Large Language Models (LLMs) are increasingly being deployed on edge devices for long-context settings, creating a growing need for fast and efficient long-context inference. In these scenarios, the Key-Value (KV) cache is the primary bottleneck in terms of both GPU memory and latency, as the full KV cache must be loaded for each decoding step. While speculative decoding is a widely accepted technique to accelerate autoregressive decoding, existing methods often struggle to achieve significant speedups due to inefficient KV cache optimization strategies and result in low acceptance rates. To address these challenges, we propose a novel self-speculative decoding framework, QuantSpec, where the draft model shares the architecture of the target model but employs a hierarchical 4-bit quantized KV cache and 4-bit quantized weights for acceleration. QuantSpec maintains high acceptance rates ($>$90\%) and reliably provides consistent end-to-end speedups upto $\sim2.5\times$, outperforming other self-speculative decoding methods that use sparse KV cache for long-context LLM inference. QuantSpec also reduces the memory requirements by $\sim 1.3\times$ compared to these alternatives.

Despite enthusiasm for Multi-Agent LLM Systems (MAS), their performance gains on popular benchmarks are often minimal. This gap highlights a critical need for a principled understanding of why MAS fail. Addressing this question requires systematic identification and analysis of failure patterns. We introduce MAST-Data, a comprehensive dataset of 1600+ annotated traces collected across 7 popular MAS frameworks. MAST-Data is the first multi-agent system dataset to outline the failure dynamics in MAS for guiding the development of better future systems. To enable systematic classification of failures for MAST-Data, we build the first Multi-Agent System Failure Taxonomy (MAST). We develop MAST through rigorous analysis of 150 traces, guided closely by expert human annotators andvalidated by high inter-annotator agreement (κ = 0.88). This process identifies 14 unique modes, clustered into 3 categories: (i) system design issues, (ii) inter-agent misalignment, and (iii) task verification. To enable scalable annotation, we develop an LLM-as-a-Judge pipeline with high agreement with human annotations. We leverage MAST and MAST-Data to analyze failure patterns across models (GPT4, Claude 3, Qwen2.5, CodeLlama) and tasks (coding, math, general agent), demonstrating improvement headrooms from better MAS design. Our analysis provides insights revealing that identified failures require more sophisticated solutions, highlighting a clear roadmap for future research. We publicly release our comprehensive dataset (MAST-Data), the MAST, and our LLM annotator to facilitate widespread research and development in MAS.

Deep networks tend to learn spurious feature-label correlations in real-world supervised learning tasks. This vulnerability is aggravated in distillation, where a student model may have lesser representational capacity than the corresponding teacher model. Often, knowledge of specific spurious correlations is used to reweight instances & rebalance the learning process. We propose a novel early readout mechanism whereby we attempt to predict the label using representations from earlier network layers. We show that these early readouts automatically identify problem instances or groups in the form of confident, incorrect predictions. Leveraging these signals to modulate the distillation loss on an instance level allows us to substantially improve not only group fairness measures across benchmark datasets, but also overall accuracy of the student model. We also provide secondary analyses that bring insight into the role of feature learning in supervision and distillation.

Concept bottleneck models (CBMs) are interpretable neural networks that first predict labels for human-interpretable concepts relevant to the prediction task, and then predict the final label based on the concept label predictions. We extend CBMs to interactive prediction settings where the model can query a human collaborator for the label to some concepts. We develop an interaction policy that, at prediction time, chooses which concepts to request a label for so as to maximally improve the final prediction. We demonstrate that a simple policy combining concept prediction uncertainty and influence of the concept on the final prediction achieves strong performance and outperforms static approaches as well as active feature acquisition methods proposed in the literature. We show that the interactive CBM can achieve accuracy gains of 5-10% with only 5 interactions over competitive baselines on the Caltech-UCSD Birds, CheXpert and OAI datasets.

Object trackers deployed on low-power devices need to be light-weight, however, most of the current state-of-the-art (SOTA) methods rely on using compute-heavy backbones built using CNNs or Transformers. Large sizes of such models do not allow their deployment in low-power conditions and designing compressed variants of large tracking models is of great importance. This paper demonstrates how highly compressed light-weight object trackers can be designed using neural architectural pruning of large CNN and Transformer based trackers. Further, a comparative study on architectural choices best suited to design lightweight trackers is provided. A comparison between SOTA trackers using CNNs, Transformers as well as the combination of the two is presented to study their stability at various compression ratios. Finally results for extreme pruning scenarios going as low as 1% in some cases are shown to study the limits of network pruning in object tracking. This work provides deeper insights into designing highly efficient trackers from existing SOTA methods.1

Simplicity bias is the concerning tendency of deep networks to over-depend on simple, weakly predictive features, to the exclusion of stronger, more complex features. This causes biased, incorrect model predictions in many real-world applications, exacerbated by incomplete training data containing spurious feature-label correlations. We propose a direct, interventional method for addressing simplicity bias in DNNs, which we call the *feature sieve*. We aim to automatically identify and suppress easily-computable spurious features in lower layers of the network, thereby allowing the higher network levels to extract and utilize richer, more meaningful representations. We provide concrete evidence of this differential suppression & enhancement of *relevant* features on both controlled datasets and real-world images, and report substantial gains on many real-world debiasing benchmarks (11.4% relative gain on Imagenet-A; 3.2% on BAR, etc). Crucially, we outperform many baselines that incorporate knowledge about known spurious or biased attributes, despite our method not using any such information. We believe that our *feature sieve* work opens up exciting new research directions in automated adversarial feature extraction & representation learning for deep networks.

