collaborators

7 papers

cs.CR2026

The Missing Layer: Specification Infrastructure for AI Oversight

Satyam Kumar, Saurabh Jha

AI safety has a missing layer. Interpretability, formal methods, security engineering, evaluation methodology, and reinforcement-learning safety each produce substantial work, but…

cs.LG2026

CADENCE: Closing the Reasoning Gap via Coverage-Adaptive On-Policy Distillation

Satyam Kumar, Saurabh Jha

On-policy knowledge distillation transfers reasoning from large teachers to compact students, but existing approaches suffer three compounding failure modes: (i) cold-start collaps…

cs.AR2026

Forge-UGC: FX optimization and register-graph engine for universal graph compiler

Satyam Kumar, Saurabh Jha

We present Forge-UGC (FX Optimization and Register-Graph Engine for Universal Graph Compilation), a four-phase compiler for transformer deployment on heterogeneous accelerator hard…

cs.LG2026

RAMP: Reinforcement Adaptive Mixed Precision Quantization for Efficient On Device LLM Inference

Arpit Singh Gautam, Saurabh Jha

Post training quantization is essential for deploying large language models (LLMs) on resource constrained hardware, yet state of the art methods enforce uniform bit widths across…

cs.DC2026

StreamServe: Adaptive Speculative Flows for Low-Latency Disaggregated LLM Serving

Satyam Kumar, Arpit Singh Gautam, Kailash Talreja +1

Efficient LLM serving must balance throughput and latency across diverse, bursty workloads. We introduce StreamServe, a disaggregated prefill decode serving architecture that combi…

cs.CL2026

The Energy of Falsehood: Detecting Hallucinations via Diffusion Model Likelihoods

Arpit Singh Gautam, Kailash Talreja, Saurabh Jha

Large Language Models (LLMs) frequently hallucinate plausible but incorrect assertions, a vulnerability often missed by uncertainty metrics when models are confidently wrong. We pr…