3.7k citations · 11.8k across the 85 of their papers we have counts for
91 papers
SkillSmith: Learning to Compose Parametric Skills and Textual Knowledge
Lucio M. Dery, Benedict Aaron Tjandra, Siavash Samiei +4
Agentic systems driven by large language models (LLMs) regularly feature two key mechanisms to autonomously solve complex problems: synthesizing text-based knowledge and procedures…
Context Training with Active Information Seeking
Zeyu Huang, Adhiguna Kuncoro, Qixuan Feng +4
Most existing large language models (LLMs) are expensive to adapt after deployment, especially when a task requires newly produced information or niche domain knowledge. Recent wor…
Decoupled DiLoCo for Resilient Distributed Pre-training
Arthur Douillard, Keith Rush, Yani Donchev +14
Modern large-scale language model pre-training relies heavily on the single program multiple data (SPMD) paradigm, which requires tight coupling across accelerators. Due to this co…
Latent Space Communication via K-V Cache Alignment
Lucio M. Dery, Zohar Yahav, Henry Prior +3
Solving increasingly complex problems with large language models (LLMs) necessitates a move beyond individual models and towards multi-model systems that can effectively collaborat…
Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
Gheorghe Comanici, Eric Bieber, Mike Schaekermann +3431
In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our…
Communication-Efficient Language Model Training Scales Reliably and Robustly: Scaling Laws for DiLoCo
Zachary Charles, Gabriel Teston, Lucio Dery +5
As we scale to more massive machine learning models, the frequent synchronization demands inherent in data-parallel approaches create significant slowdowns, posing a critical chall…
Streaming DiLoCo with overlapping communication: Towards a Distributed Free Lunch
Arthur Douillard, Yanislav Donchev, Keith Rush +11
Training of large language models (LLMs) is typically distributed across a large number of accelerators to reduce training time. Since internal states and parameter gradients need…
DiPaCo: Distributed Path Composition
Arthur Douillard, Qixuan Feng, Andrei A. Rusu +7
Progress in machine learning (ML) has been fueled by scaling neural network models. This scaling has been enabled by ever more heroic feats of engineering, necessary for accommodat…
Asynchronous Local-SGD Training for Language Modeling
Bo Liu, Rachita Chhaparia, Arthur Douillard +5
Local stochastic gradient descent (Local-SGD), also referred to as federated averaging, is an approach to distributed optimization where each device performs more than one SGD upda…
DiLoCo: Distributed Low-Communication Training of Language Models
Arthur Douillard, Qixuan Feng, Andrei A. Rusu +6
Large language models (LLM) have become a critical component in many applications of machine learning. However, standard approaches to training LLM require a large number of tightl…
A Data Source for Reasoning Embodied Agents
Jack Lanchantin, Sainbayar Sukhbaatar, Gabriel Synnaeve +3
Recent progress in using machine learning models for reasoning tasks has been driven by novel model architectures, large-scale pre-training protocols, and dedicated reasoning datas…
Transforming Human-Centered AI Collaboration: Redefining Embodied Agents Capabilities through Interactive Grounded Language Instructions
Shrestha Mohanty, Negar Arabzadeh, Julia Kiseleva +6
Human intelligence's adaptability is remarkable, allowing us to adjust to new tasks and multi-modal environments swiftly. This skill is evident from a young age as we acquire new a…
Learning to Reason and Memorize with Self-Notes
Jack Lanchantin, Shubham Toshniwal, Jason Weston +2
Large language models have been shown to struggle with multi-step reasoning, and do not retain previous reasoning steps for future use. We propose a simple method for solving both…
Multi-Party Chat: Conversational Agents in Group Settings with Humans and Models
Jimmy Wei, Kurt Shuster, Arthur Szlam +3
Current dialogue research primarily studies pairwise (two-party) conversations, and does not address the everyday setting where more than two speakers converse together. In this wo…
Infusing Commonsense World Models with Graph Knowledge
Alexander Gurung, Mojtaba Komeili, Arthur Szlam +2
While language models have become more capable of producing compelling language, we find there are still gaps in maintaining consistency, especially when describing events in a dyn…
Collecting Interactive Multi-modal Datasets for Grounded Language Understanding
Shrestha Mohanty, Negar Arabzadeh, Milagro Teruel +9
Human intelligence can remarkably adapt quickly to new tasks and environments. Starting from a very young age, humans acquire new skills and learn how to solve new tasks either by…
CLIP-Fields: Weakly Supervised Semantic Fields for Robotic Memory
Nur Muhammad Mahi Shafiullah, Chris Paxton, Lerrel Pinto +2
We propose CLIP-Fields, an implicit scene model that can be used for a variety of tasks, such as segmentation, instance identification, semantic search over space, and view localiz…
BlenderBot 3: a deployed conversational agent that continually learns to responsibly engage
Kurt Shuster, Jing Xu, Mojtaba Komeili +15
We present BlenderBot 3, a 175B parameter dialogue model capable of open-domain conversation with access to the internet and a long-term memory, and having been trained on a large…
IGLU 2022: Interactive Grounded Language Understanding in a Collaborative Environment at NeurIPS 2022
Julia Kiseleva, Alexey Skrynnik, Artem Zholus +14
Human intelligence has the remarkable ability to adapt to new tasks and environments quickly. Starting from a very young age, humans acquire new skills and learn how to solve new t…
Interactive Grounded Language Understanding in a Collaborative Environment: IGLU 2021
Julia Kiseleva, Ziming Li, Mohammad Aliannejadi +18
Human intelligence has the remarkable ability to quickly adapt to new tasks and environments. Starting from a very young age, humans acquire new skills and learn how to solve new t…
Many Episode Learning in a Modular Embodied Agent via End-to-End Interaction
Yuxuan Sun, Ethan Carlson, Rebecca Qian +2
In this work we give a case study of an embodied machine-learning (ML) powered agent that improves itself via interactions with crowd-workers. The agent consists of a set of module…
Language Models that Seek for Knowledge: Modular Search & Generation for Dialogue and Prompt Completion
Kurt Shuster, Mojtaba Komeili, Leonard Adolphs +3
Language models (LMs) have recently been shown to generate more factual responses by employing modularity (Zhou et al., 2021) in combination with retrieval (Adolphs et al., 2021).…
Can I see an Example? Active Learning the Long Tail of Attributes and Relations
Tyler L. Hayes, Maximilian Nickel, Christopher Kanan +2
There has been significant progress in creating machine learning models that identify objects in scenes along with their associated attributes and relationships; however, there is…
Am I Me or You? State-of-the-Art Dialogue Models Cannot Maintain an Identity
Kurt Shuster, Jack Urbanek, Arthur Szlam +1
State-of-the-art dialogue models still often stumble with regards to factual accuracy and self-contradiction. Anecdotally, they have been observed to fail to maintain character ide…