51 citations · 177 across the 13 of their papers we have counts for
3 papers · 1 filter
FTB Graph: Determining and Validating First-token Broadcasters and Language-Identity Head Circuits in Multilingual Language Models
Arjun Pillai, Christian Hoang, Anjelo Laroza
Large language models operating in multilingual contexts must resolve target response languages early in generation, yet the causal circuitry governing first-token language identit…
Quantum Sensors for Chemistry and Materials Science
Piotr Put, Arjun Pillai, Xuan Hoang Le +2
The advancement of chemistry and materials science relies on transformative analytical tools which can overcome the sensitivity, spatial resolution, and throughput limitations of c…
First-Token Broadcasters: Mechanistic Origins of Language Identity and Distributed Robustness in Transformers
Arjun Pillai, Christian Hoang, Anjelo Jann Laroza
Why do multilingual language models sometimes generate in the wrong language, and why is this so hard to fix? We introduce Language Identity Head Ablation (LIHA), a causal interven…