, for example, shows an improvement in accuracy of $\approx 7\%$ and a reduction in the generalization gap by $\approx 22\%$, while being $\approx 1/3$ rd the size of the dense baseline model in few-shot restricted imagenet training. We further observe interesting nonmonotonic transitions in the generalization gap in the high number of megabatches-based ALMA. The code and experiment dashboards can be accessed at \url{https://github.com/landskape-ai/Progressive-Pruning} and \url{https://wandb.ai/landskape/APP}, respectively.

John Chong Min Tan, Prince Saroj, Bharat Runwal et al.

TaskGen is an open-sourced agentic framework which uses an Agent to solve an arbitrary task by breaking them down into subtasks. Each subtask is mapped to an Equipped Function or another Agent to execute. In order to reduce verbosity (and hence token usage), TaskGen uses StrictJSON that ensures JSON output from the Large Language Model (LLM), along with additional features such as type checking and iterative error correction. Key to the philosophy of TaskGen is the management of information/memory on a need-to-know basis. We empirically evaluate TaskGen on various environments such as 40x40 dynamic maze navigation with changing obstacle locations (100% solve rate), TextWorld escape room solving with dense rewards and detailed goals (96% solve rate), web browsing (69% of actions successful), solving the MATH dataset (71% solve rate over 100 Level-5 problems), Retrieval Augmented Generation on NaturalQuestions dataset (F1 score of 47.03%)

Bharat Runwal, Vivek, Sandeep Kumar

Graph neural network (GNN) is achieving remarkable performances in a variety of application domains. However, GNN is vulnerable to noise and adversarial attacks in input data. Making GNN robust against noises and adversarial attacks is an important problem. The existing defense methods for GNNs are computationally demanding and are not scalable. In this paper, we propose a generic framework for robustifying GNN known as Weighted Laplacian GNN (RWL-GNN). The method combines Weighted Graph Laplacian learning with the GNN implementation. The proposed method benefits from the positive semi-definiteness property of Laplacian matrix, feature smoothness, and latent features via formulating a unified optimization framework, which ensures the adversarial/noisy edges are discarded and connections in the graph are appropriately weighted. For demonstration, the experiments are conducted with Graph convolutional neural network(GCNN) architecture, however, the proposed framework is easily amenable to any existing GNN architecture. The simulation results with benchmark dataset establish the efficacy of the proposed method, both in accuracy and computational efficiency. Code can be accessed at https://github.com/Bharat-Runwal/RWL-GNN.

Jinghan Jia, Yihua Zhang, Yimeng Zhang et al.

Large Language Models (LLMs) have highlighted the necessity of effective unlearning mechanisms to comply with data regulations and ethical AI practices. LLM unlearning aims at removing undesired data influences and associated model capabilities without compromising utility beyond the scope of unlearning. While interest in studying LLM unlearning is growing, the impact of the optimizer choice for LLM unlearning remains unexplored. In this work, we shed light on the significance of optimizer selection in LLM unlearning for the first time, establishing a clear connection between second-order optimization and influence unlearning (a classical approach using influence functions to update the model for data influence removal). This insight propels us to develop a second-order optimization-based LLM unlearning framework, termed Second-Order UnLearning (SOUL), which extends the static, one-shot model update using influence unlearning to a dynamic, iterative unlearning process. Our extensive experiments show that SOUL consistently outperforms conventional first-order methods across various unlearning tasks, models, and metrics, indicating that second-order optimization offers an effective and broadly applicable solution for LLM unlearning. Codes are available at https://github.com/OPTML-Group/SOUL.

Diganta Misra, Muawiz Chaudhary, Agam Goyal et al.

