10 papers
Program-as-Weights: A Programming Paradigm for Fuzzy Functions
Wentao Zhang, Liliana Hotsko, Woojeong Kim +3
Many everyday programming tasks resist clean rule-based implementation, such as alerting on important log lines, repairing malformed JSON, or ranking search results by intent, and…
GOPO: Policy Optimization using Ranked Rewards
Kyuseong Choi, Dwaipayan Saha, Woojeong Kim +2
Standard reinforcement learning from human feedback (RLHF) trains a reward model on pairwise preference data and then uses it for policy optimization. However, while reward models…
Auto-Adaptive PINNs with Applications to Phase Transitions
Kevin Buck, Woojeong Kim
We propose an adaptive sampling method for the training of Physics Informed Neural Networks (PINNs) which allows for sampling based on an arbitrary problem-specific heuristic which…
OverFill: Two-Stage Models for Efficient Language Model Decoding
Woojeong Kim, Junxiong Wang, Jing Nathan Yan +2
Large language models (LLMs) excel across diverse tasks but face significant deployment challenges due to high inference costs. LLM inference comprises prefill (compute-bound) and…
DICE-BENCH: Evaluating the Tool-Use Capabilities of Large Language Models in Multi-Round, Multi-Party Dialogues
Kyochul Jang, Donghyeon Lee, Kyusik Kim +4
Existing function-calling benchmarks focus on single-turn interactions. However, they overlook the complexity of real-world scenarios. To quantify how existing benchmarks address p…
Approximating Language Model Training Data from Weights
John X. Morris, Junjie Oscar Yin, Woojeong Kim +2
Modern language models often have open weights but closed training data. We formalize the problem of data approximation from model weights and propose several baselines and metrics…
From Chat Logs to Collective Insights: Aggregative Question Answering
Wentao Zhang, Woojeong Kim, Yuntian Deng
Conversational agents powered by large language models (LLMs) are rapidly becoming integral to our daily interactions, generating unprecedented amounts of conversational data. Such…
A numerical scheme for a multi-scale model of thrombus in arteries
Woojeong Kim
In this article, the time-discretization of the fluid structure interaction model in the three-dimensional boundary domain is taken into account, which explains the mechanical inte…
NetFlowGen: Leveraging Generative Pre-training for Network Traffic Dynamics
Jiawei Zhou, Woojeong Kim, Zhiying Xu +2
Understanding the traffic dynamics in networks is a core capability for automated systems to monitor and analyze networking behaviors, reducing expensive human efforts and economic…
SEAL: Suite for Evaluating API-use of LLMs
Woojeong Kim, Ashish Jagmohan, Aditya Vempaty
Large language models (LLMs) have limitations in handling tasks that require real-time access to external APIs. While several benchmarks like ToolBench and APIGen have been develop…