Related papers: PaSe: An Extensible and Inspectable DSL for Micro-…

One of the main challenges in distributed computing is building interfaces and APIs that allow programmers with limited background in distributed systems to write scalable, performant, and fault-tolerant applications on large clusters. In…

Sparse autoencoders (SAEs) trained on large language model activations output thousands of features that enable mapping to human-interpretable concepts. The current practice for analyzing these features primarily relies on inspecting…

Controllable human image animation aims to generate videos from reference images using driving videos. Due to the limited control signals provided by sparse guidance (e.g., skeleton pose), recent works have attempted to introduce additional…

Work on knowledge graphs and graph-based data management often focus either on declarative graph query languages or on frameworks for graph analytics, where there has been little work in trying to combine both approaches. However, many…

We propose InsNet, an expressive insertion-based text generator with efficient training and flexible decoding (parallel or sequential). Unlike most existing insertion-based text generation works that require re-encoding of the context after…

Significant progress has been made in scene understanding which seeks to build 3D, metric and object-oriented representations of the world. Concurrently, reinforcement learning has made impressive strides largely enabled by advances in…

Writing a platform for reactive applications which enforces operational constraints is difficult, and has been approached in various ways. In this experience report, we detail an approach using an embedded DSL which can be used to specify…

Motivated by the hypothesis that neural network representations encode abstract, interpretable features as linearly accessible, approximately orthogonal directions, sparse autoencoders (SAEs) have become a popular tool in interpretability.…

A popular new method in mechanistic interpretability is to train high-dimensional sparse autoencoders (SAEs) on neuron activations and use SAE features as the atomic units of analysis. However, the body of evidence on whether SAE feature…

Large Language Models (LLMs) encode factual knowledge within hidden parametric spaces that are difficult to inspect or control. While Sparse Autoencoders (SAEs) can decompose hidden activations into more fine-grained, interpretable…

Unsupervised segmentation approaches have increasingly leveraged foundation models (FM) to improve salient object discovery. However, these methods often falter in scenes with complex, multi-component morphologies, where fine-grained…

Artists and video game designers often construct 2D animations using libraries of sprites -- textured patches of objects and characters. We propose a deep learning approach that decomposes sprite-based video animations into a disentangled…

Extending large language models (LLMs) to process longer inputs is crucial for a wide range of applications. However, the substantial computational cost of transformers and limited generalization of positional encoding restrict the size of…

Symbolic regression plays a crucial role in modern scientific research thanks to its capability of discovering concise and interpretable mathematical expressions from data. A key challenge lies in the search for parsimonious and…

Self-adaptive systems (SAS) can reconfigure at run time in response to changing situations to express acceptable behaviors in the face of uncertainty. With respect to game design, such situations may include user input, emergent behaviors,…

High-level synthesis (HLS) performs well for simple memory access patterns, such as for sequential accesses that can be turned into bursts, or for memory accesses into small datasets that can be stored in scratchpads. This limits HLS to…

Federated learning (FedL) has emerged as a popular technique for distributing model training over a set of wireless devices, via iterative local updates (at devices) and global aggregations (at the server). In this paper, we develop…

To keep up with increasing dataset sizes and model complexity, distributed training has become a necessity for large machine learning tasks. Parameter servers ease the implementation of distributed parameter management---a key concern in…

In sequence to sequence learning, the self-attention mechanism proves to be highly effective, and achieves significant improvements in many tasks. However, the self-attention mechanism is not without its own flaws. Although self-attention…

We introduce PULSE, a sub-microsecond optical circuit-switched data centre network architecture controlled by distributed hardware schedulers. PULSE is a flat architecture that uses parallel passive coupler-based broadcast and select…