Abstract Interpreting high‐resolution satellite imagery could be an expensive and time‐consuming task for human eyes. Computer Vision and Deep Learning techniques can help to solve this major problem by applying detection algorithms, which can ease the task of analysing such images for the benefit of humans. It can help in changing the way we comprehend and anticipate the economic activity around the world. Such techniques help us to observe the urban development in high security areas such as national and international borders. Constant progressions in improving and making satellites deployment, a cost‐effective process to strengthen the networks of satellite orbiting the earth is one of the reasons such tasks can be easily solved with the help of high‐resolution images. Current computer vision research works have achieved significant milestones in accuracy and speed but, there are still room for improvements. In this paper, we addressed some of these methods to bring them to a combined pipeline and proposed a set of improvements to further improve the speed and the accuracy of the detections. We proposed a unified framework, which combines several object detection algorithms and the state‐of‐art architecture of YoloV4 along with the TensorFlow object detection API. This framework can detect small and well as large objects with improved speed and accuracy by using two detectors for different scales. Evaluation ran on these high‐resolution images yield mAP of 85.6% F1‐score of 0.84.

Gradient based meta-learning methods are prone to overfit on the meta-training set, and this behaviour is more prominent with large and complex networks. Moreover, large networks restrict the application of meta-learning models on low-power edge devices. While choosing smaller networks avoid these issues to a certain extent, it affects the overall generalization leading to reduced performance. Clearly, there is an approximately optimal choice of network architecture that is best suited for every meta-learning problem, however, identifying it beforehand is not straight-forward. In this paper, we present Metadock, a task-specific dynamic kernel selection strategy for designing compressed CNN models that generalize well on unseen tasks in meta-learning. Our method is based on the hypothesis that for a given set of similar tasks, not all kernels of the network are needed by each individual task. Rather, each task uses only a fraction of the kernels, and the selection of the kernels per task can be learnt dynamically as a part of the inner update steps. Metadockcompresses the meta-model as well as the task-specific inner models, thus providing significant reduction in model size for each task, and through constraining the number of active kernels for every task, it implicitly mitigates the issue of meta-overfitting. We show that for the same inference budget, pruned versions of large CNN models obtained using our approach consistently outperform the conventional choices of CNN models. Metadock couples well with popular meta-learning approaches such as iMAML [22]. The efficacy of our method is validated on CIFAR-fs [1] and mini-ImageNet [28] datasets, and we have observed that our approach can provide improvements in model accuracy of up to 2% on standard meta-learning benchmark, while reducing the model size by more than 75%. Our code is available at https://github.com/transmuteAI/MetaDOCK.

Continual learning (CL) aims to develop techniques by which a single model adapts to an increasing number of tasks encountered sequentially, thereby potentially leveraging learnings across tasks in a resource-efficient manner. A major challenge for CL systems is catastrophic forgetting, where earlier tasks are forgotten while learning a new task. To address this, replay-based CL approaches maintain and repeatedly retrain on a small buffer of data selected across encountered tasks. We propose Gradient Coreset Replay (GCR), a novel strategy for replay buffer selection and update using a carefully designed optimization criterion. Specifically, we select and maintain a ‘coreset’ that closely approximates the gradient of all the data seen so far with respect to current model parameters, and discuss key strategies needed for its effective application to the continual learning setting. We show significant gains (2%-4% absolute) over the state-of-the-art in the well-studied offline continual learning setting. Our findings also effectively transfer to online / streaming CL settings, showing up to 5% gains over existing approaches. Finally, we demonstrate the value of supervised contrastive loss for continual learning, which yields a cumulative gain of up to 5% accuracy when combined with our subset selection strategy.

Deeper and wider CNNs are known to provide improved performance for deep learning tasks. However, most such networks have poor performance gain per parameter increase. In this paper, we investigate whether the gain observed in deeper models is purely due to the addition of more optimization parameters or whether the physical size of the network as well plays a role. Further, we present a novel rescaling strategy for CNNs based on learnable repetition of its parameters. Based on this strategy, we rescale CNNs without changing their parameter count, and show that learnable sharing of weights itself can provide significant boost in the performance of any given model without changing its parameter count. We show that small base networks when rescaled, can provide performance comparable to deeper networks with as low as 6% of optimization parameters of the deeper one. The relevance of weight sharing is further highlighted through the example of group-equivariant CNNs. We show that the significant improvements obtained with group-equivariant CNNs over the regular CNNs on classification problems are only partly due to the added equivariance property, and part of it comes from the learnable repetition of network weights. For rot-MNIST dataset, we show that up to 40% of the relative gain reported by state-of-the-art methods for rotation equivariance could actually be due to just the learnt repetition of weights.11Code is publicly available at https://github.com/transmuteAI/RepeatNet