collaborators

5 papers

cs.CL2026

Not All LLM Reasoning is Visible in the Chain-of-Thought

Vatsal Baherwani, Tom Goldstein, Ashwinee Panda

A key question for AI safety is whether a language model expresses all of its reasoning in its output tokens. We demonstrate a concrete failure mode where frontier models exhibit i…

cs.LG2026

Emergent Capabilities Arise Randomly from Learning Sparse Attention Patterns

Vatsal Baherwani, Zixi Chen, Shikai Qiu +2

Neural scaling laws for transformer language models predict smooth improvements in pretraining loss with increasing parameters, but downstream capabilities such as in-context learn…

cs.CV2025

Characterizing Motion Encoding in Video Diffusion Timesteps

Vatsal Baherwani, Yixuan Ren, Abhinav Shrivastava

Text-to-video diffusion models synthesize temporal motion and spatial appearance through iterative denoising, yet how motion is encoded across timesteps remains poorly understood.…

cs.LG2025

Dense Backpropagation Improves Training for Sparse Mixture-of-Experts

Ashwinee Panda, Vatsal Baherwani, Zain Sarwar +4

Mixture of Experts (MoE) pretraining is more scalable than dense Transformer pretraining, because MoEs learn to route inputs to a sparse set of their feedforward parameters. Howeve…

cs.LG2025

DynaGuard: A Dynamic Guardian Model With User-Defined Policies

Monte Hoover, Vatsal Baherwani, Neel Jain +7

Guardian models play a crucial role in ensuring the safety and ethical behavior of user-facing AI applications by enforcing guardrails and detecting harmful content. While standard…