Mayur Naik

Professor@University of Pennsylvania·美国·OpenReview

研究方向

Neuro-symbolic learning and reasoning · Machine learning for code · Differentiable programming · Compilers and programming languages for LLMs · LLM specialization

5.6

LASER: A Neuro-Symbolic Framework for Learning Spatio-Temporal Scene Graphs with Weak Supervision

ICLR 2025Poster

三作

6.0

Dolphin: A Programmable Framework for Scalable Neurosymbolic Learning

ICLR 2025Rejected

5.8

IRIS: LLM-Assisted Static Analysis for Detecting Security Vulnerabilities

ICLR 2025Poster

三作

4.3

Activations Aren't Cheap in LoRA, Weights Are

合作者 (20)

Mayur Naik

QLCoder: A Query Synthesizer For Static Analysis of Security Vulnerabilities

On Improving Neurosymbolic Learning by Exploiting the Representation Space

Delta Activations: A Representation for Finetuned Large Language Models

DOLPHIN: A Programmable Framework for Scalable Neurosymbolic Learning

Once Upon an Input: Reasoning via Per-Instance Program Synthesis

ESCA: Contextualizing Embodied Agents via Scene-Graph Generation

LASER: A Neuro-Symbolic Framework for Learning Spatio-Temporal Scene Graphs with Weak Supervision

Dolphin: A Programmable Framework for Scalable Neurosymbolic Learning

IRIS: LLM-Assisted Static Analysis for Detecting Security Vulnerabilities

Activations Aren't Cheap in LoRA, Weights Are