4 / papers
Inference where the compute budget is real — on-device, at the edge, off the datacentre. 4 open-access papers by Pranay Mahendrakar, each with a permanent DOI.
In 2023 a large language model was reported to solve text-based analogy problems zero-shot at or above the level of college students. Two critiques followed. They showed that performance on letter-string analogies collapses when the…
A retrieval-augmented system that answers only when its retrieved context suffices needs a predicate that says when it does. Three incompatible predicates are in circulation under the word "sufficient", and results obtained under different…
Contradiction detection occupies an unusual position among proposals for making a language model check itself. Comparing two of a model's own outputs appears to need no external oracle, which makes it look like the one exception to the…
Spiking neural networks (SNNs) on neuromorphic hardware are routinely proposed as the path to energy-efficient on-device inference for language models. The framing usually treats this as a single technical question. It is not. There are…
Pranay Mahendrakar has published 4 open-access papers on Systems & Hardware: "Three Rounds on Emergent Analogy", "Sufficient for Whose Answer? Three Predicates Under One Word in Selective Retrieval-Augmented Generation, and Why the Only One Tied to Correctness Cannot Be Computed Where the Gate Runs", "Consistency Is Not Correctness" and "Neuromorphic Computing for On-Device LLM Inference". Each is deposited on Zenodo with a permanent DOI and is free to read.
