Sarath Chandar

Understanding Stability of Outer Nesterov Momentum on Stochastic Pseudo-Gradients

Nishant Sharma, Laura Gomezjurado Gonzalez, Ganesh Talluri +2

Large neural networks are commonly trained across many workers, and communication between them is a major cost. Workers therefore take several local optimization steps before sendi…

Global Exponential Convergence of Two-Layer Linear Network Training

Stephen Y Zhang, Gabriel Peyré

We prove global exponential (linear) convergence with an explicit rate in the rich scaling for wide two-layer linear networks trained with smooth Polyak-Lojasiewicz predictor losse…

What Does Layer-Importance Reveal About Transformers and State-Space Models?

Istabrak Abbes, Nizar Islah, Irina Rish +1

Transformers and state-space models (SSMs) are the two dominant families of sequence models, and a central open question is how far the analytical knowledge built for transformers…

Sparse Koopman Autoencoders Identify Local Dynamical Regimes in Multibasin Systems

Aidan Li, Uday Kiran Reddy Tadipatri, Mahan Fathi +2

Koopman autoencoders (KAEs) seek a higher-dimensional latent representation in which nonlinear dynamics evolve linearly. However, many interesting systems have multiple basins of a…

Dialectics of Alignment: Harnessing Unsafe Knowledge for Dynamic Safety Routing

Maryam Hashemzadeh, Jerry Huang, Minseon Kim +2

The prevailing paradigm in large language model (LLM) alignment operates via erasure, filtering unsafe data or training models to strictly refuse harmful prompts. While effective a…

CoPeP: Benchmarking Continual Pretraining for Protein Language Models

Darshan Patil, Pranshu Malviya, Mathieu Reymond +2

Protein language models (pLMs) have recently gained significant attention for their ability to uncover relationships between sequence, structure, and function from evolutionary sta…

The Expressive Limits of Diagonal SSMs for State-Tracking

Mehran Shakerinava, Behnoush Khavari, Siamak Ravanbakhsh +1

State-Space Models (SSMs) have recently been shown to achieve strong empirical performance on a variety of long-range sequence modeling tasks while remaining efficient and highly-p…

Neural Coherence : Find higher performance to out-of-distribution tasks from few samples

Simon Guiroy, Mats Richter, Sarath Chandar +1

To create state-of-the-art models for many downstream tasks, it has become common practice to fine-tune a pre-trained large vision model. However, it remains an open question of ho…

The Markovian Thinker: Architecture-Agnostic Linear Scaling of Reasoning

Milad Aghajohari, Kamran Chitsaz, Amirhossein Kazemnejad +4

Reinforcement learning (RL) has recently become a strong recipe for training reasoning LLMs that produce long chains of thought (LongCoT). Yet the standard RL "thinking environment…

GRPO- $λ$ : Credit Assignment improves LLM Reasoning

Prasanna Parthasarathi, Mathieu Reymond, Boxing Chen +2

Large language models (LLMs) are increasingly deployed for tasks requiring complex reasoning, prompting significant interest in improving their reasoning abilities through post-tra…

CrystalGym: A New Benchmark for Materials Discovery Using Reinforcement Learning

Prashant Govindarajan, Mathieu Reymond, Antoine Clavaud +3

In silico design and optimization of new materials primarily relies on high-accuracy atomic simulators that perform density functional theory (DFT) calculations. While recent works…

Parity Requires Unified Input Dependence and Negative Eigenvalues in SSMs

Behnoush Khavari, Mehran Shakerinava, Jayesh Khullar +4

Recent work has shown that LRNN models such as S4D, Mamba, and DeltaNet lack state-tracking capability due to either time-invariant transition matrices or restricted eigenvalue ran…

Revisiting Replay and Gradient Alignment for Continual Pre-Training of Large Language Models

Istabrak Abbes, Gopeshh Subbaraj, Matthew Riemer +6

Training large language models (LLMs) typically involves pre-training on massive corpora, only to restart the process entirely when new data becomes available. A more efficient and…

Optimizers Qualitatively Alter Solutions And We Should Leverage This

Razvan Pascanu, Clare Lyle, Ionut-Vlad Modoranu +6

Due to the nonlinear nature of Deep Neural Networks (DNNs), one can not guarantee convergence to a unique global minimum of the loss when using optimizers relying only on local inf…

Context-Aware Assistant Selection for Improved Inference Acceleration with Large Language Models

Jerry Huang, Prasanna Parthasarathi, Mehdi Rezagholizadeh +1

Despite their widespread adoption, large language models (LLMs) remain prohibitive to use under resource constraints, with their ever growing sizes only increasing the barrier for…

cs.LG2024★ 1 cited

BindGPT: A Scalable Framework for 3D Molecular Design via Language Modeling and Reinforcement Learning

Artem Zholus, Maksim Kuznetsov, Roman Schutski +4

Generating novel active molecules for a given protein is an extremely challenging task for generative models that requires an understanding of the complex physical interactions bet…

Sub-goal Distillation: A Method to Improve Small Language Agents

Maryam Hashemzadeh, Elias Stengel-Eskin, Sarath Chandar +1

While Large Language Models (LLMs) have demonstrated significant promise as agents in interactive tasks, their substantial computational requirements and restricted number of calls…

Intelligent Switching for Reset-Free RL

Darshan Patil, Janarthanan Rajendran, Glen Berseth +1

In the real world, the strong episode resetting mechanisms that are needed to train agents in simulation are unavailable. The \textit{resetting} assumption limits the potential of…