30 citations · 91 across the 44 of their papers we have counts for
47 papers
Communication-Efficient LLM Adaptation over Decentralized GPU Meshes
Sameera Ramasinghe, Shamane Siriwardhana, Thalaiyasingam Ajanthan +8
Decentralized training enables large-model training over low-end GPUs and internet-grade connections, but communication along both data-parallel and pipeline-parallel axes becomes…
The Gate Always Closes: On Injecting Auxiliary Signals into Frozen Vision-Language Models
Moshiur Farazi, Sameera Ramasinghe, Bekir Sait Ciftler +2
Auxiliary signal pathways in VLMs are routinely fitted with learnable gates so the optimiser can decide how much of the signal to admit. We find that the optimiser almost always de…
Agora: Collective and Permissionless Internet-Scale Pretraining of Large Language Models
Gil Avraham, Violetta Shevchenko, Hadi Mohaghegh Dolatabadi +9
Training large language models at the multi-billion to trillion parameter scale is confined to datacenters, where data-parallel (DP) and model-parallel (MP) techniques presume homo…
Factored Gossip DiLoCo: Reducing Blocking Communication in DiLoCo
Chamin Hewa Koneputugodage, Thalaiyasingam Ajanthan, Sameera Ramasinghe +7
To make large-scale distributed training practical outside high-bandwidth datacenters, we must reduce blocking, high-volume synchronization. While DiLoCo communicates infrequently,…
Taming Curvature: Architecture Warm-Up for Stable Transformer Training
Sameera Ramasinghe, Ajanthan Thalaiyasingam, Hadi Mohaghegh Dolatabadi +6
Training billion-parameter Transformers is often brittle, with transient loss spikes and divergence that waste compute. Even though the recently developed Edge of Stability (EoS) t…
Mixtures of Subspaces for Bandwidth Efficient Context Parallel Training
Sameera Ramasinghe, Ajanthan Thalaiyasingam, Hadi Mohaghegh Dolatabadi +5
Pretraining language models with extended context windows enhances their ability to leverage rich information during generation. Existing methods split input sequences into chunks,…