Neil Shah

Message

stumbling, crumbling, haphazardly falling

3mo

Neil Shah

stumbling, crumbling, haphazardly falling

Two More Methods for Consistency Training and Some New Ways to Apply It

by David Africa, Sukrati_Gautam, Neil Shah, and arav-dhoot

Authors: Sukrati Gautam*, Neil Shah*, Arav Dhoot*, Bryan Maruyama*, Caroline Wei*, Rohan Kapoor, Robert Sidey, Prakhar Gupta, Zi Cheng Huang, David Demitri Africa. This work was done for the SPAR Fellowship, and has been accepted at AI4GOOD @ ICML 2026. It was supervised by David Africa. TL;DR * We introduce...

Jun 5•16

Sealing Conditional Misalignment in Inoculation Prompting with Consistency Training

by David Africa, Sukrati_Gautam, and Neil Shah

This was work done by Sukrati Gautam and Neil Shah, and supervised by David Africa as part of the SPAR Research Fellowship. TLDR: We find a new way to use consistency training: by “sealing up” the leaky backdoor introduced by the inoculation prompt, as well as related conditional misalignment, and...

May 19•42

Gemma Gets Help: Mitigating Frustration and Self-Deletion with Consistency Training

by David Africa and Neil Shah

This was work done by Neil Shah and supervised by David Africa as part of the SPAR Research Fellowship. TLDR: Soligo et al. (2026), found that various Gemma and Gemini models became frustrated after being rejected several times on a diverse problem set. They also showed that DPO was able...

Apr 20•25