In an age dominated by resource-intensive foundation models, the ability to efficiently adapt to downstream tasks is crucial. Visual Prompting (VP), drawing inspiration from the prompting techniques employed in Large Language Models (LLMs), has emerged as a pivotal method for transfer learning in the realm of computer vision. As the importance of efficiency continues to rise, research into model compression has become indispensable in alleviating the computational burdens associated with training and deploying over-parameterized neural networks. A primary objective in model compression is to develop sparse and/or quantized models capable of matching or even surpassing the performance of their over-parameterized, full-precision counterparts. Although previous studies have explored the effects of model compression on transfer learning, its impact on visual prompting-based transfer remains unclear. This study aims to bridge this gap, shedding light on the fact that model compression detrimentally impacts the performance of visual prompting-based transfer, particularly evident in scenarios with low data volume. Furthermore, our findings underscore the adverse influence of sparsity on the calibration of downstream visual-prompted models. However, intriguingly, we also illustrate that such negative effects on calibration are not present when models are compressed via quantization. This empirical investigation underscores the need for a nuanced understanding beyond mere accuracy in sparse and quantized settings, thereby paving the way for further exploration in Visual Prompting techniques tailored for sparse and quantized models.

Bharat Runwal, Ashish Agrawal, Anurag Roy et al.

We present PRISM, a comprehensive empirical study of mid-training design choices for large language models. Through controlled experiments across seven base models spanning four families (Granite, LLaMA, Mistral, Nemotron-H), two architecture types (dense Transformer and attention-Mamba hybrid), and scales from 3B to 24B parameters, we show that mid-training on approximately 27B high-quality tokens yields consistent gains of +15 to +40 points on math, +5 to +12 points on code, and +6 to +13 points on science benchmarks while preserving general performance. The full PRISM to RL pipeline improves macro-average across six reasoning benchmarks from under 12 to 29-42 (a 3-4x improvement), whereas RL applied directly to most of the base models remains substantially less effective, with AIME scores near zero. Data composition matters most at mid-training, not RL: including science data during mid-training unlocks +17 to +28 point GPQA-Diamond gains during RL, while changing the RL mix produces less than 2 point differences. Mechanistically, mid-training densely restructures over 90% of model weights, while RL makes sparse, front-loaded refinements to approximately 5% of parameters. Representation analysis (CKA) confirms that RL consistently preserves mid-training's representational geometry (over 0.998 CKA) across architectures. Crucially, RL applies identical weight changes regardless of starting point, yet only succeeds on mid-trained models, consistent with mid-training placing the model in a configuration from which RL can effectively improve performance. Our results demonstrate that retention-aware mid-training is highly effective for reliable reasoning enhancement and provide practical guidance for designing robust mid-training pipelines.

Bharat Runwal, Tejaswini Pedapati, Pin-Yu Chen

Pretrained Language Models (PLMs) have become the de facto starting point for fine-tuning on downstream tasks. However, as model sizes continue to increase, traditional fine-tuning of all the parameters becomes challenging. To address this, parameter-efficient fine-tuning (PEFT) methods have gained popularity as a means to adapt PLMs effectively. In parallel, recent studies have revealed the presence of activation sparsity within the intermediate outputs of the multilayer perceptron (MLP) blocks in transformers. Low activation density enables efficient model inference on sparsity-aware hardware. Building upon this insight, in this work, we propose a novel density loss that encourages higher activation sparsity (equivalently, lower activation density) in the pre-trained models. We demonstrate the effectiveness of our approach by utilizing mainstream PEFT techniques, including QLoRA, LoRA, Adapter, and Prompt/Prefix Tuning, to facilitate efficient model adaptation across diverse downstream tasks. Experiments show that our proposed method, \textbf{DEFT} (Density-Efficient Fine-Tuning), can consistently reduce activation density by up to \textbf{44.94\%} on RoBERTa$_\mathrm{Large}$ and by \textbf{53.19\%} (encoder density) and \textbf{90.60\%} (decoder density) on Flan-T5$_\mathrm{XXL}$ (\textbf{11B}) compared to PEFT, using GLUE and QA (SQuAD) benchmarks respectively. We also introduce \textbf{ADA-DEFT}, an adaptive variant of our DEFT approach, which achieves significant memory and runtime savings during inference. For instance, ADA-DEFT reduces runtime by \textbf{8.79\%}and memory usage by \textbf{17.46\%} in Flan-T5$_\mathrm{XL}$, and by \textbf{2.79\%} and \textbf{2.54\%} respectively in Flan-T5$_\mathrm{XXL}$. Additionally, we showcase that DEFT works complementarily with quantized and pruned models.