1 code implementation • 13 Feb 2025 • Siyan Zhao, Mingyi Hong, Yang Liu, Devamanyu Hazarika, Kaixiang Lin
We introduce PrefEval, a benchmark for evaluating LLMs' ability to infer, memorize and adhere to user preferences in a long-context conversational setting.
no code implementations • 17 Dec 2024 • Yifei Zhou, Qianlan Yang, Kaixiang Lin, Min Bai, Xiong Zhou, Yu-Xiong Wang, Sergey Levine, Erran Li
We validate PAE on challenging vision-based web navigation, using both real-world and self-hosted websites from WebVoyager and WebArena. To the best of our knowledge, this work represents the first effective learning system to apply autonomous task proposal with RL for agents that generalizes real-world human-annotated benchmarks with SOTA performances.
no code implementations • 10 Jan 2024 • Dennis Ulmer, Elman Mansimov, Kaixiang Lin, Justin Sun, Xibin Gao, Yi Zhang
This metric is used to filter the generated conversational data that is fed back in LLM for training.
1 code implementation • 21 May 2023 • Rami Aly, Xingjian Shi, Kaixiang Lin, Aston Zhang, Andrew Gordon Wilson
We observe, in the context of classification tasks, that instruction finetuned language models exhibit remarkable prompt robustness, and we subsequently propose a simple method to eliminate the need for handcrafted prompts, named AuT-Few.
no code implementations • 8 Nov 2022 • Soumajyoti Sarkar, Kaixiang Lin, Sailik Sengupta, Leonard Lausen, Sheng Zha, Saab Mansour
While prior research studies have tried to adapt these multilingual models for dialectal variants of Arabic, it still remains a challenging problem owing to the lack of sufficient monolingual dialectal data and parallel translation data of such dialectal variants.
no code implementations • 26 Aug 2022 • Vasu Sharma, Prasoon Goyal, Kaixiang Lin, Govind Thattai, Qiaozi Gao, Gaurav S. Sukhatme
We propose a multimodal (vision-and-language) benchmark for cooperative and heterogeneous multi-agent learning.
Multi-agent Reinforcement Learning
reinforcement-learning
+2
2 code implementations • 27 Feb 2022 • Xiaofeng Gao, Qiaozi Gao, Ran Gong, Kaixiang Lin, Govind Thattai, Gaurav S. Sukhatme
Language-guided Embodied AI benchmarks requiring an agent to navigate an environment and manipulate objects typically allow one-way communication: the human user gives a natural language command to the agent, and the agent can only follow the command passively.
1 code implementation • 24 Jan 2022 • Zhiwei Jia, Kaixiang Lin, Yizhou Zhao, Qiaozi Gao, Govind Thattai, Gaurav Sukhatme
With the proposed Affordance-aware Multimodal Neural SLAM (AMSLAM) approach, we obtain more than 40% improvement over prior published work on the ALFRED benchmark and set a new state-of-the-art generalization performance at a success rate of 23. 48% on the test unseen scenes.
no code implementations • 21 Jan 2022 • Tongzhou Mu, Kaixiang Lin, Feiyang Niu, Govind Thattai
We present a two-step hybrid reinforcement learning (RL) policy that is designed to generate interpretable and robust hierarchical policies on the RL problem with graph-based input.
3 code implementations • 10 Nov 2021 • Yizhou Zhao, Kaixiang Lin, Zhiwei Jia, Qiaozi Gao, Govind Thattai, Jesse Thomason, Gaurav S. Sukhatme
However, current simulators for Embodied AI (EAI) challenges only provide simulated indoor scenes with a limited number of layouts.
1 code implementation • NeurIPS 2020 • Zhuangdi Zhu, Kaixiang Lin, Bo Dai, Jiayu Zhou
To further accelerate the learning procedure, we regulate the policy update with an inverse action model, which assists distribution matching from the perspective of mode-covering.
2 code implementations • 1 Jan 2021 • Boyang Liu, Ding Wang, Kaixiang Lin, Pang-Ning Tan, Jiayu Zhou
Unsupervised anomaly detection plays a crucial role in many critical applications.
no code implementations • 24 Nov 2020 • Dong Chen, Kaian Chen. Zhaojian Li, Tianshu Chu, Rui Yao, Feng Qiu, Kaixiang Lin
Specifically, we consider the decentralized inverter-based secondary voltage control problem in distributed generators (DGs), which is first formulated as a cooperative multi-agent reinforcement learning (MARL) problem.
Deep Reinforcement Learning
Multi-agent Reinforcement Learning
+2
no code implementations • 16 Sep 2020 • Zhuangdi Zhu, Kaixiang Lin, Anil K. Jain, Jiayu Zhou
Reinforcement learning is a learning paradigm for solving sequential decision-making problems.
no code implementations • 11 Jul 2020 • Zhaonan Qu, Kaixiang Lin, Zhaojian Li, Jiayu Zhou, Zhengyuan Zhou
For strongly convex and convex problems, we also characterize the corresponding convergence rates for the Nesterov accelerated FedAvg algorithm, which are the first linear speedup guarantees for momentum variants of FedAvg in convex settings.
1 code implementation • 1 Apr 2020 • Zhuangdi Zhu, Kaixiang Lin, Bo Dai, Jiayu Zhou
SAIL bridges the advantages of IL and RL to reduce the sample complexity substantially, by effectively exploiting sup-optimal demonstrations and efficiently exploring the environment to surpass the demonstrated performance.
1 code implementation • ICLR 2020 • Kaixiang Lin, Jiayu Zhou
Sample inefficiency is a long-lasting problem in reinforcement learning (RL).
2 code implementations • 19 Feb 2018 • Liyang Xie, Kaixiang Lin, Shu Wang, Fei Wang, Jiayu Zhou
Generative Adversarial Network (GAN) and its variants have recently attracted intensive research interests due to their elegant theoretical foundation and excellent empirical performance as generative models.
1 code implementation • 18 Feb 2018 • Kaixiang Lin, Renyu Zhao, Zhe Xu, Jiayu Zhou
Large-scale online ride-sharing platforms have substantially transformed our lives by reallocating transportation resources to alleviate traffic congestion and promote transportation efficiency.
no code implementations • 18 Feb 2018 • Fengyi Tang, Kaixiang Lin, Ikechukwu Uchendu, Hiroko H. Dodge, Jiayu Zhou
Even though there is mild cognitive decline in MCI patients, they have normal overall cognition and thus is challenging to distinguish from normal aging.
1 code implementation • 19 Feb 2017 • Kaixiang Lin, Shu Wang, Jiayu Zhou
Motivated by human collaborative learning, in this paper we propose a collaborative deep reinforcement learning (CDRL) framework that performs adaptive knowledge transfer among heterogeneous learning agents.