chandar-lab

RLHive

Recall2Imagine

Recall to Imagine, a model-based RL algorithm with superhuman memory. Oral (1.2%) @ ICLR 2024

PatchUp

IIRC

IIRC: Incremental Implicitly Refined Classification

Lifelong-Hanabi

A Continual Multi-agent RL testbed based on Hanabi

LoCA

CGOptimizer

CMOptimizer

CriticalGradientOptimization

Critical Gradient Optimization.

EpiK-Eval

Benchmark to evaluate the capability of language models to consolidate and recall information from multiple training documents.

LoCA2

COE

healthy-data-diet

Reduce gender bias in machine learning models.

RL-Tuner-CP

tgi-for-mila

A toolkit for running text-generation-inference on Mila and Compute Canada

adaptive-hanabi

crystal-design

Reinforcement Learning for Crystal Structure Design

FASP

We study the effect of attention head pruning on fairness in large language models

INF8245e-assignments-public

INF8250ae-assignments-2023

INF8250e-assignments-public

SubGoal_Distillation_LLM

Code for the paper Sub-goal Distillation: A Method to Improve Small Language Agents

INF8245e-assignments-2023

Lookbehind-SAM

Implementation of Lookbehind-SAM: k steps back, 1 step forward (ICML 2024)

r2i.github.io

RISC