Publications

MTI-LLM@NeurIPS'25

BrowserArena: Evaluating LLM Agents on Real-World Web Navigation Tasks

Sagnik Anupam, Davis Brown, Shuo Li, Eric Wong, Hamed Hassani, Osbert Bastani An arena-style benchmark for evaluating web agents on realistic web navigation tasks.

PDF Code

MOSS@ICML'25

Effective Reinforcement Learning for Reasoning in Language Models

Lianghuan Huang*, Shuo Li*, Sagnik Anupam, Insup Lee, Osbert Bastani Pre-emptive sampling and gradient filtering for improving GRPO performance.

PDF Code

DL4C@ICLR'25

LLM Program Optimization via Retrieval Augmented Search

Sagnik Anupam, Alexander Shypula, Osbert Bastani A contextual retrieval-augmented search approach for program optimization using LLMs.

PDF

MATH-AI@NeurIPS'24

MathDSL: A Domain-Specific Language for Concise Mathematical Solutions Via Program Synthesis

Sagnik Anupam, Maddy Bowers, Omar Costilla-Reyes, Armando Solar-Lezama A domain-specific language for solving mathematical equations with neurosymbolic program synthesis.

PDF Code

PoMA Vol. 45 (1)

Detection of Phonation Events for Feature-Cue-Based Analysis Using Gaussian Mixture Models

Sagnik Anupam, Jeung-Yoon Choi, Stefanie Shattuck-Hufnagel An algorithm for detecting phonation events in speech using Gaussian mixture models.

PDF

Telecom. Sys. Vol. 76 (1)

Phishing Website Detection Using Support Vector Machines and Nature-Inspired Optimization Algorithms

Sagnik Anupam, Arpan Kumar Kar Phishing website detection using SVMs optimized with nature-inspired algorithms.

PDF

MESE Vol. 6 (1)

Flood Forecasting Using a Hybrid Extreme Learning Machine-Particle Swarm Optimization Algorithm (ELM-PSO) Model

Sagnik Anupam, Padmini Pani A hybrid ELM-PSO model for flood prediction using hydrological data.

PDF

arXiv 2510.03515

RAPID: An Efficient Reinforcement Learning Algorithm for Small Language Models

Lianghuan Huang, Sagnik Anupam, Insup Lee, Shuo Li, Osbert Bastani An efficient approach for using off-policy samples in GRPO.

PDF