19 citations · 67 across the 21 of their papers we have counts for
23 papers
Critical-State RL: Diagnosing Trainable States for Multi-Turn Tool Use
Zixiang Chen, Wenting Zhao, Zhepeng Cen +9
Multi-turn tool-use failures can hinge on a single model call, yet reward variation alone does not reveal which call would benefit from training. When rewards depend on later inter…
Salesforce Koa: An Enterprise Language Model for Agentic Tool Use
Zixiang Chen, Sufeng Niu, Yingchi Liu +23
We present Salesforce Koa, an enterprise language model built by post-training the open-weight Nemotron-3-Super-120B foundation model with reinforcement learning using Group Relati…
Pushing Forward Pareto Frontiers of Proactive Agents with Behavioral Agentic Optimization
Yihang Yao, Zhepeng Cen, Haohong Lin +6
Proactive large language model (LLM) agents aim to actively plan, query, and interact over multiple turns, enabling efficient task completion beyond passive instruction following a…
LoCoBench-Agent: An Interactive Benchmark for LLM Agents in Long-Context Software Engineering
Jielin Qiu, Zuxin Liu, Zhiwei Liu +18
As large language models (LLMs) evolve into sophisticated autonomous agents capable of complex software development tasks, evaluating their real-world capabilities becomes critical…
Webscale-RL: Automated Data Pipeline for Scaling RL Data to Pretraining Levels
Zhepeng Cen, Haolin Chen, Shiyu Wang +8
Large Language Models (LLMs) have achieved remarkable success through imitation learning on vast text corpora, but this paradigm creates a training-generation gap and limits robust…
LoCoBench: A Benchmark for Long-Context Large Language Models in Complex Software Engineering
Jielin Qiu, Zuxin Liu, Zhiwei Liu +14
The emergence of long-context language models with context windows extending to millions of tokens has created new opportunities for sophisticated code understanding and software d…
Signal, Image, or Symbolic: Exploring the Best Input Representation for Electrocardiogram-Language Models Through a Unified Framework
William Han, Chaojing Duan, Zhepeng Cen +7
Recent advances have increasingly applied large language models (LLMs) to electrocardiogram (ECG) interpretation, giving rise to Electrocardiogram-Language Models (ELMs). Condition…
Behavior Injection: Preparing Language Models for Reinforcement Learning
Zhepeng Cen, Yihang Yao, William Han +2
Reinforcement learning (RL) has emerged as a powerful post-training technique to incentivize the reasoning ability of large language models (LLMs). However, LLMs can respond very i…
Safety is Not Only About Refusal: Reasoning-Enhanced Fine-tuning for Interpretable LLM Safety
Yuyou Zhang, Miao Li, William Han +3
Large Language Models (LLMs) are vulnerable to jailbreak attacks that exploit weaknesses in traditional safety alignment, which often relies on rigid refusal heuristics or represen…
Your Language Model May Think Too Rigidly: Achieving Reasoning Consistency with Symmetry-Enhanced Training
Yihang Yao, Zhepeng Cen, Miao Li +6
Large Language Models (LLMs) have demonstrated strong reasoning capabilities across various tasks. However, even minor variations in query phrasing, despite preserving the underlyi…
Bridging the Training-Inference Gap in LLMs by Leveraging Self-Generated Tokens
Zhepeng Cen, Yao Liu, Siliang Zeng +4
Language models are often trained to maximize the likelihood of the next token given past tokens in the training dataset. However, during inference time, they are utilized differen…
OASIS: Conditional Distribution Shaping for Offline Safe Reinforcement Learning
Yihang Yao, Zhepeng Cen, Wenhao Ding +5
Offline safe reinforcement learning (RL) aims to train a policy that satisfies constraints using a pre-collected dataset. Most current methods struggle with the mismatch between im…
Feasibility Consistent Representation Learning for Safe Reinforcement Learning
Zhepeng Cen, Yihang Yao, Zuxin Liu +1
In the field of safe reinforcement learning (RL), finding a balance between satisfying safety constraints and optimizing reward performance presents a significant challenge. A key…
Learning from Sparse Offline Datasets via Conservative Density Estimation
Zhepeng Cen, Zuxin Liu, Zitong Wang +3
Offline reinforcement learning (RL) offers a promising direction for learning policies from pre-collected datasets without requiring further interactions with the environment. Howe…
Gradient Shaping for Multi-Constraint Safe Reinforcement Learning
Yihang Yao, Zuxin Liu, Zhepeng Cen +4
Online safe reinforcement learning (RL) involves training a policy that maximizes task efficiency while satisfying constraints via interacting with the environments. In this paper,…
Constraint-Conditioned Policy Optimization for Versatile Safe Reinforcement Learning
Yihang Yao, Zuxin Liu, Zhepeng Cen +4
Safe reinforcement learning (RL) focuses on training reward-maximizing agents subject to pre-defined safety constraints. Yet, learning versatile safe policies that can adapt to var…
Datasets and Benchmarks for Offline Safe Reinforcement Learning
Zuxin Liu, Zijian Guo, Haohong Lin +8
This paper presents a comprehensive benchmarking suite tailored to offline safe reinforcement learning (RL) challenges, aiming to foster progress in the development and evaluation…
Constrained Decision Transformer for Offline Safe Reinforcement Learning
Zuxin Liu, Zijian Guo, Yihang Yao +4
Safe reinforcement learning (RL) trains a constraint satisfaction policy by interacting with the environment. We aim to tackle a more challenging problem: learning a safe policy fr…
Trustworthy Reinforcement Learning Against Intrinsic Vulnerabilities: Robustness, Safety, and Generalizability
Mengdi Xu, Zuxin Liu, Peide Huang +4
A trustworthy reinforcement learning algorithm should be competent in solving challenging real-world problems, including {robustly} handling uncertainties, satisfying {safety} cons…
On the Robustness of Safe Reinforcement Learning under Observational Perturbations
Zuxin Liu, Zijian Guo, Zhepeng Cen +4
Safe reinforcement learning (RL) trains a policy to maximize the task reward while satisfying safety constraints. While prior works focus on the performance optimality, we find tha…
Test Against High-Dimensional Uncertainties: Accelerated Evaluation of Autonomous Vehicles with Deep Importance Sampling
Mansur Arief, Zhepeng Cen, Zhenyuan Liu +4
Evaluating the performance of autonomous vehicles (AV) and their complex subsystems to high precision under naturalistic circumstances remains a challenge, especially when failure…
Constrained Variational Policy Optimization for Safe Reinforcement Learning
Zuxin Liu, Zhepeng Cen, Vladislav Isenbaev +4
Safe reinforcement learning (RL) aims to learn policies that satisfy certain constraints before deploying them to safety-critical applications. Previous primal-dual style approache…
Will This Idea Spread Beyond Academia? Understanding Knowledge Transfer of Scientific Concepts across Text Corpora
Hancheng Cao, Mengjie Cheng, Zhepeng Cen +2
What kind of basic research ideas are more likely to get applied in practice? There is a long line of research investigating patterns of knowledge transfer, but it generally focuse…