11 papers
MIRAGE: Multi-Perspective Creative Language Model Reasoning with Reinforcement Learning Guidance
Arash Lagzian, Srinivas Anumasa, Dianbo Liu
Recent advances in Large Language Models (LLMs) have revolutionized artificial intelligence and how human interact with AIs. Despite impressive advancements, LLMs struggle with com…
Hierarchical Data Selection via Manifold Coverage and Sparse Feature Coverage in LLM Post-training
Peng Sun, Yi Yang, Antong Zhang +7
As supervised fine-tuning data continues to scale, selecting high-value subsets from large candidate pools is crucial for reducing training cost and improving model performance. Ex…
EulerLoRA: Rank-Driven Jump Dynamics for Calibrated Parameter-Efficient Fine-Tuning
Srinivas Anumasa, Dianbo Liu
Low-Rank Adaptation (LoRA) enables parameter-efficient fine-tuning, but standard LoRA produces a single deterministic model and does not directly support predictive uncertainty est…
Breaking Diversity Collapse in Spiking Pseudo-Ensembles for Efficient OOD Detection in Remote Sensing
Srinivas Anumasa, Rushi Shah, Qiran Zou +1
Spiking Neural Networks (SNNs) are attractive for resource-constrained remote-sensing systems, but reliable out-of-distribution (OOD) detection remains challenging. Deep ensembles…
FML-bench: A Controlled Study of AI Research Agent Strategies from the Perspective of Search Dynamics
Qiran Zou, Hou Hei Lam, Wenhao Zhao +11
AI research agents accelerate ML research by automating hypothesis generation, experimentation, and empirical refinement. Existing agent strategies range from greedy hill-climbing…
Navigating heterogeneous protein landscapes through geometry-aware smoothing
Srinivas Anumasa, Barath Chandran, Tingting Chen +15
The evolutionary fitness landscape of biological molecules is extremely sparse and heterogeneous, with functional sequences forming isolated dense ``islands'' within a vast combina…
Laplacian Score Sharpening for Mitigating Hallucination in Diffusion Models
Barath Chandran. C, Srinivas Anumasa, Dianbo Liu
Diffusion models, though successful, are known to suffer from hallucinations that create incoherent or unrealistic samples. Recent works have attributed this to the phenomenon of m…
Data-Dependent Smoothing for Protein Discovery with Walk-Jump Sampling
Srinivas Anumasa, Barath Chandran. C, Tingting Chen +1
Diffusion models have emerged as a powerful class of generative models by learning to iteratively reverse the noising process. Their ability to generate high-quality samples has ex…
Human-like Content Analysis for Generative AI with Language-Grounded Sparse Encoders
Yiming Tang, Arash Lagzian, Srinivas Anumasa +9
The rapid development of generative AI has transformed content creation, communication, and human development. However, this technology raises profound concerns in high-stakes doma…
Multi-Novelty: Improve the Diversity and Novelty of Contents Generated by Large Language Models via inference-time Multi-Views Brainstorming
Arash Lagzian, Srinivas Anumasa, Dianbo Liu
Large Language Models (LLMs) demonstrate remarkable proficiency in generating accurate and fluent text. However, they often struggle with diversity and novelty, leading to repetiti…
Auto-Discovery-Bench: Diagnosing Structured State Tracking in Oracle-Guided Discovery
Tingting Chen, Beibei Lin, Srinivas Anumasa +5
Interactive discovery requires agents to maintain and update structured beliefs over many rounds of feedback. Before evaluating agents in noisy, open-ended scientific environments,…