Learn what’s next for AI at Research Forum on Sept. 3; WizardArena simulates human-annotated chatbot games; MInference speeds pre-filling for long-context LLMs via dynamic sparse attention; Reef: Fast succinct non-interactive zero-knowledge regex proofs.
As large language models (LLMs) continue to advance, aligning these models with human preferences has emerged as a critical challenge. Traditional alignment methods, relying on human or LLM annotated datasets, are limited by their resource-intensive nature, inherent subjectivity, and the risk of feedback loops that…
This extended report presents DDS, a novel disaggregated storage architecture enabled by emerging networking hardware, namely DPUs (Data Processing Units). DPUs can optimize the latency and CPU consumption of disaggregated storage servers. However, utilizing DPUs for DBMSs requires careful design of the network and storage…
Differentiable causal discovery has made significant advancements in the learning of directed acyclic graphs. However, its application to real-world datasets remains restricted due to the ubiquity of latent confounders and the requirement to learn maximal ancestral graphs (MAGs). To date, existing differentiable MAG learning algorithms…
Large language models (LLMs) have empowered intelligent agents to execute intricate tasks within domain-specific software such as browsers and games. However, when applied to general-purpose software systems like operating systems, LLM agents face three primary challenges. Firstly, the action space is vast and dynamic, posing…
Large Language Models (LLMs) have revolutionized the field of Natural Language Generation (NLG) by demonstrating an impressive ability to generate human-like text. However, their widespread us age introduces challenges that necessitate thoughtful examination, ethical scrutiny, and responsible practices. In this study, we delve into these…
Large Language Model (LLM) workloads have distinct prefill and decode phases with different compute and memory requirements which should ideally be accounted for when scheduling input queries across different LLM instances in a cluster. However existing scheduling algorithms treat LLM workloads as monolithic jobs without…
Learning to simulate the visual world from large-scale videos. In-context learning for vision data has been underexplored compared with that in natural language. Previous works studied image in-context learning, urging models to generate a single image guided by demonstrations. In this project, we propose and…
We develop accurate and efficient solutions for large-scale retrieval tasks where novel (zero-shot) items can arrive continuously at a rapid pace. Conventional Siamese-style approaches embed both queries and items through a small encoder and retrieve the items lying closest to the query. While this approach…
Speaker: Torsten HoeflerHost: Karin Strauss Machine learning presents a great opportunity for Climate simulation and research. We will discuss some ideas from the Earth Virtualization Engines summit in Berlin and several research results ranging from ensemble prediction and bias correction of simulation output, extreme compression…
Model maker and fabricator Lex Story helps bring research to life through prototyping. He discusses his take on failure; the encouragement and advice that has supported his pursuit of art and science; and the sabbatical that might inspire his next career move.
Order flow modeling stands as the most fundamental and essential financial task, as orders embody the minimal unit within a financial market. However, current approaches often result in unsatisfactory fidelity in generating order flow, and their generation lacks controllability, thereby limiting their application scenario. In…
