4 papers
Faynt: Scaling and Optimizing Policies for Competitive Melee
Ali Janati, Nikita Kuzmin, Rohit Swamy +1
We introduce Faynt, a family of 10M- and 75M-parameter Transformer policies for Super Smash Bros. Melee, each controlling all 26 characters with a single checkpoint. After reinforc…
Router Sensitivity Under Lightweight Fine-Tuning Identifies Prunable Experts in Mixture-of-Experts Models
Ali Janati, Kaoutar El Maghraoui, Chengke Zou +4
Mixture-of-Experts (MoE) models decouple parameter count from per-token compute, but deployment still requires hosting every expert in memory. Recent theory shows that experts whos…
Post-Grokking Collapse at the Representation-Readout Interface in Muon-Trained Transformers
Ali Janati, Kaoutar El Maghraoui, Andrei Kanavalau +1
Muon-trained modular-arithmetic transformers can lose accuracy while retaining linearly decodable task information. Adjacent swaps localize five captured unnormalized failures to A…
Uncertainty-Aware Multimodal Emotion Recognition through Dirichlet Parameterization
Rémi Grzeczkowicz, Eric Soriano, Ali Janati +4
In this work, we present a lightweight and privacy-preserving Multimodal Emotion Recognition (MER) framework designed for deployment on edge devices. To demonstrate framework's ver…